The Data Quality Act
Edward G. Nilges
remove interpretations and data on public website that these interests
feel to be incorrect or unscientific.
What's interesting is that corporations will not be required to meet
the same standards for data quality in electronic media that range
from known deception in investor information to poor consumer credit
reporting and sloppy data processing that has enabled identity theft.
"Equal protection of the laws" means equal protection of the laws.
Therefore, I'd call upon concerned people to apply this act to
corporate communications.
For some time, business software has been consistently informed by a
sloppy ethic of "get it done yesterday." The resulting procedures
create messes that data processing technicians are normally expected
to work unpaid hours to fix.
What about CORPORATE data quality?
Richard Heathfield
>
> The "Data Quality Act" was recently passed under corporate pressure to
> remove interpretations and data on public website that these interests
> feel to be incorrect or unscientific.
Could you please take this to a regional political newsgroup?
comp.programming is an international technical newsgroup. Since it is
international, the passing of a particular law in a particular country
is not relevant to many of its participants. Furthermore,
comp.programming is for discussing technical programming matters, not
legal matters.
<snip>
--
Richard Heathfield : bin...@eton.powernet.co.uk
"Usenet is a strange place." - Dennis M Ritchie, 29 July 1999.
C FAQ: http://www.eskimo.com/~scs/C-faq/top.html
K&R answers, C books, etc: http://users.powernet.co.uk/eton
Tim Sinkins
Richard Heathfield wrote:
>
> "Edward G. Nilges" wrote:
> >
> > The "Data Quality Act" was recently passed under corporate pressure to
> > remove interpretations and data on public website that these interests
> > feel to be incorrect or unscientific.
>
> Could you please take this to a regional political newsgroup?
> comp.programming is an international technical newsgroup. Since it is
> international, the passing of a particular law in a particular country
> is not relevant to many of its participants. Furthermore,
> comp.programming is for discussing technical programming matters, not
> legal matters.
I wonder what he though the connection to alt.global-warming was.
Perhaps he thinks that's just a general winge about the govermnet ng.
Tim
Programmer Dude
> The "Data Quality Act" was recently passed under corporate pressure to
> remove interpretations and data on public website that these interests
> feel to be incorrect or unscientific.
Eh? I don't know about corporate pressure, but it applies to the
accuracy of *scientific* information on *government* websites.
> What's interesting is that corporations will not be required to meet
> the same standards for data quality in electronic media that range
> from known deception in investor information to poor consumer credit
> reporting and sloppy data processing that has enabled identity theft.
Cavet Emptor. So?
> "Equal protection of the laws" means equal protection of the laws.
> Therefore, I'd call upon concerned people to apply this act to
> corporate communications.
No thanks. The DQA sounds like a horrible idea, anyway. I'd be
firmly in the "anti-" group on this one.
> For some time, business software has been consistently informed by a
> sloppy ethic of "get it done yesterday." The resulting procedures
> create messes that data processing technicians are normally expected
> to work unpaid hours to fix.
The horse is long dead, Ed. Get over it.
--
|_ CJSonnack <Ch...@Sonnack.com> _____________| How's my programming? |
|_ http://www.Sonnack.com/ ___________________| Call: 1-800-DEV-NULL |
|_____________________________________________|_______________________|
Opinions expressed herein are my own and may not represent those of my employer.
Edward G. Nilges
This shows how a misunderstood "focus" creates bias. For the
arguments of those who refuse to interpret data including the recent
calving of a Rhode-Island size iceberg and 70 degrees in Chicago in
January includes the pseudo-scientific refusal to widen the "focus."
The connection to the warming ng is obvious. The Data Quality Act was
quietly rammed-through by corporations who were concerned that
government data and the associated interpretation of same were
persuading the general public that corporate policies were
contributing to global warming.
Yet these same companies do not manifest any real committment to
physical safety and data quality, for when their own engineering and
their own technical employees ask for sufficient time to incorporate
physical safety into products, and reliability into software and data,
they are in many cases prevented from doing so by "the bottom line."
For example, the Exxon Valdez incident could have been prevented by a
modification to tanker design called "double hulling", but this was
not done because it could not be shown, at least prior to the
incident, to improve Exxon's profitability.
There are many similar examples in corporate data processing, and they
are found in comp.risks. Even more so than marine engineers (who have
independent professional bodies that can address safety issues)
software developers are continually under the gun to prove that their
efforts are enhancing the bottom line, and only the bottom line. In
many companies, 'adding unneeded features' is a terminating offense.
A deliberate narrowing of "focus" masquerading as adult responsibility
is part of the problem set. It's also a convenient dodge for
dullards.
Edward G. Nilges
> "Edward G. Nilges" wrote:
> >
> > The "Data Quality Act" was recently passed under corporate pressure to
> > remove interpretations and data on public website that these interests
> > feel to be incorrect or unscientific.
>
> Could you please take this to a regional political newsgroup?
> comp.programming is an international technical newsgroup. Since it is
> international, the passing of a particular law in a particular country
> is not relevant to many of its participants. Furthermore,
> comp.programming is for discussing technical programming matters, not
> legal matters.
>
> <snip>
Hello, Richard.
I am sorry, sir, but I think your contention has no merit.
The correctness of public data, in the USA and world-wide, and the
correctness of corporate data, in the USA and world-wide is today an
artifact of the correctness of the software that produces it, and as
such international programmers need to be concerned with it.
With all due respect, the very idea that "technical programming" is
unconcerned with correctness of software or the data it produces is
part of the problem, for it generates the psychology that insists that
if software runs to completion without crashing, and the user is
"happy" in some undefined (and indeed undefinable) way, the
programmer's job is done.
This very psychology is one of the reason for the decline in prestige
of "mere" programming.
The so-called "accounting" scandals which have afflicted corporate
governance and stock valuations here in the United States and, of
course, world-wide (as a consequence of United States dominance)
appear to be in reality and in part artifacts of the mental confusion
of the responsible CPAs who have not mastered product that is produced
by apolitical and irresponsible coders.
For example, Excel makes it easy to distort trend-lines in a
completely irresponsible fashion. Countless annual reports have been
deliberately distorted by its engine, which gives no normative advice
as to the mathematics of different trend-lines.
With all due respect, I stand prepared to win Yet Another border war
with you should you choose to begin one, for the problem is precisely
narrowness of focus and indeed of spirit.
Edward G. Nilges
> "Edward G. Nilges" wrote:
>
> > The "Data Quality Act" was recently passed under corporate pressure to
> > remove interpretations and data on public website that these interests
> > feel to be incorrect or unscientific.
>
> Eh? I don't know about corporate pressure, but it applies to the
> accuracy of *scientific* information on *government* websites.
And, as such, violates the 14th Amendment to the United States
Constitution.
>
> > What's interesting is that corporations will not be required to meet
> > the same standards for data quality in electronic media that range
> > from known deception in investor information to poor consumer credit
> > reporting and sloppy data processing that has enabled identity theft.
>
> Cavet Emptor. So?
A misspelled Latin tag is remembered when the rest of culture is
destroyed, because "caveat emptor" allows dogs to get away with
deception.
"Caveat emptor" does NOT apply to investor information. The Uniform
Commercial Code does, and it requires and does not request fair
dealing.
As a result of "caveat emptor", the slime at NASDAQ can't sell stocks
and the market is in the toilet, whereas the rest of the economy,
comprised of honest working people, is recovering.
>
> > "Equal protection of the laws" means equal protection of the laws.
> > Therefore, I'd call upon concerned people to apply this act to
> > corporate communications.
>
> No thanks. The DQA sounds like a horrible idea, anyway. I'd be
> firmly in the "anti-" group on this one.
That's good, but it happens to be law. Thanks to "libertarian"
flaccidity and leave-us-aloneism, these sorts of laws are passed all
the time, because libertarian clowns prefer to live in a cloud cuckoo
land.
Eric Swanson
>
>Tim Sinkins <timothy...@alcatel-ke.de> wrote in message
news:<3CFCBFC1...@alcatel-ke.de>...
>> Richard Heathfield wrote:
>> >
>> > "Edward G. Nilges" wrote:
>> > >
>> > > The "Data Quality Act" was recently passed under corporate pressure to
>> > > remove interpretations and data on public website that these interests
>> > > feel to be incorrect or unscientific.
>> >
>> > Could you please take this to a regional political newsgroup?
>> > comp.programming is an international technical newsgroup. Since it is
>> > international, the passing of a particular law in a particular country
>> > is not relevant to many of its participants. Furthermore,
>> > comp.programming is for discussing technical programming matters, not
>> > legal matters.
>>
>> I wonder what he though the connection to alt.global-warming was.
>
>This shows how a misunderstood "focus" creates bias. For the
>arguments of those who refuse to interpret data including the recent
>calving of a Rhode-Island size iceberg and 70 degrees in Chicago in
>January includes the pseudo-scientific refusal to widen the "focus."
>
>The connection to the warming ng is obvious. The Data Quality Act was
>quietly rammed-through by corporations who were concerned that
>government data and the associated interpretation of same were
>persuading the general public that corporate policies were
>contributing to global warming.
There is already a petition to quash the U.S. National Assessment on
Climate Change, filed by the "Center for Regulatory Effectiveness":
http://www.thecre.com/quality/20020211_climate-letter.html#start
They filed long before the agencies could produce the required guidelines.
--
Eric Swanson --- E-mail address: e_sw...@skybest.com :-)
--------------------------------------------------------------
Al Dunbar
More likely it was simply misread as "alt-global-warNing" :-)
/Al
Al Dunbar
"Edward G. Nilges" <spino...@yahoo.com> wrote in message
news:f5dda427.02060...@posting.google.com...
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message
news:<3CFCAB84...@eton.powernet.co.uk>...
> > "Edward G. Nilges" wrote:
> > >
> > > The "Data Quality Act" was recently passed under corporate pressure to
> > > remove interpretations and data on public website that these interests
> > > feel to be incorrect or unscientific.
> >
> > Could you please take this to a regional political newsgroup?
> > comp.programming is an international technical newsgroup. Since it is
> > international, the passing of a particular law in a particular country
> > is not relevant to many of its participants. Furthermore,
> > comp.programming is for discussing technical programming matters, not
> > legal matters.
> >
> > <snip>
>
> Hello, Richard.
>
> I am sorry, sir, but I think your contention has no merit.
>
> The correctness of public data, in the USA and world-wide, and the
> correctness of corporate data, in the USA and world-wide is today an
> artifact of the correctness of the software that produces it, and as
> such international programmers need to be concerned with it.
Perhaps, but your post referred to a law passed in the US, which, while it
may impinge on the larger problem you note above, is only one small part.
Narrowness of focus is a two-edged sword.
> With all due respect, the very idea that "technical programming" is
> unconcerned with correctness of software or the data it produces is
> part of the problem, for it generates the psychology that insists that
> if software runs to completion without crashing, and the user is
> "happy" in some undefined (and indeed undefinable) way, the
> programmer's job is done.
>
> This very psychology is one of the reason for the decline in prestige
> of "mere" programming.
>
> The so-called "accounting" scandals which have afflicted corporate
> governance and stock valuations here in the United States and, of
> course, world-wide (as a consequence of United States dominance)
> appear to be in reality and in part artifacts of the mental confusion
> of the responsible CPAs who have not mastered product that is produced
> by apolitical and irresponsible coders.
>
> For example, Excel makes it easy to distort trend-lines in a
> completely irresponsible fashion. Countless annual reports have been
> deliberately distorted by its engine, which gives no normative advice
> as to the mathematics of different trend-lines.
>
> With all due respect, I stand prepared to win Yet Another border war
> with you should you choose to begin one, for the problem is precisely
> narrowness of focus and indeed of spirit.
No doubt, friend Edward, Richard will be glad to meet you at the border
between your two countries - as long as you get there first.
But while we are talking about the effect of one countries policies and laws
on other countries, could I put in a word against the huge tariffs placed by
the US on Canadian softwood products because of the claim that our guys are
propped up by illegal subsidies, while at the same time granting subsidies
to its own farmers? Or would that be a bit OT?
/Al
Al Dunbar
> The "Data Quality Act" was recently passed under corporate pressure to
> remove interpretations and data on public website that these interests
> feel to be incorrect or unscientific.
>
> What's interesting is that corporations will not be required to meet
> the same standards for data quality in electronic media that range
> from known deception in investor information to poor consumer credit
> reporting and sloppy data processing that has enabled identity theft.
>
> "Equal protection of the laws" means equal protection of the laws.
> Therefore, I'd call upon concerned people to apply this act to
> corporate communications.
Hey, I tried that earlier today, but was laughed out of the office for
presuning that a US law could be used that way in Canada. Actually, they
arrested me, but forgot to "Mirandize" me! I am still awaiting a call back
from that DA guy on "Law and Order" to explain to our police how they need
to observe the laws of all other countries. I wonder why he has not called?
/Al
Richard Heathfield
should be dragged through a topicality debate unless they so choose.]
"Edward G. Nilges" wrote:
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFCAB84...@eton.powernet.co.uk>...
> > "Edward G. Nilges" wrote:
> > >
> > > The "Data Quality Act" was recently passed under corporate pressure to
> > > remove interpretations and data on public website that these interests
> > > feel to be incorrect or unscientific.
> >
> > Could you please take this to a regional political newsgroup?
> > comp.programming is an international technical newsgroup. Since it is
> > international, the passing of a particular law in a particular country
> > is not relevant to many of its participants. Furthermore,
> > comp.programming is for discussing technical programming matters, not
> > legal matters.
> >
> > <snip>
>
> Hello, Richard.
>
> I am sorry, sir, but I think your contention has no merit.
I'm not surprised. As usual, you have completely failed to understand
the nature of the comp.programming newsgroup.
>
> The correctness of public data, in the USA and world-wide, and the
> correctness of corporate data, in the USA and world-wide is today an
> artifact of the correctness of the software that produces it, and as
> such international programmers need to be concerned with it.
Similarly, the amount of caffeine in coffee undoubtedly has an effect on
the quality of software written by coffee-drinking programmers. That
doesn't mean that caffeine levels are topical in comp.programming.
It is true that the correctness of data depends to some extent on the
correctness of the software that produces it. It is also true, however,
that comp.programming doesn't care whether the data is public,
corporate, personal, or completely random - what it is concerned with is
the art of programming, not the art of
programming-for-corporate-and-political-United-States-of-America.
>
> With all due respect, the very idea that "technical programming" is
> unconcerned with correctness of software or the data it produces is
> part of the problem, for it generates the psychology that insists that
> if software runs to completion without crashing, and the user is
> "happy" in some undefined (and indeed undefinable) way, the
> programmer's job is done.
You set up a strawman to knock him down. Your description of the testing
process is a travesty of the truth, and - if you /are/ a programmer
yourself - I sincerely hope you know that already.
>
> This very psychology is one of the reason for the decline in prestige
> of "mere" programming.
Maybe it is, and maybe it isn't. I don't do "mere" programming, so I
don't know. Psychology isn't really topical for comp.programming.
<off-topic stuff about the USA snipped>
> For example, Excel makes it easy to distort trend-lines in a
> completely irresponsible fashion.
So does a pencil and paper. This is not a programming issue. The issues
here are the knowledge and honesty of the person generating the
"information", and the gullibility of the person receiving the
"information". None of these issues have much, if anything, to do with
programming (or indeed global warming!).
> Countless annual reports have been
> deliberately distorted by its engine, which gives no normative advice
> as to the mathematics of different trend-lines.
This is an issue for people who prepare annual reports, not an issue for
programmers.
>
> With all due respect, I stand prepared to win Yet Another border war
> with you should you choose to begin one, for the problem is precisely
> narrowness of focus and indeed of spirit.
You might think of it as a war, and you might think you're winning. But
when you spout off-topic crap in this newsgroup, everyone loses.
Narrowness of focus is *essential* to Usenet, and narrowness of spirit
is, would you believe, off-topic here.
Gerry Quinn
>"Edward G. Nilges" wrote:
>
Look on the bright side - neither will jackasses posting to usenet.
- Gerry Quinn
verulam
> cjso...@mmm.com (Programmer Dude) wrote in message news:<3CFD1985...@mmm.com>...
> > "Edward G. Nilges" wrote:
> And, as such, violates the 14th Amendment to the United States
> Constitution.
ahah programming at last, I suggest you malloc slightly more space if
your stepping out of bounds on the 14th element in your array.
> That's good, but it happens to be law. Thanks to "libertarian"
> flaccidity and leave-us-aloneism, these sorts of laws are passed all
> the time, because libertarian clowns prefer to live in a cloud cuckoo
> land.
It's not law here so ner.
Cuckoo land eh <big grin>
Edward G. Nilges
> spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > cjso...@mmm.com (Programmer Dude) wrote in message news:<3CFD1985...@mmm.com>...
> > > "Edward G. Nilges" wrote:
> [snippety snip snip]
> > And, as such, violates the 14th Amendment to the United States
> > Constitution.
>
> ahah programming at last, I suggest you malloc slightly more space if
> your stepping out of bounds on the 14th element in your array.
The political problem of malloc is that it FAILS to discriminate and
mallocs mere bytes, which can then hold any kind of pernicious text.
I prefer to create objects for this reason.
Edward G. Nilges
Now, here I am indeed at fault. I did not mean to imply that our
Constitution should obtain in Canada. But don't you fellows have
something like the abstract principle expressed in "equal protection
of the laws." I don't remember Sergeant York of the Yukon or his
faithful dog reading the Metis their rights, but I believe English
common law applies in Canada, and it contains at least the idea of
equal protection.
We had to pass the 14th amendment to get the common law principle in
writing because our slavedriver class thought that laws only applied
to white people. However, since American slaves fled in signal
instances to Canada, I can only conclude that you fellows have a
progressive legal system, barring the occasional anti-terrorism
hysteria such as obtained under Trudeau: an anti-terrorism hysteria
which we undergo today.
>
> /Al
Edward G. Nilges
Hey, hi Al. We need to get rights in writing down here. Keeps the
slavedrivers honest.
Edward G. Nilges
> "Edward G. Nilges" <spino...@yahoo.com> wrote in message
> news:f5dda427.02060...@posting.google.com...
> > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message
> news:<3CFCAB84...@eton.powernet.co.uk>...
> > > "Edward G. Nilges" wrote:
> > > >
> > > > The "Data Quality Act" was recently passed under corporate pressure to
> > > > remove interpretations and data on public website that these interests
> > > > feel to be incorrect or unscientific.
> > >
> > > Could you please take this to a regional political newsgroup?
> > > comp.programming is an international technical newsgroup. Since it is
> > > international, the passing of a particular law in a particular country
> > > is not relevant to many of its participants. Furthermore,
> > > comp.programming is for discussing technical programming matters, not
> > > legal matters.
> > >
> > > <snip>
> >
> > Hello, Richard.
> >
> > I am sorry, sir, but I think your contention has no merit.
> >
> > The correctness of public data, in the USA and world-wide, and the
> > correctness of corporate data, in the USA and world-wide is today an
> > artifact of the correctness of the software that produces it, and as
> > such international programmers need to be concerned with it.
>
> Perhaps, but your post referred to a law passed in the US, which, while it
> may impinge on the larger problem you note above, is only one small part.
> Narrowness of focus is a two-edged sword.
Yes it is. Perhaps as the USA fades like Holland in the 18th century
I shall be less able to generalize my concerns. Nonetheless, you
write code in Canada. Do you write code that is audited for
correctness by an independent authority?
International trade is a vast and confusing subject and also off-topic
in a way that the Data Quality Act is not. Shame on you. Seriously,
I think we need truly free trade in all things. That would end this
hypocrisy in which the Bush administration calls for "free
trade"...and passes tarriffs which enable Inland Steel to reopen and
make steel the old-fashioned way, just when Wolf Lake in northern
Indiana was beginning to fully recover from the old-fashioned way.
But, I am in good faith talking about the USA's Data Quality act. Do
you seriously not want to see the discussion of Moronic Inferno issues
at all?
>
> /Al
Edward G. Nilges
Of course not. However, programming is not pure mathematics: it is,
in Knuth's words, "concrete" mathematics.
For the same reason a programmer should not be permitted to even CODE
a Or b without being able to describe, accurately, the evaluation (in
particular whether his language short-circuits) programming happens in
a scientific and industrial context.
Indeed, it is a nerd fantasy that programming exists in a vacuum.
If the exogenous issue arrives at a sufficient scale of importance,
the suffiency itself being open to your discussion (which happens to
make your rather predictable contributions valuable) then it belongs.
Suppose there was a vacuum tube Internet in 1948 wherein physicists
logged on to several groups concerned with fission and fusion bomb
technology.
Suppose that MOST of the groups were
sci.physics.blowthemalltohell.neutrinos,
sci.physics.blowthemalltohell.cladding,
sci.physics.blowthemalltohell.testingthegoddamnthing: groups
concerned, that is, with various subgoals of the Los Alamos project.
Suppose, also, that there was a father group called, of course,
sci.physics.blowthemalltohell.
Suppose J. Robert Oppenheimer logged on and posted the thread "SHOULD
we blow them all to hell?", or perhaps "SHOULD we use thermonuclear
fusion to blow them all to goddamn hell?"
In this *gedanken* experiment, I PREDICT that Oppie would have done
this and with equal certitude, I PREDICT that Ed Teller would have
taken upon Dickie Heathfield's thankless task: of enforcing
BOUNDARIES, for, don't you know, without BOUNDARIES, we'd be...free,
or something like that.
And, with all due respect, I suggest that Ed Teller would have been
disingenuous as indeed Dickie is here being disingenuous. For we know
that the real Ed Teller thought it axiomatic that we needed the
capability to blow them all to hell, using thermonuclear technology.
This *gedanken* experiment SHOWS how focus is political. Thank you
for your attention.
>
> It is true that the correctness of data depends to some extent on the
> correctness of the software that produces it. It is also true, however,
> that comp.programming doesn't care whether the data is public,
> corporate, personal, or completely random - what it is concerned with is
> the art of programming, not the art of
> programming-for-corporate-and-political-United-States-of-America.
Strange: because bizness IS so concerned. I submit that it uses this
fact to ensure that programming skill is used, in a global sense, for
bizness ends, because the business of America is bizness.
This is *en passant* an answer to Al Dunbar's concerns, of national
focus.
Ben Pfaff
> Suppose there was a vacuum tube Internet in 1948 wherein physicists
> logged on to several groups concerned with fission and fusion bomb
> technology.
>
> Suppose that MOST of the groups were
> sci.physics.blowthemalltohell.neutrinos,
> sci.physics.blowthemalltohell.cladding,
> sci.physics.blowthemalltohell.testingthegoddamnthing: groups
I didn't know that physicists had so much pent-up anger against
malls.
Edward G. Nilges
Tell that to Gerald Weinberg. It was PRECISELY his lack of focus, in
his 1972 book The Psychology of Computer Programming, that was so
valuable and useful, and it is why so much "focused" writing on UML
and use-cases is less valuable.
>
> <off-topic stuff about the USA snipped>
>
> > For example, Excel makes it easy to distort trend-lines in a
> > completely irresponsible fashion.
>
> So does a pencil and paper. This is not a programming issue. The issues
Nope. Basically, it is IMPOSSIBLE to use David Stockman's "magic
asterisk" which he used at Reagan's OMB to get Reagan's targets, using
pencil and paper.
> here are the knowledge and honesty of the person generating the
> "information", and the gullibility of the person receiving the
> "information". None of these issues have much, if anything, to do with
> programming (or indeed global warming!).
Untrue. Consider monetary rounding alone. What was an individual
fraud in the 1960s (where programmers were charged with sending
fractional pennies to their bank accounts) becomes a CORPORATE fraud
whenever a "business rule" in a telecom company dictates (as actually
happens) that calls of less than 60 seconds shall be rounded up. The
ability to do so electronically REMOVES a common practice of
long-distance systems prior to complete automation, where the operator
was empowered to forgive charges if the caller rang her and made the
case that the call was interrupted, or the person he wanted to speak
to was not home.
What you completely fail to see is an aspect of software "empowerment"
that is enhanced corporate control in disguise. This is ABOUT
programming.
>
> > Countless annual reports have been
> > deliberately distorted by its engine, which gives no normative advice
> > as to the mathematics of different trend-lines.
>
> This is an issue for people who prepare annual reports, not an issue for
> programmers.
>
This committs to me a legal error. If a programmer writes the
accounting software from which the annual report is derived, you
cannot seriously maintain that IF he's paid off by criminals to
systematically provide false reports, he's off the hook. This
exception proves the rule: we interpret the programmer in the
exceptional case to be at fault because we know that the ordinary
programmer in the normal case is a full participant in the speech-act
that an annual report comprises.
The production of a correct and honest annual report is a group effort
in which the CPAs who sign it have to depend on the honesty of the
accountants who specify the reporting system. These accountants, as
"users" have to depend on the honesty of the coders who write the
general ledger.
It is to me a fantasy, encouraged by the cruel downgrading of the
programming profession, that their is some magical break at the doors
of the data center, beyond which we can play Doom, write a line of
code, have a cup of coffee, and pretend that we are not responsible.
> >
> > With all due respect, I stand prepared to win Yet Another border war
> > with you should you choose to begin one, for the problem is precisely
> > narrowness of focus and indeed of spirit.
>
> You might think of it as a war, and you might think you're winning. But
> when you spout off-topic crap in this newsgroup, everyone loses.
> Narrowness of focus is *essential* to Usenet, and narrowness of spirit
> is, would you believe, off-topic here.
Yes, now dat we gots de Internet we must become nasty little clerks.
verulam
> verula...@hotmail.com (verulam) wrote in message news:<a78137c0.0206...@posting.google.com>...
> > spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > > cjso...@mmm.com (Programmer Dude) wrote in message news:<3CFD1985...@mmm.com>...
> > > > "Edward G. Nilges" wrote:
> [snippety snip snip]
> > > And, as such, violates the 14th Amendment to the United States
> > > Constitution.
> >
> > ahah programming at last, I suggest you malloc slightly more space if
> > your stepping out of bounds on the 14th element in your array.
>
> The political problem of malloc is that it FAILS to discriminate and
> mallocs mere bytes, which can then hold any kind of pernicious text.
> I prefer to create objects for this reason.
was I too subtle?
tell me what language do you think I'm talking about here....
Richard Heathfield
why they should be dragged through a topicality debate unless they so
simple courtesies of Usenet.]
"Edward G. Nilges" wrote:
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFDA74B...@eton.powernet.co.uk>...
> > [Followups set to comp.programming only - I see no reason why they
> > should be dragged through a topicality debate unless they so choose.]
> >
> > "Edward G. Nilges" wrote:
> > >
> > > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFCAB84...@eton.powernet.co.uk>...
> > > > "Edward G. Nilges" wrote:
> > > > >
<snip>
> > >
> > > Hello, Richard.
> > >
> > > I am sorry, sir, but I think your contention has no merit.
> >
> > I'm not surprised. As usual, you have completely failed to understand
> > the nature of the comp.programming newsgroup.
Mr Nilges, like Mr Macaulay before him, has "occasional flashes of
silence, that make his conversation perfectly delightful". By not
rebutting this point when he had the clear chance to do so, Mr Nilges
implicitly agrees with it.
> > > The correctness of public data, in the USA and world-wide, and the
> > > correctness of corporate data, in the USA and world-wide is today an
> > > artifact of the correctness of the software that produces it, and as
> > > such international programmers need to be concerned with it.
> >
> > Similarly, the amount of caffeine in coffee undoubtedly has an effect on
> > the quality of software written by coffee-drinking programmers. That
> > doesn't mean that caffeine levels are topical in comp.programming.
>
> Of course not. However, programming is not pure mathematics: it is,
> in Knuth's words, "concrete" mathematics.
Quite so. We are in full agreement here.
> For the same reason a programmer should not be permitted to even CODE
> a Or b without being able to describe, accurately, the evaluation (in
> particular whether his language short-circuits) programming happens in
> a scientific and industrial context.
Nonsense. My son is currently writing a computer game. He's 11 years
old. I am supervising the quality of his work and, indeed, doing much of
it myself, but he is certainly contributing. Am I to make him write a
sociological essay every time he learns a new programming concept, and
to forbid him to practise real world use of that concept in his game
until the essay is handed in? Stuff and nonsense!
>
> Indeed, it is a nerd fantasy that programming exists in a vacuum.
Let me get the terminology straight, here: nerds are those people who
have zits, jamjar specs, brains the size of planets and social lives the
size of walnuts, right? Well, I don't know anyone like that. Not even
one person. Perhaps you're less fortunate.
It is a peculiarly Nilgesque fantasy that all aspects of anything
remotely related to information technology are topical for
comp.programming.
>
> If the exogenous issue arrives at a sufficient scale of importance,
Please define "exogenous". It looks to me as if it should mean
"originating from outside". Is that right? I'll answer as if it means
that, and you can correct me if I'm wrong.
Now, if it does mean that, then really it doesn't matter how important
*you* think it is or how important *I* think it is - what matters is how
important it is to those people who read this group for its intended
purpose - discussions about programming.
> the suffiency itself being open to your discussion (which happens to
> make your rather predictable contributions valuable) then it belongs.
No, sir. It only belongs if it fits the charter.
>
> Suppose there was a vacuum tube Internet in 1948 wherein physicists
> logged on to several groups concerned with fission and fusion bomb
> technology.
There wasn't.
>
> Suppose that MOST of the groups were
> sci.physics.blowthemalltohell.neutrinos,
> sci.physics.blowthemalltohell.cladding,
> sci.physics.blowthemalltohell.testingthegoddamnthing: groups
> concerned, that is, with various subgoals of the Los Alamos project.
I suspect there are plenty of groups dealing with the sociology of
technology.
I spent a little time trying to find some for you, even though it's
*your* job to find a newsgroup where your stuff is topical. But guess
what? My news server carries ***so many groups*** that Lotus 1-2-3
couldn't handle them all in a convenient form. Never mind - you're
clearly interested in legal and political issues, so here are my grep
hits for legal, law, politics, political; I suggest you try some of
these groups instead of comp.programming when you want to discuss legal
and political issues:
5col.politics
ab.politics
alabama.politics
alt.arabic.politics
alt.culture.arabic.non-politics
alt.culture.turkish.politics
alt.discuss.politics
alt.fan.jenn.politics
alt.hemp.politics
alt.irc.efnet.politics
alt.irc.efnet.politics.suck
alt.irc.efnet.politics.suck.suck
alt.irc.efnet.politics.suck.suck.suck
alt.men.politics
alt.politics
alt.politics.assassination
alt.politics.black
alt.politics.black.helicopters
alt.politics.british
alt.politics.bush
alt.politics.canada
alt.politics.canada.party-of-the-unacanceller
alt.politics.carlos-may
alt.politics.clinton
alt.politics.communism
alt.politics.correct
alt.politics.corruption
alt.politics.corruption.mena
alt.politics.datahighway
alt.politics.democrats
alt.politics.democrats.clinton
alt.politics.democrats.d
alt.politics.democrats.governors
alt.politics.democrats.house
alt.politics.democrats.senate
alt.politics.drinking-age
alt.politics.ec
alt.politics.economics
alt.politics.elections
alt.politics.equality
alt.politics.europe
alt.politics.europe.misc
alt.politics.europe.party-of-the-unacanceller
alt.politics.fans
alt.politics.fans.izquierdaunida
alt.politics.gooley
alt.politics.gossip
alt.politics.greens
alt.politics.harry-browne
alt.politics.homosexual
alt.politics.homosexuality
alt.politics.homosexuality.hatemongers
alt.politics.homosexuality.hatemongers.phelps
alt.politics.homosexuality.hatemongers.rev-white
alt.politics.howard-stern
alt.politics.immigration
alt.politics.india
alt.politics.india.communist
alt.politics.india.progressive
alt.politics.international
alt.politics.italy
alt.politics.jaffo
alt.politics.jaffo.dammit
alt.politics.jason-steiner
alt.politics.kibo
alt.politics.korea
alt.politics.larouche
alt.politics.liberal
alt.politics.liberal.bleed
alt.politics.liberal.bleed.bleed
alt.politics.liberal.bleed.bleed.bleed
alt.politics.liberalism
alt.politics.libertarian
alt.politics.libertarian.creative
alt.politics.libertarian.gay
alt.politics.marrou
alt.politics.media
alt.politics.media.latimes
alt.politics.media.latimes.bias
alt.politics.meijer
alt.politics.micronations
alt.politics.middle-east
alt.politics.middle-east.party-of-the-unacanceller
alt.politics.national-socialist
alt.politics.national-socialist.config
alt.politics.national-socialist.control
alt.politics.national-socialist.texas
alt.politics.nationalism
alt.politics.nationalism.albino
alt.politics.nationalism.black
alt.politics.nationalism.texas
alt.politics.nationalism.white
alt.politics.nationalism.white.announce
alt.politics.neil-simpson
alt.politics.newt
alt.politics.newt.grinch
alt.politics.newt.grinch.grinch
alt.politics.newt.grinch.grinch.grinch
alt.politics.org
alt.politics.org.batf
alt.politics.org.btaf
alt.politics.org.ccr
alt.politics.org.cfr
alt.politics.org.cia
alt.politics.org.covert
alt.politics.org.fbi
alt.politics.org.misc
alt.politics.org.nsa
alt.politics.org.nsa.echelon
alt.politics.org.suopo
alt.politics.org.un
alt.politics.org.yps
alt.politics.org.yps.dem
alt.politics.org.yps.gop
alt.politics.pacific-rim
alt.politics.pacific-rim.party-of-the-unacanceller
alt.politics.perot
alt.politics.radical-left
alt.politics.reform
alt.politics.religion
alt.politics.republic-of
alt.politics.republic-of.texas
alt.politics.republicans
alt.politics.rightgrrl
alt.politics.satanism
alt.politics.scorched-earth
alt.politics.sex
alt.politics.shadow-government
alt.politics.shadow-government.cryptarchy
alt.politics.shadow-government.cryptarchy.rumjuggler
alt.politics.shelfbutt
alt.politics.socialism
alt.politics.socialism.democratic
alt.politics.socialism.libertarian
alt.politics.socialism.mao
alt.politics.socialism.national-socialist
alt.politics.socialism.trotsky
alt.politics.socialist
alt.politics.socialist.nazi
alt.politics.soviet
alt.politics.the-revolution
alt.politics.turn-left
alt.politics.unacanceller
alt.politics.united-we-stand
alt.politics.usa
alt.politics.usa.candidate
alt.politics.usa.candidate.president
alt.politics.usa.candidate.president.whedon
alt.politics.usa.congress
alt.politics.usa.constitution
alt.politics.usa.constitution.gun-rights
alt.politics.usa.misc
alt.politics.usa.mock
alt.politics.usa.mock.government
alt.politics.usa.newt-gingrich
alt.politics.usa.party-of-the-unacanceller
alt.politics.usa.republican
alt.politics.usa.usa-parliament
alt.politics.vietnamese
alt.politics.warpowers
alt.politics.warpowers.act
alt.politics.white-power
alt.politics.world
alt.politics.world.federalism
alt.politics.youth
alt.politics.yugoslavia
alt.sex.politics
aol.neighborhood.il.chicago.politics
asu.politics
asu.politics.speech
asu.politics.talk
at.blackbox.bbe.politics
aus.politics
aus.politics.guns
austin.politics
az.politics
ba.politics
bbs.fido38.politics
bc.politics
be.politics
bermuda.politics
bhm.politics
bit.listserv.politics
ca.politics
can.politics
chi.politics
chinese.talk.politics
clari.news.politics
clari.news.usa.gov.politics
clari.usa.gov.politics
clari.usa.politics
clari.usa.politics.personalities
clari.web.usa.gov.politics
clari.web.usa.politics
clari.web.usa.politics.personalities
clari.web.world.gov.politics
clari.web.world.gov.politics.personalities
clari.world.gov.politics
clari.world.gov.politics.personalities
cmi.politics
cna.politics
cna.politics.domestic
cna.politics.foreign
cna.politics.mainland
co.politics
co.politics.amend2
co.politics.amend2.discuss
co.politics.amend2.info
dc.politics
dfw.politics
edm.politics
england.politics
england.politics.conservative
england.politics.environment
england.politics.europe
england.politics.immigration
england.politics.misc
eug.local.connectivity.politics
eug.politics
eunet.politics
evv.politics
fido.belg.politics
fido.politics
fido7.su.mn.politics
fidonet.politics
fj.soc.politics
fl.politics
free.politics
free.politics.arena1
free.talk.persian.politics
free.uk.politics
free.uk.politics.animal-rights
free.uk.politics.libertarian
free.uk.politics.parties
free.uk.politics.parties.conservative
free.uk.politics.parties.equal-parenting
free.uk.politics.parties.green
free.uk.politics.parties.lib-dem
free.uk.politics.parties.mebyon-kernow
free.uk.politics.parties.monster-raving-loony
free.uk.politics.parties.new-labour
free.uk.politics.parties.plaid-cymru
free.uk.politics.parties.scottish-nationalist
free.uk.politics.parties.socialist-party
fsu.freenet.politics
fsu.freenet.pubforum.politics
fsu.freenet.pubforum.politics.grassroots
fsu.freenet.pubforum.politics.local
gaia.fido.gop_politics
gaia.fido.politics
git.politics
git.talk.politics
gvnc.politics
han.politics
hawaii.politics
hk.politics
houston.politics
hr.politics
hsv.politics
ia.talk.politics
ict.politics
ie.politics
info.firearms.politics
jerusalem.politics
li.politics
man.politics
mars.politics
md.politics
me.politics
mn.politics
nalive.news.politics
ne.politics
news.admin.politics
news.admin.politics.european-union
news.admin.politics.mideast
news.misc.games.chess.politics
news.misc.politics
news.misc.politics.animals
newsguy.pub.us.politics
newsguy.pub.world.politics
newsguy.world.politics
nh.politics
ni.politics
nj.politics
ny.politics
nyc.politics
nz.politics
nz.politics.announce
ont.politics
or.politics
othernet.sganet.politics
ott.politics
pa.ne.politics
pa.politics
pnet.local.politics
pnet.local.politics.talk
pnet.talk.politics
rec.games.chess.politics
relcom.politics
ri.politics
sac.politics
sanet.talk.politics
sat.politics
sbay.politics
sci.lang.politics
scn.politics
scot.politics
scruz.politics
sdnet.politics
seattle.politics
sff.discuss.politics
sff.future.politics
sk.politics
sk.politics.conservativism
sk.politics.liberal
slo.politics
soc.culture.pakistan.politics
soc.politics
soc.politics.anti-fascism
soc.politics.arms-d
soc.politics.marxism
tacoma.politics
talk.politics
talk.politics.animals
talk.politics.assassination
talk.politics.china
talk.politics.circumcision
talk.politics.cis
talk.politics.clinton
talk.politics.crypto
talk.politics.drugs
talk.politics.european-union
talk.politics.extremism
talk.politics.guns
talk.politics.immigration
talk.politics.internet
talk.politics.libertarian
talk.politics.local
talk.politics.medicine
talk.politics.mics
talk.politics.mideast
talk.politics.misc
talk.politics.natl-socialism
talk.politics.peace
talk.politics.rent-control
talk.politics.soviet
talk.politics.space
talk.politics.taxation
talk.politics.theory
talk.politics.tibet
talk.politics.usa
talk.politics.usa.constitution
triad-online.politics-and-government
trial.talk.politics
trial.talk.politics.peace
triangle.politics
tw.bbs.soc.politics
tw.bbs.soc.politics.dpp
tw.bbs.soc.politics.kmt
tw.bbs.soc.politics.np
tw.infotimes.politics
tw.infotimes.politics.world
tx.politics
tx.politics.republic
ucb.politics
ucb.politics.progressive
uiuc.misc.politics
uiuc.politics
uk.politics
uk.politics.animals
uk.politics.announce
uk.politics.censorship
uk.politics.constitution
uk.politics.crime
uk.politics.drugs
uk.politics.economics
uk.politics.electoral
uk.politics.environment
uk.politics.guns
uk.politics.misc
uk.politics.parliament
uk.politics.philosophy
ukr.politics
us.politics
us.politics.abortion
us.politics.bob-dole
us.politics.elections
us.politics.phil-gramm
us.sc.charleston.politics
us.sc.chester.politics
us.sc.columbia.politics
us.sc.florence.politics
us.sc.gsp.politics
us.sc.lancaster.politics
us.sc.rockhill.politics
va.politics
viwa.politics
wa.politics
wales.politics
wales.politics.assembly
wales.politics.general
wash.politics
worldlynx.politics
wpg.politics
za.politics
zipnews.gov.us.national.politics
zipnews.gov.world.politics
clari.editorial.commentary.political
clari.news.crime.murders.political
clari.web.editorial.commentary.political
clari.web.news.crime.murders.political
gaia.fido.z1_political
acc.sbell.usa-today.law
afp.law
alt.airline.class.action.lawsuit
alt.antiques.delaware
alt.antiques.delaware.joe
alt.banjo.clawhammer
alt.bonehead.phil-lawlor
alt.building.law
alt.cats.declawing-debate
alt.crimehip.lit.authors.czeslaw-milosz
alt.culture.malawi
alt.flame.tim.law
alt.flame.tim-law
alt.games.lucas-arts.outlaws
alt.games.outlaws
alt.greek-lawyers
alt.law
alt.law.war-crimes
alt.law.war-crimes.tribunals
alt.law-enforcement
alt.law-enforcement.O_P_P
alt.law-enforcement.R_C_M_P
alt.law-enforcement.corruption
alt.law-enforcement.coruption
alt.law-enforcement.interpol
alt.law-enforcement.london-ontario
alt.law-enforcement.london_ontario
alt.law-enforcement.m
alt.law-enforcement.sarnia-ontario
alt.law-enforcement.sarnia_ontario
alt.law-enforcement.toronto_ontario
alt.law-enforcement.traffic
alt.law-enforcement.windsor-ontario
alt.law-enforcement.windsor_ontario
alt.lawyers
alt.lawyers.sue
alt.lawyers.sue.sue
alt.lawyers.sue.sue.sue
alt.philosophy.law
alt.self-help-law
alt.tv.la-law
alt.tv.law-and-order
alt.tv.martial-law
alt.uk.law
alt.uk.law.and
alt.uk.law.and.order
alt.uk.law.intellectual-property
apc.elaw
apc.elaw.public
apc.elaw.public.int
area.list.law-and-order
bit.listserv.ada-law
bit.listserv.lawsch
bit.listserv.lawsch.internships
bit.listserv.lawsch-l
clari.local.california.sfbay.law
clari.local.delaware
clari.nb.law
clari.news.law
clari.news.law.crime
clari.news.law.crime.sex
clari.news.law.prison
clari.news.law.supreme
clari.news.law_enforce
clari.news.usa.law
clari.news.usa.law.supreme
clari.usa.law
clari.usa.law.misc
clari.usa.law.supreme
clari.web.local.california.sfbay.law
clari.web.news.law_enforce
clari.web.usa.law
clari.web.usa.law.misc
clari.web.usa.law.supreme
clari.world.law
cn.bbs.sci.law
cs-monolit.press.business.law
cu.courses.laws5223
cu.courses.laws8428
delaware
delaware.bulletin
delaware.forsale
delaware.general
delaware.nerds
delaware.test
delaware.weather
dux.pressa.law
fido.su.civil_law
fido7.aids.law
fido7.civil-law
fido7.law
fido7.pvt.law
fido7.su.civil-law
fido7.su.civil_law
fj.sci.law
fj.soc.law
free.uk.profession.law
gaia.fido.bbslaw
gaia.fido.law
gov.us.topic.gov-jobs.offered.law-enforce
gov.us.topic.law
gov.us.topic.law.pub-contract
hrnet.intllaw
law
law.class
law.class.conlaw2-goldstei
law.class.evidence-wiley
law.class.secreg-bainbridge
law.court
law.court.federal
law.court.federal.archive
law.court.federal.chunk-tag
law.court.federal.database
law.court.federal.home-page
law.library
law.library.future
law.library.future.ctl
law.listserv
law.listserv.buddies-list
law.listserv.compcons
law.listserv.edlaw
law.listserv.election-law
law.listserv.lnet-tax
law.listserv.net-lawyers
law.school
law.school.admin
law.school.anti-trust
law.school.antitrust
law.school.clinic
law.school.clinic.info-law
law.school.computer
law.school.computer.apps
law.school.conlaw
law.school.copyright
law.school.corps
law.school.crim
law.school.evidence
law.school.gradtax
law.school.gradtax.partner
law.school.gradtax.s-corp
law.school.legal-prof
law.school.sources
law.school.sources.hyson
law.school.sources.perritt
law.school.tax
law.school.tax.basic
law.school.tax.business
law.school.tax.busiplan
newsguy.us.state.delaware
ntu.eve.law
ntu.law
ntu.law.economics
ntu.law.law
ntu.law.politic
ntu.law.sociology
ntu.law.st
relcom.comp.law
su.class.law314
su.class.law623
tnn.law
tnn.law.copyright
tnn.law.patent
tw.bbs.sci.law
tx3.enumclaw
ucb.boalt.high-tech-law-j
ukr.law
umn.law
umn.law.class
umn.law.class.5109
umn.law.computers
usa-today.law
utexas.class.cs105.c++-dianelaw
utexas.law
yale.forum.law-school
alt.california.illegals
alt.fan.dean-stark.advice.legal
aol.commerce.legal-services
aol.commerce.legal-services.wanted
aus.legal
aus.legal.moderated
can.legal
corel.support.wps.legal
corelsupport.wordperfectsuite-legal
england.legal
fedreg.legal
fido7.mo.legal
fido7.mo.legal.soft
fido7.ru.internet.nelegal
law.school.legal-prof
misc.immigration.illegal-aliens
misc.legal
misc.legal.beagles
misc.legal.computing
misc.legal.moderated
misc.legal.self-represent
mygale.mylegal
news.groups.legal
news.groups.legal.moderated
pdaxs.services.legal
pnet.talk.legal
rec.aviation.legal
scot.legal
tnn.interv.ngolegal
trial.misc.legal
trial.misc.legal.software
uk.legal
us.legal
us.legal.self-represent
alt.sci.sociology
ntu.law.sociology
sci.sociology
>
> Suppose, also, that there was a father group called, of course,
> sci.physics.blowthemalltohell.
>
> Suppose J. Robert Oppenheimer logged on and posted the thread "SHOULD
> we blow them all to hell?", or perhaps "SHOULD we use thermonuclear
> fusion to blow them all to goddamn hell?"
>
> In this *gedanken* experiment, I PREDICT that Oppie would have done
> this and with equal certitude, I PREDICT that Ed Teller would have
> taken upon
Your predictions are unfalsifiable and thus of no value in scientific
discourse.
> Dickie Heathfield's
Childish ad hominem attacks again? Really, I thought you might have
learned your lesson from last time. If you want to be taken seriously
(and you clearly do), then don't throw your credibility (such as it is)
into the fire by trying to mock people's names. It's about as logical as
mocking their shoe size.
> thankless task: of enforcing
> BOUNDARIES, for, don't you know, without BOUNDARIES, we'd be...free,
> or something like that.
Oh, don't misunderstand me. You're perfectly free to post your off-topic
stuff here. Please don't forget, however, that I am equally free to post
my opinion about how it would be better for all concerned, including
you, if you would find a newsgroup where your stuff is topical.
>
> And, with all due respect, I suggest that Ed Teller would have been
> disingenuous as indeed Dickie is here being disingenuous.
Another ad hominem attack? If you had a serious point, you'd make it,
instead of fooling around.
<OT stuff snipped>
>
> This *gedanken* experiment SHOWS how focus is political.
It shows *nothing* except your unfalsifiable and thus pointless
interpretation of it.
> Thank you
> for your attention.
Does that mean you're done now?
> > It is true that the correctness of data depends to some extent on the
> > correctness of the software that produces it. It is also true, however,
> > that comp.programming doesn't care whether the data is public,
> > corporate, personal, or completely random - what it is concerned with is
> > the art of programming, not the art of
> > programming-for-corporate-and-political-United-States-of-America.
>
> Strange: because bizness IS so concerned.
But this newsgroup isn't "bizness" - it's comp.programming.
> I submit that it uses this
> fact to ensure that programming skill is used, in a global sense, for
> bizness ends, because the business of America is bizness.
That is not the business of this newsgroup, however. The business of
this newsgroup is to discuss a technical subject, that of programming.
>
> This is *en passant* an answer to Al Dunbar's concerns, of national
> focus.
Not really. Well, not a correct answer, anyway. You see, America is not
topical in comp.programming.
<snip>
> > > For example, Excel makes it easy to distort trend-lines in a
> > > completely irresponsible fashion.
> >
> > So does a pencil and paper. This is not a programming issue. The issues
> > here are the knowledge and honesty of the person generating the
> > "information", and the gullibility of the person receiving the
> > "information". None of these issues have much, if anything, to do with
> > programming (or indeed global warming!).
Ah, silence from Mr Nilges on this point, I see. Thus, he implicitly
agrees with the point that this paragraph makes - i.e. that his article
is off-topic in this newsgroup. If he knows it's off-topic, why does he
post it here?
> >
> > > Countless annual reports have been
> > > deliberately distorted by its engine, which gives no normative advice
> > > as to the mathematics of different trend-lines.
> >
> > This is an issue for people who prepare annual reports, not an issue for
> > programmers.
Again, silent assent from Mr Nilges.
> >
> > >
> > > With all due respect, I stand prepared to win Yet Another border war
> > > with you should you choose to begin one, for the problem is precisely
> > > narrowness of focus and indeed of spirit.
> >
> > You might think of it as a war, and you might think you're winning. But
> > when you spout off-topic crap in this newsgroup, everyone loses.
> > Narrowness of focus is *essential* to Usenet, and narrowness of spirit
> > is, would you believe, off-topic here.
Again, silent assent from Mr Nilges.
Ian Woods
> verula...@hotmail.com (verulam) wrote in message
> news:<a78137c0.0206...@posting.google.com>...
>> spino...@yahoo.com (Edward G. Nilges) wrote in message
>> news:<f5dda427.02060...@posting.google.com>...
>> > cjso...@mmm.com (Programmer Dude) wrote in message
>> > news:<3CFD1985...@mmm.com>...
>> > > "Edward G. Nilges" wrote: [snippety snip snip]
>> > And, as such, violates the 14th Amendment to the United States
>> > Constitution.
>>
>> ahah programming at last, I suggest you malloc slightly more space if
>> your stepping out of bounds on the 14th element in your array.
>
> The political problem of malloc is that it FAILS to discriminate and
> mallocs mere bytes, which can then hold any kind of pernicious text.
> I prefer to create objects for this reason.
Not so. It allocates storage for an object, or informs you that it cannot
allocate storage for an object the size you requested. Importantly, it
doesn't specify that the allocated space for your object "holds any kind of
pernicious text" - merely that it's value is indeterminate. Here is the
relavent section about mallocs behaviour from ISO/IEC 9899:1999...
7.20.3.3 The malloc function
Synopsis
1 #include <stdlib.h>
void *malloc(size_t size);
Description
2 The malloc function allocates space for an object whose size is specified
by size and whose value is indeterminate.
Returns
3 The malloc function returns either a null pointer or a pointer to the
allocated space.
Ian Woods
Programmer Dude
> The political problem of malloc...
The "political problem of malloc"?!?!
Bwa-ha-ha-ha-ha....
Kook!
Richard Heathfield
>
<snip>
>
How true.
> I did not mean to imply that our
> Constitution should obtain in Canada.
In alt.global-warming, I presume, "our" means "referring to the members
of alt.global-warming" (unless there is a specific indication to the
contrary); in comp.programming, presumably, it means "referring to the
members of comp.programming", again unless there is a specific
indication to the contrary. Where an article is cross-posted between the
two groups, then, "our" surely means "referring to people who are
members of *both* groups, alt.global-warming *and* comp.programming",
unless there is a specific indication to the contrary. What, then, is
this Constitution of which you speak, which applies to all those who
subscribe to both alt.global-warming and comp.programming, except
(apparently) those who live in Canada? Even without the qualification,
the chances are high that you already have an almost empty set.
> But don't you fellows have
Who do you mean by "you fellows"? People who do not read either
newsgroup? Clearly not. People who read one newsgroup but not the other?
Maybe. People who subscribe to both groups but who also live in Canada?
Ah, this seems most likely, but again you're probably looking at an
empty set.
<snip OT stuff>
> We had to pass the 14th amendment
(a) The 14th amendment to what?
(b) If and only if it is relevant to programming, please explain who you
mean by "we", precisely. Were you, personally, involved?
<political stuff snipped>
Richard Heathfield
should be dragged through a topicality debate unless they so choose.
settings) suggest that he disagrees. I don't know why.]
"Edward G. Nilges" wrote:
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFDA74B...@eton.powernet.co.uk>...
> >
> > "Edward G. Nilges" wrote:
<snip>
> > > This very psychology is one of the reason for the decline in prestige
> > > of "mere" programming.
> >
> > Maybe it is, and maybe it isn't. I don't do "mere" programming, so I
> > don't know. Psychology isn't really topical for comp.programming.
>
> Tell that to Gerald Weinberg.
You tell him. I'm busy.
> It was PRECISELY his lack of focus, in
> his 1972 book The Psychology of Computer Programming, that was so
> valuable and useful, and it is why so much "focused" writing on UML
> and use-cases is less valuable.
It doesn't take a 30-year-old book to tell me UML is less valuable. But
at least the discussion of the utility of UML for program design is
topical here. Do you want to talk about the utility of UML? If so, I'm
all ears.
> >
> > <off-topic stuff about the USA snipped>
> >
> > > For example, Excel makes it easy to distort trend-lines in a
> > > completely irresponsible fashion.
> >
> > So does a pencil and paper. This is not a programming issue. The issues
>
> Nope. Basically, it is IMPOSSIBLE to use David Stockman's "magic
> asterisk" which he used at Reagan's OMB to get Reagan's targets, using
> pencil and paper.
If David Stockman's magic asterisk is a programming technique,
discussion of its intent, implementation, and usage are topical here. Is
/that/ what you want to talk about?
>
> > here are the knowledge and honesty of the person generating the
> > "information", and the gullibility of the person receiving the
> > "information". None of these issues have much, if anything, to do with
> > programming (or indeed global warming!).
>
> Untrue.
I strongly disagree, but let's see what you have to say...
> Consider monetary rounding alone. What was an individual
> fraud in the 1960s (where programmers were charged with sending
> fractional pennies to their bank accounts) becomes a CORPORATE fraud
> whenever a "business rule" in a telecom company dictates (as actually
> happens) that calls of less than 60 seconds shall be rounded up.
This still looks like a matter of honesty (or at least its corporate
shadow, "compliance"), not a matter of programming. The programming here
is not difficult to do either way. The difficulty, if there is any, is
in the mind of the person writing the specification. Is he honest, or
not? This is not a technical issue, but a moral one.
> The
> ability to do so electronically REMOVES a common practice of
> long-distance systems prior to complete automation, where the operator
> was empowered to forgive charges if the caller rang her and made the
> case that the call was interrupted, or the person he wanted to speak
> to was not home.
Same applies. This isn't a programming issue but a corporate customer
service issue. If the phone company wants to make this feature available
to its customers, then it is free so to do. That decision, however, is
not a matter of programming. It's up to the phone company.
>
> What you completely fail to see is an aspect of software "empowerment"
> that is enhanced corporate control in disguise. This is ABOUT
> programming.
Show us. Write a computer program that disguises enhanced corporate
control as an aspect of software "empowerment", and post the source here
in a relatively well-known programming language. Many of us will then
cheerfully comment upon it for you.
>
> >
> > > Countless annual reports have been
> > > deliberately distorted by its engine, which gives no normative advice
> > > as to the mathematics of different trend-lines.
> >
> > This is an issue for people who prepare annual reports, not an issue for
> > programmers.
> >
> This committs to me a legal error. If a programmer writes the
> accounting software from which the annual report is derived, you
> cannot seriously maintain that IF he's paid off by criminals to
> systematically provide false reports, he's off the hook.
Of course I don't maintain that. But this is still an honesty issue, not
a programming issue.
> This
> exception proves the rule:
No, it doesn't prove anything.
> we interpret the programmer in the
> exceptional case to be at fault because we know that the ordinary
> programmer in the normal case is a full participant in the speech-act
> that an annual report comprises.
This is beginning to look like nonsense again.
> The production of a correct and honest annual report is a group effort
> in which the CPAs who sign it have to depend on the honesty of the
> accountants who specify the reporting system. These accountants, as
> "users" have to depend on the honesty of the coders who write the
> general ledger.
Right. It's an honesty issue.
> It is to me a fantasy, encouraged by the cruel downgrading of the
> programming profession, that their is some magical break at the doors
> of the data center, beyond which we can play Doom, write a line of
> code, have a cup of coffee, and pretend that we are not responsible.
Nobody is pretending, as far as I can see, that programmers are not
responsible for the code they write. That's one reason why it's good to
have a newsgroup where programmers can discuss technical issues.
> > > With all due respect, I stand prepared to win Yet Another border war
> > > with you should you choose to begin one, for the problem is precisely
> > > narrowness of focus and indeed of spirit.
> >
> > You might think of it as a war, and you might think you're winning. But
> > when you spout off-topic crap in this newsgroup, everyone loses.
> > Narrowness of focus is *essential* to Usenet, and narrowness of spirit
> > is, would you believe, off-topic here.
>
> Yes, now dat we gots de Internet we must become nasty little clerks.
Not at all. Let us, instead, refrain from childish ad hominem attacks
and become considerate Netizens who find appropriate newsgroups in which
to post.
Programmer Dude
> The correctness of public data, in the USA and world-wide, and the
> correctness of corporate data, in the USA and world-wide is today an
> artifact of the correctness of the software that produces it,..
The correctness of the software is one small piece of the overall
puzzle. Once again, you're demonstrating your fundimental confusion
between business issues that have nothing to do with programming per
se and Actual Programming Issues.
> With all due respect, the very idea that "technical programming" is
> unconcerned with correctness of software or the data it produces is
> part of the problem,
The idea that programming is "unconcerned with correctness" is, in my
experience, a deviant one which appears driven by your hurt feelings
over ancient perceived ills. Indeed, one of the Holy Grails of the art
is the ability to mathematically prove program correctness.
(In my opinion, this is unattainable for most real world programs,
but that's another discussion. But in brief, it has mainly to do
with my perception that programming is--at least currently--as much
an art form as a science.)
> ...for it generates the psychology that insists that if software runs
> to completion without crashing, and the user is "happy" in some
> undefined (and indeed undefinable) way, the programmer's job is done.
Complete, utter nonsense.
Maybe YOU work that way; no one I know does.
> This very psychology is one of the reason for the decline in prestige
> of "mere" programming.
More nonsense. I suspect it has a great deal to do with the perception
that 'any nine-year-old' can program... "After, all, my sister's kid is
programming all the time...it must be pretty easy."
I think it also has a great deal to do with that our industry is
unregulated. ANYONE can claim to be a programmer, and many lack
the most basic skills.
It's also still an arcane profession not always understood by those
who manage, and TWM frequently lack the ability to differentiate
wheat and chaff.
> The so-called "accounting" scandals which have afflicted corporate
> governance and stock valuations here in the United States and, of
> course, world-wide (as a consequence of United States dominance)
> appear to be in reality and in part artifacts of the mental confusion
> of the responsible CPAs who have not mastered product that is produced
> by apolitical and irresponsible coders.
>
> For example, Excel makes it easy to distort trend-lines in a
> completely irresponsible fashion. Countless annual reports have been
> deliberately distorted by its engine, which gives no normative advice
> as to the mathematics of different trend-lines.
In other words, ignorant or devious people can corrupt or manipulate
information. Wow, that IS news! And, oh, so relevant to programming.
Rule #84: Know your tools.
When powerful tools are made easy to use by the masses is it any
wonder accidents occur?
Programmer Dude
>>> This very psychology is one of the reason for the decline in prestige
>>> of "mere" programming.
>>
>> Maybe it is, and maybe it isn't. I don't do "mere" programming, so I
>> don't know. Psychology isn't really topical for comp.programming.
>
> Tell that to Gerald Weinberg.
If he posts off-topic stuff here, we will. Believe me, we will!
> Untrue. Consider monetary rounding alone. What was an individual
> fraud in the 1960s (where programmers were charged with sending
> fractional pennies to their bank accounts) becomes a CORPORATE fraud
> whenever a "business rule" in a telecom company dictates (as actually
> happens) that calls of less than 60 seconds shall be rounded up.
You continue to confuse business practice with programming. The telco
could perform the same beneficial adjustment if there were human beings
watching clocks and writing down numbers.
The *decision* to manage a business in a certain way is a choice made
by human beings. It has nothing to do with programming.
> The ability to do so electronically REMOVES a common practice of
> long-distance systems prior to complete automation, where the operator
> was empowered to forgive charges if the caller rang her and made the
> case that the call was interrupted, or the person he wanted to speak
> to was not home.
Nothing has changed. If no one is home and the phone is not answered,
you do not pay for the call. If you care to call your telco's service
department and complain about a disconnect, they may well forgive the
call. Of course, I've never HAD a disconnect; things work a bit better
today than yesteryear.
> What you completely fail to see is an aspect of software "empowerment"
> that is enhanced corporate control in disguise. This is ABOUT programming.
Wrong, Ed. Completely, utterly, totally wrong.
They Who Manage use whatever tools are available. They've done so since
the dawn of Corporate Management. This is NOT about programming.
>>> Countless annual reports have been deliberately distorted by its
>>> engine, which gives no normative advice as to the mathematics of
>>> different trend-lines.
>>
>> This is an issue for people who prepare annual reports, not an issue
>> for programmers.
>
> This committs to me a legal error. If a programmer writes the
> accounting software from which the annual report is derived, you
> cannot seriously maintain that IF he's paid off by criminals to
> systematically provide false reports, he's off the hook.
Trancendental nonsense. "Paid of by criminals"?!?! What the hell
color is the sky in your world? I imagine it's lurid, indeed.
Again (and again and again) you confuse basic human behaviors with a
craft that may--or may not--contain those behaviors. Said behaviors
are *completely* orthagonal to the craft. Said behaviors therefore
*clearly* are not related to said craft.
(As I write this, I realize you'll never absorb this, but one hopes
the onreaders will.)
> This exception proves the rule: we interpret the programmer in the
> exceptional case to be at fault because we know that the ordinary
> programmer in the normal case is a full participant in the speech-act
> that an annual report comprises.
If you mean simply that a craftsperson is responsible for their work,
this is (1) self-evident and (2) not related to programming in any
specific way whatsoever.
> The production of a correct and honest annual report is a group effort
> in which the CPAs who sign it have to depend on the honesty of the
> accountants who specify the reporting system. These accountants, as
> "users" have to depend on the honesty of the coders who write the
> general ledger.
So,.... you meet many of these 'dishonest' programmers being 'paid off
by criminals' do you? Maybe you should try one of those tin-foil hats
people of your ilk seem to find so beneficial. I've heard the CIA
uses microwaves to control thoughts....you gotta watch out for that!
> It is to me a fantasy, encouraged by the cruel downgrading of the
> programming profession, that their is some magical break at the doors
> of the data center, beyond which we can play Doom, write a line of
> code, have a cup of coffee, and pretend that we are not responsible.
I'm glad you have some sense you live in a fantasy world.
The next step is to Seek Help. I understand they have excellent
medications available these days. And almost no side effects!
That's got to be better than worrying about these criminal programmer
gangs who are out to get you.
Edward G. Nilges
I'm well aware of what malloc does. Actual experience shows that
malloc and free cause memory leaks, and that objects, which don't
allow you to malloc, don't cause as many problems.
malloc was an early 1990s solution to problems whose complexity
outruns its capabilities.
It shouldn't be used in an applications program. It should only be
used in an OO runtime to support higher-level capabilities.
Edward G. Nilges
> spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.0206...@posting.google.com>...
> > > spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > > > cjso...@mmm.com (Programmer Dude) wrote in message news:<3CFD1985...@mmm.com>...
> > > > > "Edward G. Nilges" wrote:
> [snippety snip snip]
> > > > And, as such, violates the 14th Amendment to the United States
> > > > Constitution.
> > >
> > > ahah programming at last, I suggest you malloc slightly more space if
> > > your stepping out of bounds on the 14th element in your array.
> >
> > The political problem of malloc is that it FAILS to discriminate and
> > mallocs mere bytes, which can then hold any kind of pernicious text.
> > I prefer to create objects for this reason.
>
> was I too subtle?
>
> tell me what language do you think I'm talking about here....
Probably, C or a similar out-of-date language.
Edward G. Nilges
Very funny, thanks for pointing out the ambiguity. Blow them all to
hell at the mall? We're going to get flagged by John Ashcroft's
Internet sniffer.
Edward G. Nilges
training time to mention linkages with topics outside the workplace
per se. I suggest that the picture of the father and son working
together SILENTLY is an American cultural artifact, for my own German
grandfather worked with his sons not in silence but instead yelling at
them about the right way, the wrong way, his way, the highway, the
virtues of Ohio Republicanism and the perfidy of the Copperheads.
Only in America are sons supposed to reinvent the wheel.
> >
> > Indeed, it is a nerd fantasy that programming exists in a vacuum.
>
> Let me get the terminology straight, here: nerds are those people who
> have zits, jamjar specs, brains the size of planets and social lives the
> size of walnuts, right? Well, I don't know anyone like that. Not even
> one person. Perhaps you're less fortunate.
No: for under the Popcult lies the reality which Cold War America
happened to share with the Soviet Union.
This is that in the Soviet Union, many of the most intelligent youths
of the imediate postwar generation took one look at the treatment of
the fathers when their fathers entered politics and law and elected
instead to become mathematicians and scientists.
I think this happened here, too. I entered programming in 1973
because Watergate made it seemed that BOTH sides (Nixon and his
opponents) were completely corrupt. I knew that Nixon stopped at
nothing to stay in power, but I also could see that the Beltway elite
hated him precisely because of his residual humanity and residual
ethics, were expressed in his conflicted, tortured "body language." I
therefore did not want to go to law school and associate with either
side, and at the time software development was so unknown as to be
rife with interesting characters.
>
> It is a peculiarly Nilgesque fantasy that all aspects of anything
> remotely related to information technology are topical for
> comp.programming.
>
It's not a fantasy, Dickie. Programming is a phenomenology for it is
ABOUT something. It is the writing of texts, which are not
things-in-the-world but *qua* texts refer to something outside
themselves, with the exception, of course, of null For loops.
> >
> > If the exogenous issue arrives at a sufficient scale of importance,
>
> Please define "exogenous". It looks to me as if it should mean
> "originating from outside". Is that right? I'll answer as if it means
> that, and you can correct me if I'm wrong.
Correct.
>
> Now, if it does mean that, then really it doesn't matter how important
> *you* think it is or how important *I* think it is - what matters is how
> important it is to those people who read this group for its intended
> purpose - discussions about programming.
These "real people" treat this ng as the place for random idiotic
questions which have, as a matter of fact, specialized ngs which the
posters are too goddamn lazy to find.
You appoint yourself only the defender of narrowness: yet I do not see
where you critique the converse error, of posting a question for which
a MORE SPECIALISED ng exists! I can only conclude that your brief is
for narrowness, not the purity of this newsgroup.
>
> > the suffiency itself being open to your discussion (which happens to
> > make your rather predictable contributions valuable) then it belongs.
>
> No, sir. It only belongs if it fits the charter.
>
We've been through this. I've reviewed the language which explicitly
makes this ng the appropriate place for boundary issues including the
goal of a programming profession.
> >
> > Suppose there was a vacuum tube Internet in 1948 wherein physicists
> > logged on to several groups concerned with fission and fusion bomb
> > technology.
>
> There wasn't.
Oooo. Touche. Ever hear of a gedanken experiment?
>
> >
> > Suppose that MOST of the groups were
> > sci.physics.blowthemalltohell.neutrinos,
> > sci.physics.blowthemalltohell.cladding,
> > sci.physics.blowthemalltohell.testingthegoddamnthing: groups
> > concerned, that is, with various subgoals of the Los Alamos project.
>
> I suspect there are plenty of groups dealing with the sociology of
> technology.
I have no interest in talking with self-appointed sods who know
NOTHING about code. These people are academic rejects in search of a
meal ticket and they have a pernicious tendency to be appointed by
management to control those of us who know our trade, in order to
prevent the worst thing in the world in American management praxis,
which is any meaningful worker control of the production process.
Furthermore, American sociology has been completely corrupted by
quants who don't even know the math which they vastly prefer to
anything like theory. There are no C. Wright Mills anymore, only
quantitative lawn trolls who I used to help with Fortran in the 1970s.
When I talk to these folks about Adorno, they in signal cases don't
know who Adorno was!
Edward G. Nilges
Missing here is the idea of structure itself: for note that your
implicit taxonomy is an unstructured set or collection of topics in an
unordered list.
In this taxonomy, C programming is followed more or less randomly by
Britney Spears, and America's sweetheart is followed, more or less
randomly, by backgammon. Oops, I thought I saw "the sociology of
programming." I thought I saw, in fine, a puddy tat.
The amusing irony is that structured texts, with well-formed
subclauses which phenomenonlogically relate chains of signifiers, are
in this Internet culture tagged with the crime of verbosity: whereas
the unordered list is not considered prolix.
In my example, the counterfactual physicists had set up a workplace
structure in which the toplevel ng was for broad, general discussions
which necessarily would involve the Kantian intersection of cognition
and ethics. In such a ng, the technical feasibility of fusion would
have been subordinated by an Oppenheimer to its morality.
And, that subsumption would not itself be a purely ethical choice.
For Oppenheimer grasped the Kantian essence, which was the profound
IRRATIONALITY of physicists, standing on the shoulders of history and
giants, constructing the physical negation of history and its
replacement by Jonathan Schell's "republic of insects and grass."
In a micro fashion, I perceive a relationship between the fact that
companies like Enron used sophisticated technology for crude and
brutal deception, and the widespread disinterest in America in mere
software correctness, which extends to distaste for logical argument
about the behavior of code, and the confusion of negative critique
with discourtesy (a confusion which damaged the career of hero
computer scientist Dijkstra.) The Kantian link is that ethical
dishonesty equates to intellectual dishonesty.
Now, I am sure that the fusion bomb "works" although note that it,
like Star Wars, can't be live tested with real data as can ordinary
code: for this would be to actually conduct nuclear war. But as a
layperson my guess is that Ed Teller was inferior as a scientist to
Oppenheimer.
>
> I spent a little time trying to find some for you, even though it's
> *your* job to find a newsgroup where your stuff is topical. But guess
> what? My news server carries ***so many groups*** that Lotus 1-2-3
You prove my point. In the absence of structure, the server is
overwhelmed.
> couldn't handle them all in a convenient form. Never mind - you're
> clearly interested in legal and political issues, so here are my grep
> hits for legal, law, politics, political; I suggest you try some of
> these groups instead of comp.programming when you want to discuss legal
> and political issues:
>
> 5col.politics
> ab.politics
> alabama.politics
Oh great, a bunch of barefoot peckerwoods will just love to hear from
me...
> alt.arabic.politics
Terrific, inshallah, just the site...God is Great...
> alt.culture.arabic.non-politics
Bwa most assuredly ha
> alt.culture.turkish.politics
> alt.discuss.politics
> alt.fan.jenn.politics
> alt.hemp.politics
Oh wow, just my ng...
> alt.irc.efnet.politics
> alt.irc.efnet.politics.suck
> alt.irc.efnet.politics.suck.suck
> alt.irc.efnet.politics.suck.suck.suck
> alt.men.politics
Lock up the ladies' room!
Mock on, mock on, Voltaire, Rousseau. You prove my point. Knowledge
becomes the list which even on computer science's OWN terms (see any
good text on relational data base theory) is unnormalized. They have
brought deep Heaven down on their heads.
>
>
>
>
>
> >
> > Suppose, also, that there was a father group called, of course,
> > sci.physics.blowthemalltohell.
> >
> > Suppose J. Robert Oppenheimer logged on and posted the thread "SHOULD
> > we blow them all to hell?", or perhaps "SHOULD we use thermonuclear
> > fusion to blow them all to goddamn hell?"
> >
> > In this *gedanken* experiment, I PREDICT that Oppie would have done
> > this and with equal certitude, I PREDICT that Ed Teller would have
> > taken upon
>
> Your predictions are unfalsifiable and thus of no value in scientific
> discourse.
>
Nonsense: this only shows that Sir Karl was popping off in deep
ignorance. For if counterfactuals are of use in mathematics and
science, this shows that a logician with Popper's committment to
falsifiability has to address their usability.
Logic here is not used as logic, instead as a police action to control
the form of Orientalism that most worries the metropolis: when the
Oriental shows that Enlightenment itself is not necessarily
metropolitan.
> > Dickie Heathfield's
>
> Childish ad hominem attacks again? Really, I thought you might have
> learned your lesson from last time. If you want to be taken seriously
> (and you clearly do), then don't throw your credibility (such as it is)
> into the fire by trying to mock people's names. It's about as logical as
> mocking their shoe size.
>
>
> > thankless task: of enforcing
> > BOUNDARIES, for, don't you know, without BOUNDARIES, we'd be...free,
> > or something like that.
>
> Oh, don't misunderstand me. You're perfectly free to post your off-topic
> stuff here. Please don't forget, however, that I am equally free to post
> my opinion about how it would be better for all concerned, including
> you, if you would find a newsgroup where your stuff is topical.
>
Thanks for your tolerance. And, I suggest, that the discussion is
healthy for it allows both of us to clarify our views.
Civis Romanus sum: I have no desire to be coded as a monster who is
off topic for the sake of being off topic, for the simple reason I am
not.
>
> >
> > And, with all due respect, I suggest that Ed Teller would have been
> > disingenuous as indeed Dickie is here being disingenuous.
>
> Another ad hominem attack? If you had a serious point, you'd make it,
> instead of fooling around.
Using "Dickie" is mere discourtesy, not technical ad hominem.
Instead, I shall just drop the usage because this is your second
request that you do so. I have neglected an earlier committment to do
so.
>
> <OT stuff snipped>
> >
> > This *gedanken* experiment SHOWS how focus is political.
>
> It shows *nothing* except your unfalsifiable and thus pointless
> interpretation of it.
Popper, again...
Popper was wrong.
Programmer Dude
> I'm well aware of what malloc does. Actual experience shows that
> malloc and free cause memory leaks,...
malloc() and free() **cause** memory leaks?
Heh, *my* Actual Experience shows that it's due to bad programming.
> ...and that objects, which don't allow you to malloc, don't cause
> as many problems.
Really.
void Hmmmmmmm (void)
{
SomeClass* new_object ;
new_object = new SomeClass();
new_object = new SomeClass();
new_object = new SomeClass();
new_object = new SomeClass();
new_object = new SomeClass();
new_object = new SomeClass();
}
Oh, by the way:
class SomeClass {
void* p ;
public:
SomeClass () { p = malloc (10000); }
};
Good thing objects (which don't allow you to malloc) are so trouble-
free, eh?
> malloc was an early 1990s solution to problems whose complexity
> outruns its capabilities.
Heh. Whadda ya think is under the hood of most 'new' implementations?
What, really, is the difference between:
==================================================================
void Huge_Buffer_1 (size_t sz)
{
char* buf = malloc (sz);
/* do stuff with buf */
return; /* are we forgetting anything????? */
}
==================================================================
And:
==================================================================
class HugeBuffer {
char* buf ;
public:
virtual ReadBuffer (char* data, size_t offset, size_t amount) = 0;
virtual WriteBuffer (char* data, size_t offset, size_t amount) = 0;
HugeBuffer (size_t sz): buf(NULL) { buf = new char[sz]; }
virtual ~HugeBuffer () { delete[] buf; }
};
void Huge_Buffer_2 (size_t sz)
{
HugeBufferSubClass* hbsc = new HugeBuffer(sz);
/* do stuff with hbsc */
return; /* are we STILL forgetting something????? */
}
==================================================================
> It shouldn't be used in an applications program. It should only be
> used in an OO runtime to support higher-level capabilities.
Just because you can't use your toys effectively, don't think for a
moment we all can't.
Programmer Dude
>> Nonsense. My son is currently writing a computer game. He's 11 years
>> old. I am supervising the quality of his work and, indeed, doing much of
>> it myself, but he is certainly contributing. Am I to make him write a
>> sociological essay every time he learns a new programming concept, and
>> to forbid him to practise real world use of that concept in his game
>> until the essay is handed in? Stuff and nonsense!
>
> Interesting: for in older cultures, skilled workers would indeed use
> training time to mention linkages with topics outside the workplace
> per se.
And you assume Richard isn't?
> I suggest that the picture of the father and son working together
> SILENTLY is an American cultural artifact,....
And you assume Richard is American?
> These "real people" treat this ng as the place for random idiotic
> questions which have, as a matter of fact, specialized ngs which the
> posters are too goddamn lazy to find.
Then you're in good company.
> I have no interest in talking with self-appointed sods who know
> NOTHING about code.
Yet you expect us to deal with a self-appointed sod who knows NOTHING
about USENET. Clearly demonstrated by your utter inability to snip
the huge list of political newsgroups Richard posted.
If we pool our resources and send you a dollar will you buy a clue?
Please??
> blah, blah, blah,...praxis,...
HA! I wondered how long it'd be before you used praxis in a sentence.
(Knew it wouldn't be long; it appears a favorite hobby horse.)
> Furthermore, American sociology has been completely corrupted by
"Completely", eh? No, no paranoia there...
> quants who don't even know the math which they vastly prefer to
> anything like theory. There are no C. Wright Mills anymore, only
> quantitative lawn trolls who I used to help with Fortran in the 1970s.
> When I talk to these folks about Adorno, they in signal cases don't
> know who Adorno was!
Do you understand you're displaying all the signs of The True Kook?
Everyone else is an idiot.
No one gets it but you; only you perceive the Truth.
No sense of USENET netiquette at all (755 lines of unsnipped wastage).
Definite Kook.
Why should we take you seriously when you can't even form an appropriate
post? Should be much simpler than forming a correct program.
John Doherty
spino...@yahoo.com (Edward G. Nilges) wrote:
> I entered programming in 1973 because Watergate made it seemed that
> BOTH sides (Nixon and his opponents) were completely corrupt...
Man, you are one whacked dude.
--
Richard Heathfield
>
> Ian Woods <new...@wuggy.org> wrote in message news:<Xns9223F21B1298...@217.32.252.50>...
> > spino...@yahoo.com (Edward G. Nilges) wrote in
> > news:f5dda427.02060...@posting.google.com:
> >
> > >
> > > The political problem of malloc is that it FAILS to discriminate and
> > > mallocs mere bytes, which can then hold any kind of pernicious text.
> > > I prefer to create objects for this reason.
> >
> > Not so. It allocates storage for an object, or informs you that it cannot
> > allocate storage for an object the size you requested. Importantly, it
> > doesn't specify that the allocated space for your object "holds any kind of
> > pernicious text" - merely that it's value is indeterminate. Here is the
> > relavent section about mallocs behaviour from ISO/IEC 9899:1999...
> >
> > 7.20.3.3 The malloc function
> >
> > Synopsis
> >
> > 1 #include <stdlib.h>
> > void *malloc(size_t size);
>
> I'm well aware of what malloc does. Actual experience shows that
> malloc and free cause memory leaks,
Nonsense! Bad programming causes memory leaks. The malloc function
allocates memory, and the free function deallocates it. It is the
programmer's responsibility to use them correctly. It's a poor workman
who blames his tools.
> and that objects, which don't
> allow you to malloc, don't cause as many problems.
Objects cannot stop me from using malloc. Nor do they.
>
> malloc was an early 1990s solution to problems whose complexity
> outruns its capabilities.
The malloc function dates at least as far back as 1978 (K&R1). Are you
always this careless with your "facts"?
>
> It shouldn't be used in an applications program.
Why not? It's fine if it's used *properly*.
> It should only be
> used in an OO runtime to support higher-level capabilities.
Stuff and nonsense. Still, I suppose at least we're back on topic.
Richard Heathfield
>
> verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
<snip>
> >
> > tell me what language do you think I'm talking about here....
>
> Probably, C or a similar out-of-date language.
And C would be ou- of-date compared to... what, exactly? What language
of comparable power, portability and flexibility is more up-to-date than
C?
Richard Heathfield
"Edward G. Nilges" wrote:
> Richard Heathfield wrote...
> > "Edward G. Nilges" wrote:
<snip>
> > > For the same reason a programmer should not be permitted to even CODE
> > > a Or b without being able to describe, accurately, the evaluation (in
> > > particular whether his language short-circuits) programming happens in
> > > a scientific and industrial context.
> >
> > Nonsense. My son is currently writing a computer game. He's 11 years
> > old. I am supervising the quality of his work and, indeed, doing much of
> > it myself, but he is certainly contributing. Am I to make him write a
> > sociological essay every time he learns a new programming concept, and
> > to forbid him to practise real world use of that concept in his game
> > until the essay is handed in? Stuff and nonsense!
> >
> Interesting: for in older cultures, skilled workers would indeed use
> training time to mention linkages with topics outside the workplace
> per se. I suggest that the picture of the father and son working
> together SILENTLY is an American cultural artifact, for my own German
> grandfather worked with his sons not in silence but instead yelling at
> them about the right way, the wrong way, his way, the highway, the
> virtues of Ohio Republicanism and the perfidy of the Copperheads.
>
> Only in America are sons supposed to reinvent the wheel.
This is a complete non sequitur. Did you read what I wrote? Apparently
not. As for working silently, well, we don't yell at each other all the
time, but we're far from silent. I think you just make stuff up.
>
> > >
> > > Indeed, it is a nerd fantasy that programming exists in a vacuum.
> >
> > Let me get the terminology straight, here: nerds are those people who
> > have zits, jamjar specs, brains the size of planets and social lives the
> > size of walnuts, right? Well, I don't know anyone like that. Not even
> > one person. Perhaps you're less fortunate.
>
> No: for under the Popcult lies the reality which Cold War America
> happened to share with the Soviet Union.
Your reply has nothing to do with programming.
<snip>
> I think this happened here, too. I entered programming in 1973
Sure about that? You don't seem to know much about it, after all.
<irrelevant stuff snipped>
> [...] at the time software development was so unknown as to be
> rife with interesting characters.
It still is.
>
> >
> > It is a peculiarly Nilgesque fantasy that all aspects of anything
> > remotely related to information technology are topical for
> > comp.programming.
> >
> It's not a fantasy, Dickie.
Another childish ad hominem attack. You seem to resort to these when
you're most stuck for technical arguments.
> Programming is a phenomenology for it is
> ABOUT something. It is the writing of texts, which are not
> things-in-the-world but *qua* texts refer to something outside
> themselves, with the exception, of course, of null For loops.
What do you mean by "null For loops"? Is this one? Does it refer to
something outside itself?
for(p = q; p; p = p->next) { }
> > > If the exogenous issue arrives at a sufficient scale of importance,
> >
> > Please define "exogenous". It looks to me as if it should mean
> > "originating from outside". Is that right? I'll answer as if it means
> > that, and you can correct me if I'm wrong.
>
> Correct.
> >
> > Now, if it does mean that, then really it doesn't matter how important
> > *you* think it is or how important *I* think it is - what matters is how
> > important it is to those people who read this group for its intended
> > purpose - discussions about programming.
>
> These "real people" treat this ng as the place for random idiotic
> questions which have, as a matter of fact, specialized ngs which the
> posters are too goddamn lazy to find.
This is, in fact, largely true (although the people to whom you refer
tend to be the writers to the newsgroup rather than the readers of it).
>
> You appoint yourself only the defender of narrowness: yet I do not see
> where you critique the converse error, of posting a question for which
> a MORE SPECIALISED ng exists!
<shrug> When I remember, I point C people to clc, C++ people to clc++,
Java people to cljp, Unix people to cup, DOS people to
comp.os.msdos.programmer, Windows people to
comp.os.ms-windows.programmer.win32, etc. Admittedly not all the time,
because I don't always remember.
> I can only conclude that your brief is
> for narrowness, not the purity of this newsgroup.
I can only conclude that you're not brief.
>
> >
> > > the suffiency itself being open to your discussion (which happens to
> > > make your rather predictable contributions valuable) then it belongs.
> >
> > No, sir. It only belongs if it fits the charter.
> >
> We've been through this. I've reviewed the language which explicitly
> makes this ng the appropriate place for boundary issues including the
> goal of a programming profession.
We've been through this, as you say. You may have reviewed the language,
but that doesn't mean that the language means what you want it to mean.
It means what it says - this newsgroup is for discussing programming,
not politics.
>
> > >
> > > Suppose there was a vacuum tube Internet in 1948 wherein physicists
> > > logged on to several groups concerned with fission and fusion bomb
> > > technology.
> >
> > There wasn't.
>
> Oooo. Touche. Ever hear of a gedanken experiment?
Sure. So what?
> >
> > >
> > > Suppose that MOST of the groups were
> > > sci.physics.blowthemalltohell.neutrinos,
> > > sci.physics.blowthemalltohell.cladding,
> > > sci.physics.blowthemalltohell.testingthegoddamnthing: groups
> > > concerned, that is, with various subgoals of the Los Alamos project.
> >
> > I suspect there are plenty of groups dealing with the sociology of
> > technology.
>
> I have no interest in talking with self-appointed sods who know
> NOTHING about code.
If not even *you* will talk with yourself, why should we?
<political stuff snipped>
> Furthermore, American sociology
...is off-topic here.
<snip>
Programmer Dude
>>> Suppose there was a vacuum tube Internet in 1948 wherein physicists
>>> logged on to several groups concerned with fission and fusion bomb
>>> technology.
Considering the secrecy surrounding those projects, they'd all have been
shot for treason.
Even if secrecy were maintained, I VERY much doubt either reknowned
physicst would find much real value in USENET and certainly not in the
opinions of the unwashed masses.
USENET is best compared to a very large park where everyone stands on
their own soapbox proclaiming their testiment. A few listen and a few
learn a thing or two. But to a huge degree, USENET is write-only.
If found your thought experiment completely specious, but am willing
to do a bit of deconstruction...
> Suppose there was a vacuum tube Internet in 1948 wherein physicists
> logged on to several groups concerned with fission and fusion bomb
> technology.
Okay. Are we assuming it's relatively private and restricted? Lets.
> Suppose that MOST of the groups were
> sci.physics.blowthemalltohell.neutrinos,
> sci.physics.blowthemalltohell.cladding,
> sci.physics.blowthemalltohell.testingthegoddamnthing: groups
> concerned, that is, with various subgoals of the Los Alamos project.
I reject this assumption (thereby invalidating most of what follows,
but we'll slog through nevertheless). Why would they choose such
language? Nothing about these men suggests such an absurd possibility,
and--to the contrary--quite a bit suggests otherwise.
> Suppose, also, that there was a father group called, of course,
> sci.physics.blowthemalltohell.
Supposition rejected for the above reasons.
> Suppose J. Robert Oppenheimer logged on and posted the thread "SHOULD
> we blow them all to hell?", or perhaps "SHOULD we use thermonuclear
> fusion to blow them all to goddamn hell?"
Let's suppose more sensibly that JRO opened a discussion on whether
the bomb should be used against fellow humans....
> I PREDICT that Oppie would have done this and with equal certitude,
Rejected. You've demonstrated no basis for this prediction.
> I PREDICT that Ed Teller would have taken upon Dickie Heathfield's
> thankless task: of enforcing BOUNDARIES,
Rejected. You've demonstrated no basis for this prediction.
SO far you seem to be wrapping some weird ad homenum attack on Richard
(who asked you LAST time to refrain from the self-evidently belittling
"Dickie"). Considering that Richard is a valued poster and you're an
occasional Visiting Kook, you're not positioning yourself well here.
> And, with all due respect, I suggest that Ed Teller would have been
> disingenuous as indeed Dickie is here being disingenuous.
Rejected. I submit that Teller would have been straightforward and not
disingenuous. As is Richard. Who is correct. None of this (so far)
is very on topic, and you don't even have the excuse of topic drift.
You were off-topic from the starting gate.
On the other hand, we tend to be fairly tolerant of that here. Richard's
opinion is his own. As is yours. And mine. And theirs.
> For we know that the real Ed Teller thought it axiomatic that we needed
> the capability to blow them all to hell, using thermonuclear technology.
And I submit that rather than these idiotic games we're playing, Teller
and Oppie would have respectfully debated and argued position and, as
*scientists* done their best to support those positions.
You've done a lot of whining about the state of things and a lot of
hand wringing and hand waving, but when pressed for substance you come
off rather vaporous.
> Missing here is the idea of structure itself: for note that your
> implicit taxonomy is an unstructured set or collection of topics in an
> unordered list.
No doubt if Richard had felt like taking the time to sort the list, he
would have.
> The amusing irony is that structured texts, with well-formed
> subclauses which phenomenonlogically relate chains of signifiers, are
> in this Internet culture tagged with the crime of verbosity: whereas
> the unordered list is not considered prolix.
A. Says who?
B. Your opinion of your own writing not withstanding, the truth is
your ability to communicate and persuade is sadly lacking. Proof
that education .NE. wisdom.
> In a micro fashion, I perceive a relationship between the fact that
> companies like Enron used sophisticated technology for crude and
> brutal deception, and the widespread disinterest in America in mere
> software correctness, which extends to distaste for logical argument
> about the behavior of code, and the confusion of negative critique
> with discourtesy (a confusion which damaged the career of hero
> computer scientist Dijkstra.) The Kantian link is that ethical
> dishonesty equates to intellectual dishonesty.
Good Gravey! Last night, on one of the educational channels, I saw
a bit about The Silver Spade. A "stripping shovel", used for for the
mining of coal, that stands about 20 stories high and can move tons
of earth in a single gulp.
But, I fear, not up to the task of shoveling through the above.
> Now, I am sure that the fusion bomb "works" although note that it,
> like Star Wars, can't be live tested with real data as can ordinary
> code: for this would be to actually conduct nuclear war.
Or, say, blowing up a small island in the Pacific.
> Civis Romanus sum: I have no desire to be coded as a monster who is
> off topic for the sake of being off topic, for the simple reason I am
> not.
In this corner, you're coded as a one-note kook who's still stewing
over past hurts.
> Using "Dickie" is mere discourtesy, not technical ad hominem.
A rather fine and meaningless distinction, "Eddie".
Richard Heathfield
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFF0E72...@eton.powernet.co.uk>...
<snip>
> > I suspect there are plenty of groups dealing with the sociology of
> > technology.
>
> Missing here is the idea of structure itself: for note that your
> implicit taxonomy is an unstructured set or collection of topics in an
> unordered list.
It's *your* job to structure your universe in a way that leads you to
the right newsgroup. The best I can do is give you a few ideas to work
with.
<LOTS of off-topic nonsense snipped>
> > I spent a little time trying to find some for you, even though it's
> > *your* job to find a newsgroup where your stuff is topical. But guess
> > what? My news server carries ***so many groups*** that Lotus 1-2-3
>
> You prove my point. In the absence of structure, the server is
> overwhelmed.
No, the server isn't overwhelmed. You are. Go find the right group.
> > couldn't handle them all in a convenient form. Never mind - you're
> > clearly interested in legal and political issues, so here are my grep
> > hits for legal, law, politics, political; I suggest you try some of
> > these groups instead of comp.programming when you want to discuss legal
> > and political issues:
> >
<snip>
>
> > alt.culture.turkish.politics
> > alt.discuss.politics
> > alt.fan.jenn.politics
> > alt.hemp.politics
>
> Oh wow, just my ng...
It figures.
>
> > alt.irc.efnet.politics
> > alt.irc.efnet.politics.suck
> > alt.irc.efnet.politics.suck.suck
> > alt.irc.efnet.politics.suck.suck.suck
> > alt.men.politics
>
> Lock up the ladies' room!
>
> Mock on, mock on, Voltaire, Rousseau. You prove my point. Knowledge
> becomes the list which even on computer science's OWN terms (see any
> good text on relational data base theory) is unnormalized. They have
> brought deep Heaven down on their heads.
No, I haven't proved your point. I've just provided you with a first cut
of newsgroups where your pseudoscience might be topical. It's your job
to filter it further until you find the right group for you.
<snip>
> > > In this *gedanken* experiment, I PREDICT that Oppie would have done
> > > this and with equal certitude, I PREDICT that Ed Teller would have
> > > taken upon
> >
> > Your predictions are unfalsifiable and thus of no value in scientific
> > discourse.
> >
> Nonsense:
Prove it. Explain how your predictions can be falsified. Explain how, in
fact, they can even be said to be predictions. And do it all in a way
that is topical in comp.programming, or take it to some group where they
care.
<snip>
> Logic here is not used as logic,
When you write this kind of nonsense, I really do wonder why I bother
reasoning with you at all. PLEASE GO FIND A GROUP WHERE NONSENSE IS
TOPICAL.
<more nonsense snipped>
> > > thankless task: of enforcing
> > > BOUNDARIES, for, don't you know, without BOUNDARIES, we'd be...free,
> > > or something like that.
> >
> > Oh, don't misunderstand me. You're perfectly free to post your off-topic
> > stuff here. Please don't forget, however, that I am equally free to post
> > my opinion about how it would be better for all concerned, including
> > you, if you would find a newsgroup where your stuff is topical.
> >
> Thanks for your tolerance.
I have no tolerance for your idiocy. If I *could* stop you posting here,
I probably *would*. Freedom of speech is all very well, but IMHO you are
abusing that freedom, to the detriment of those who would seek to use
this group for its intended purpose.
> And, I suggest, that the discussion is
> healthy for it allows both of us to clarify our views.
Your view appears to be "I am American, therefore the whole world is
American, and consequently comp.programming must be particularly
interested in the sociology of American politics". Have I correctly
understood you?
>
> Civis Romanus sum:
Really? Are you sure about that?
> I have no desire to be coded as a monster who is
> off topic for the sake of being off topic,
Then stop doing it.
> for the simple reason I am
> not.
No, you're not a monster, just an idiot. Please stop being an idiot.
> > > And, with all due respect, I suggest that Ed Teller would have been
> > > disingenuous as indeed Dickie is here being disingenuous.
> >
> > Another ad hominem attack? If you had a serious point, you'd make it,
> > instead of fooling around.
>
> Using "Dickie" is mere discourtesy, not technical ad hominem.
It's certainly very childish, and does your cause no good.
> Instead, I shall just drop the usage because this is your second
> request that you do so.
Actually, all I did was point out that you are being childish. It is
entirely up to you just how childish you want to be.
> I have neglected an earlier committment to do so.
I am aware of that.
<snip>
Al Dunbar
"Edward G. Nilges" <spino...@yahoo.com> wrote in message
news:f5dda427.02060...@posting.google.com...
> > > The "Data Quality Act" was recently passed under corporate pressure to
> > > remove interpretations and data on public website that these interests
> > > feel to be incorrect or unscientific.
> > >
> > > the same standards for data quality in electronic media that range
> > > from known deception in investor information to poor consumer credit
> > > reporting and sloppy data processing that has enabled identity theft.
> > >
> > > "Equal protection of the laws" means equal protection of the laws.
> > > Therefore, I'd call upon concerned people to apply this act to
> > > corporate communications.
> >
> > Hey, I tried that earlier today, but was laughed out of the office for
> > presuning that a US law could be used that way in Canada. Actually, they
> > arrested me, but forgot to "Mirandize" me! I am still awaiting a call
back
> > from that DA guy on "Law and Order" to explain to our police how they
need
> > to observe the laws of all other countries. I wonder why he has not
called?
>
> Now, here I am indeed at fault. I did not mean to imply that our
> Constitution should obtain in Canada. But don't you fellows have
> something like the abstract principle expressed in "equal protection
> of the laws."
The "Charter of Rights and Freedoms" covers much of the ground. Charter
challenges have resulted in the overturning of laws that would seem to
impinge on rights granted by the charter. I think, in fact, that our laws
are so good that we need not turn to yours (which have not been passed here
and could not legally apply) for remedy.
> I don't remember Sergeant York of the Yukon or his
> faithful dog reading the Metis their rights, but I believe English
> common law applies in Canada, and it contains at least the idea of
> equal protection.
Quite so. Regardless of the laws, there have of course been abuses. If there
weren't, we wouldn't need the laws.
> We had to pass the 14th amendment to get the common law principle in
> writing because our slavedriver class thought that laws only applied
> to white people. However, since American slaves fled in signal
> instances to Canada, I can only conclude that you fellows have a
> progressive legal system, barring the occasional anti-terrorism
> hysteria such as obtained under Trudeau: an anti-terrorism hysteria
> which we undergo today.
I would generally concur, however, being truly Canadian, I tend to think I
must be wrong when I start to think that we must in some way be better than
some other nationality.
/Al
Al Dunbar
Exactly. And I live within about 5km of the largest mall in the world!
/Al
Al Dunbar
Actually, I find it interesting that Edward seemingly rails against "the
data quality act" while at the same time wishing that somebody (or, perhaps,
some body) would try to micromanage what programmers do.
/Al
Al Dunbar
"Edward G. Nilges" <spino...@yahoo.com> wrote in message
news:f5dda427.02060...@posting.google.com...
news:<3CFF0E72...@eton.powernet.co.uk>...
> > [Followups set to comp.programming only, AGAIN - I STILL see no reason
> > why they should be dragged through a topicality debate unless they so
> > choose, despite the apparent ignorance of this Nilges fellow of the
> > simple courtesies of Usenet.]
> >
> >
> > "Edward G. Nilges" wrote:
> > >
> > > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message
news:<3CFDA74B...@eton.powernet.co.uk>...
> > > > [Followups set to comp.programming only - I see no reason why they
> > > > should be dragged through a topicality debate unless they so
choose.]
> > > >
> > > > "Edward G. Nilges" wrote:
> > > > >
> > > > > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message
news:<3CFCAB84...@eton.powernet.co.uk>...
> > > > > > "Edward G. Nilges" wrote:
<snip>
> I think this happened here, too. I entered programming in 1973
> because Watergate made it seemed that BOTH sides (Nixon and his
> opponents) were completely corrupt.
That's odd, I became involved in local politics when I realized that some
compilers contained bugs.
Seriously, I think your statement above illlustrates what appears to be a
fact: you are more interested in talking politics than discussing
programming issues.
/Al
Al Dunbar
"Programmer Dude" <cjso...@mmm.com> wrote in message
news:3CFFCEDA...@mmm.com...
Actually, I disagree with you here. Given the bizarro world that Edward has
supposed into existence, based on a 1948 internet, I find it quite likely
that everything and everyone would behave completely illogically in it.
/Al
Al Dunbar
> "Edward G. Nilges" wrote:
>
> > The political problem of malloc...
>
> The "political problem of malloc"?!?!
Yeah, that's right up there with the sexual problem of user interfaces. Or
the periodic problem of a table of elements.
/Al (aka the garden gnome :-)
verulam
> "Edward G. Nilges" wrote:
> >
> > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
>
> <snip>
> > >
> > > tell me what language do you think I'm talking about here....
> >
> > Probably, C or a similar out-of-date language.
>
> And C would be ou- of-date compared to... what, exactly? What language
> of comparable power, portability and flexibility is more up-to-date than
> C?
he's a VB <cough> programmer
verulam
There is no generic, good for all language <troll>(C++ programmers
pipe down, I don't want to hear it (most of you don't even use OO any
way so why not use C :->))</troll>
There are even situations where Assembly language is the best choice
in a given situation
Richard Harter
<bin...@eton.powernet.co.uk> wrote:
>"Edward G. Nilges" wrote:
[snip Nilges' bilge & respondent troll food]
>> I'm well aware of what malloc does. Actual experience shows that
>> malloc and free cause memory leaks,
>
>Nonsense! Bad programming causes memory leaks. The malloc function
>allocates memory, and the free function deallocates it. It is the
>programmer's responsibility to use them correctly. It's a poor workman
>who blames his tools.
This is wrong in ways that are relevant to comp.programming.
(alt.foozball et al removed from the newsgroup line)
The maxim "It's a poor workman who blames his tools" expresses an
important truth; it also can be used, as it is in this instance, as a
defence for shoddy and inadequate tools.
Using malloc/free as a general purpose memory management methodology
is A BAD THING TO DO, bad because the methodology facilitates memory
management errors.
Richard Harter, c...@tiac.net,
http://home.tiac.net/~cri, http://www.varinoma.com
As I was saying to my knee the other day,
What's a joint like you doing in a nice guy like me?
ke...@hplb.hpl.hp.com
> malloc and free cause memory leaks,
No, they don't.
Manual store management is prone to memory leaks, and if you're using
C's malloc-and-free combination, then they will "cause" the leaks, as
in, some store allocated by malloc won't be free'd.
> and that objects, which don't allow you to malloc, don't cause as many
> problems.
Erm ... you can have languages which have both objects (by reasonable
definition) and the moral equivalent of malloc/free - for example, C++
with new/delete.
I think you mean that languages with automatic store allocation (Lisp,
Simula, Pop11, Java), which also support "objects" well, are less
prone to memory leaks. But it's not the "objects" that are the cause.
> malloc was an early 1990s solution
Early 1990s? Is this a typo? Because malloc predates the 1990s. C code
in that style predates the 1980s - K&R I has a free-store-management
example using alloc/free, starting on p173, with a publication date
of 1978.
> to problems whose complexity outruns its capabilities.
That's not clear. One can over-reach oneself in any language; malloc/free
are probably about right for C's objectives. It's hard to see how to do
better *and* stay small *and* stay general.
> It shouldn't be used in an applications program. It should only be
> used in an OO runtime to support higher-level capabilities.
Not all programming problems call for OO runtimes, and those that do
don't all call for implementations using malloc/free. I think the term
"applications program" is too fuzzy to be decisive here.
--
Chris "electric hedgehog" Dollin
C FAQs at: http://www.faqs.org/faqs/by-newsgroup/comp/comp.lang.c.html
C welcome: http://www.angelfire.com/ms3/bchambless0/welcome_to_clc.html
Richard Heathfield
Now all is clear. Thank you.
Richard Heathfield
>
> On Thu, 06 Jun 2002 21:57:47 +0100, Richard Heathfield
> <bin...@eton.powernet.co.uk> wrote:
>
> >> shows that malloc and free cause memory leaks,
> >
> >Nonsense! Bad programming causes memory leaks. The malloc function
> >allocates memory, and the free function deallocates it. It is the
> >programmer's responsibility to use them correctly. It's a poor workman
> >who blames his tools.
>
> This is wrong in ways that are relevant to comp.programming.
Hoorah! I disagree that it's wrong, but never mind that for now. At
least it's relevant.
> (alt.foozball et al removed from the newsgroup line)
Again!
>
> The maxim "It's a poor workman who blames his tools" expresses an
> important truth;
Yes. Apart from the obvious interpretation (a poor workman doesn't
realise he's a poor workman and thinks his tools are to blame), there
are other interpretations, equally valid: a poor workman buys cheap
tools, and then complains when they don't do a proper job; a poor
workman doesn't maintain his tools properly, and then complains when
yadayada.
> it also can be used, as it is in this instance, as a
> defence for shoddy and inadequate tools.
Certainly not. There is nothing shoddy or inadequate about malloc or
free.
>
> Using malloc/free as a general purpose memory management methodology
> is A BAD THING TO DO,
No, it's not.
> bad because the methodology facilitates memory
> management errors.
By the same token, iteration is a bad thing to use because it
facilitates off-by-one errors.
CBFalconer
With any luck, I have just PLONKED this alleged thread.
--
Chuck F (cbfal...@yahoo.com) (cbfal...@worldnet.att.net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home.att.net> USE worldnet address!
verulam
> It's not a fantasy, Dickie. Programming is a phenomenology for it is
> ABOUT something.
Ah so it's a movement based on this, originated about 1905 by Edmund
Husserl?
[snip]
snip your posts...
nonincrease, compendium, contraction, short, shorten, disjuction
Never fear to use little words. Big, long words name little things.
All big things have little names, such as life and death, war and
peace, dawn, day, night, hope, love and home. Learn to use little
words in a big way. (Anonymous - Reducing Jargon)
verulam
> I think it also has a great deal to do with that our industry is
> unregulated. ANYONE can claim to be a programmer, and many lack
> the most basic skills.
absolutely, I have a degree in Computer Science but I wouldn't trust a
great proportion of my fellow graduates that year with some
programming.
I'm looking at some code now that has been written by some VB idiot (I
know it's easy to bash them so I shoudn't) no consistent coding style,
global and local variables with the same name, Goto's littered all
over the place, an insane approach to logic, huge arrays of predefined
length, poorly named functions, little or no error handling and to top
it off there's 6 and a half thousand lines of the rubbish (although
you can take 1 thousand away as commented out without explanation)
verulam
> Using malloc/free as a general purpose memory management methodology
> is A BAD THING TO DO, bad because the methodology facilitates memory
> management errors.
hmm
#define new malloc
#define free delete
better?
or VB
Set adoRecord = Nothing
(if any body cares to take the 'goes out of scope is deallocated'
argument, get your memory profiler out and take a look at what happens
to ActiveX Recordsets)
This is of course the same argument used to leave pointers out of
programming languages, people have trouble with pointers so let's
leave them out so people don't cut themselves on them. oh look no
pointers right ok, a double linked list here we go erm ok let's use
agregation, oh dear if you set up a link back to the parent it creates
a circular reference, darn, erm ok let's use a variable that's the
same size as a pointer would be and write the code to set and
dereference it so we trick the object reference counter, haha it works
"People point all the time, it saves wandering over and touching
something"
How did I get to pointers? what was my argument? I'm rambling again
Edward G. Nilges
> [snippety snip snip]
> > It's not a fantasy, Dickie. Programming is a phenomenology for it is
> > ABOUT something.
>
> Ah so it's a movement based on this, originated about 1905 by Edmund
> Husserl?
No, but a reading of Husserl shows that English and American
philosophy tends to want to destroy the link between a textual
activity and its referent.
For example, symbolic logic is unconcerned as to the referents of
symbols and it was Russell's contribution to show what could (and by
implication what could not) be done with pure syntax.
Unfortunately, this tends to blind programmers to the "forensic"
nature of their work, where such activities as copying a legal data
base are speech acts with a reference that APPEAR to be purely
mechanical and purely syntactical. In the case of copying a data
base, the making of the copy creates a potential confusion after the
first change to the copy as to which text (db) is accurate, therefore
the original act of copying was a speech act that affirmed that the
copy was accurate.
Husserl emphasizes that representation is always representation of x.
I think this is at least corrective of the fallacy that we can just
code and take no responsibility for what the code does.
>
> [snip]
> snip your posts...
> nonincrease, compendium, contraction, short, shorten, disjuction
>
> Never fear to use little words. Big, long words name little things.
> All big things have little names, such as life and death, war and
> peace, dawn, day, night, hope, love and home. Learn to use little
> words in a big way. (Anonymous - Reducing Jargon)
This is a pseudo-romanticism which is at times apposite but not, I
fear, in my case. For here your objection is not to the letter count
of my words.
For if it were, your rather idiotic list of sites in the other posts
would fail your test!
Your objection to "big" words is actually an objection to novelty and
density of ideas.
It is a lazy, pompous, and ultimately disingenuous objection, for its
Romanticism sits ill with its use in corporate venues, where it is
disciplinary and minatory.
Edward G. Nilges
Indeed, this trope commonly appears as a discourse-sumptuary move in
corporate circles and academia.
In corporate venues in modern American legal culture, there is a
strong underlying drive to own intellectual property, with the absurd
result of companies whose NAMES are arcane, "big" words (where "big"
is itself a misnomer when applied to "unusual") but whose actual
EMPLOYEES, possbly but not necessarily excluding the clown who dreamed
up "Semanticon", "Analogiphone", etc, would be hard put to parse or
define the source words including semantic, and analog.
The picture is unpleasant, for it is of course the expropriation of
literacy from an American populace who listened to the Lincoln-Douglas
debates: a trashing of a discourse commons.
And, its capillary enforcement resembles nothing so much as a
disciplining of an employee for wearing the "wrong" attire. Such
flailings are in most cases an offense to human dignity and adult
choice, as Erin Brockovich made clear.
Of course, by buddy Erin did not use big words, she wore short skirts.
But in a higher sense, just as she had the right to show off her
legs, I have the right to use just as many notes as are necessary.
For what you are trying to do is censor the very idea that your
notions are indeed wrong, and eminently replaceable by mine.
There are genuine cases of prolixity. I have just spent six hours
reducing the word count of a technical article from 12000 to 4000 per
an editor's request. But precisely because I have so much writing
experience, I know when this recommendation is made in bad faith.
I admit that in these conversations I may let my hair down, and use
arcane constructions partly for the hell of it to see what happens.
But this allows me to create what I consider useful new ideas.
Indeed, it was from discussions on the Net that I realized that
Husserl might be applicable, not only to the crisis of European
science, but also to the crisis of American programming.
Indeed, the "verbosity" is an artifact of an attempt at honesty, for
it would be all too easy to just say that I am an existentialist
programmer. I actually think instead that part of the "crisis" is the
very idea that we can sit in our little cubicles and write code that
is then used to disempower and obfuscate and confuse the consumer and
employee.
Edward G. Nilges
> > > I PREDICT that Oppie would have done this and with equal certitude,
> >
> > Rejected. You've demonstrated no basis for this prediction.
>
> Actually, I disagree with you here. Given the bizarro world that Edward has
> supposed into existence, based on a 1948 internet, I find it quite likely
> that everything and everyone would behave completely illogically in it.
>
> /Al
(Sigh) Al, you are focusing on exactly the wrong thing. I merely
wanted to illustrate how, if a rather minor technical detail were
altered as a counterfactual conditional, the Los Alamos scientists
would have treated a toplevel discussion group.
And, why would a 1948 Internet be "bizarro?" Vannevar Bush predicted
hypertext in a 1946 magazine article, and there's a book, titled "The
Victoran Internet" which shows how telegraph operators used their
"Internet" in some of the same ways we use the modern Internet.
Operators, highly skilled in Morse code (which I learned in Cub
Scouting), conducted love trysts and flame wars at slack times.
Indeed, it would assert working-class solidarity over time to forget
the very idea that the Internet is completely novel, for in both the
Victorian and the modern Net, it is seen that human beings have
capabilities (such as using big words o my) above and beyond their
station in life.
And, as it happened, Oppenheimer did indeed consider the "marginal"
issue of ethics as his proper concern as chief scientist while, as it
happened, second-rate logistics pencil pushers like General Groves did
not.
Edward G. Nilges
Yes, I have a memory. That is truly bizarre.
>
> --
Edward G. Nilges
> >
> > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
>
> <snip>
> > >
> > > tell me what language do you think I'm talking about here....
> >
> > Probably, C or a similar out-of-date language.
>
> And C would be ou- of-date compared to... what, exactly? What language
> of comparable power, portability and flexibility is more up-to-date than
> C?
C++, of course: its power and flexibility are proved by the fact that
bozos write C code using C++ compilers, and its portability is proved
by the fact that its original implementation used the C preprocessor.
You are pitching slow balls indeed (hopefully the metaphor is portable
to countries playing cricket as well as baseball.)
Edward G. Nilges
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFFCD1C...@eton.powernet.co.uk>...
> > "Edward G. Nilges" wrote:
> > >
> > > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
> >
> > <snip>
> > > >
> > > > tell me what language do you think I'm talking about here....
> > >
> > > Probably, C or a similar out-of-date language.
> >
> > And C would be ou- of-date compared to... what, exactly? What language
> > of comparable power, portability and flexibility is more up-to-date than
> > C?
>
> If you do a seach on his email on google groups I think you'll notice
> he's a VB <cough> programmer
...who developed compilers in PL/I and C in a former life. Ever hear
of reincarnation? VB is where the money is. It is a low and unseemly
language which only means it's an intellectual challenge. Plus, it
helps me to identify with the programming proletariat world-wide, and
partake in their just struggle, etc., and so forth.
I do not claim that VB is a great language. However, you might take a
look at VB.Net.
Edward G. Nilges
> spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
> > > spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > > > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.0206...@posting.google.com>...
> > > > > spino...@yahoo.com (Edward G. Nilges) wrote in message news:<f5dda427.02060...@posting.google.com>...
> > > > > > cjso...@mmm.com (Programmer Dude) wrote in message news:<3CFD1985...@mmm.com>...
> > > > > > > "Edward G. Nilges" wrote:
> [snippety snip snip]
> > > > > > And, as such, violates the 14th Amendment to the United States
> > > > > > Constitution.
> > > > >
> > > > > ahah programming at last, I suggest you malloc slightly more space if
> > > > > your stepping out of bounds on the 14th element in your array.
> > > >
> > > > The political problem of malloc is that it FAILS to discriminate and
> > > > mallocs mere bytes, which can then hold any kind of pernicious text.
> > > > I prefer to create objects for this reason.
> > >
> > > was I too subtle?
> > >
> > > tell me what language do you think I'm talking about here....
> >
> > Probably, C or a similar out-of-date language.
>
> There is no generic, good for all language <troll>(C++ programmers
> pipe down, I don't want to hear it (most of you don't even use OO any
> way so why not use C :->))</troll>
Good point.
>
> There are even situations where Assembly language is the best choice
> in a given situation
Richard Heathfield is true to type: for he infers in another post that
because at this point I use VB I am an *untermensch*.
Common to this Orientalist confusion (where by "Orientalism" I mean
the racial prejudice that codes external cues and uses them in
preference to understanding) and Richard's sage advice that I avoid
big words is a refusal, of course, to engage conceptual reality.
This is the reason why the phenomenon you name above is so common.
The head-hunters and the managers look at labels and no further.
Therefore they refuse to hire the skilled OO VB.Net coder and hire
instead some bozo who writes C in C++. Because in America we think it
somehow impolite to claim knowledge above and beyond that which is
assessed by the crudest, most demotic measures, the C++ clown is given
a free pass.
Unfortunately, this was called "getting your ticket punched" in
Vietnam and we know what that led to.
Edward G. Nilges
> "Edward G. Nilges" wrote:
>
> > I'm well aware of what malloc does. Actual experience shows that
>
> malloc() and free() **cause** memory leaks?
>
> Heh, *my* Actual Experience shows that it's due to bad programming.
Good point, but if a tool generates efficient code and doesn't require
you to desk-check to be sure that for every malloc of a given sizeof,
an equivalent release exists, aren't you really responsible for using
it, given that you are being paid to save time?
>
> > ...and that objects, which don't allow you to malloc, don't cause
> > as many problems.
>
> Really.
>
> void Hmmmmmmm (void)
> {
> SomeClass* new_object ;
>
> new_object = new SomeClass();
> new_object = new SomeClass();
> new_object = new SomeClass();
> new_object = new SomeClass();
> new_object = new SomeClass();
> new_object = new SomeClass();
> }
>
> Oh, by the way:
>
> class SomeClass {
> void* p ;
> public:
> SomeClass () { p = malloc (10000); }
> };
>
> Good thing objects (which don't allow you to malloc) are so trouble-
> free, eh?
All you've proved is that you know how to write bad code, dude. Can
you write good code? More important, do you know what constitutes a
good object? The problem is that anything can be an object, including
nameless monsters.
>
> > malloc was an early 1990s solution to problems whose complexity
> > outruns its capabilities.
>
> Heh. Whadda ya think is under the hood of most 'new' implementations?
>
> What, really, is the difference between:
> ==================================================================
> void Huge_Buffer_1 (size_t sz)
> {
> char* buf = malloc (sz);
>
> /* do stuff with buf */
>
> return; /* are we forgetting anything????? */
> }
> ==================================================================
> And:
> ==================================================================
> class HugeBuffer {
> char* buf ;
> public:
> virtual ReadBuffer (char* data, size_t offset, size_t amount) = 0;
> virtual WriteBuffer (char* data, size_t offset, size_t amount) = 0;
> HugeBuffer (size_t sz): buf(NULL) { buf = new char[sz]; }
> virtual ~HugeBuffer () { delete[] buf; }
> };
>
> void Huge_Buffer_2 (size_t sz)
> {
> HugeBufferSubClass* hbsc = new HugeBuffer(sz);
>
> /* do stuff with hbsc */
>
> return; /* are we STILL forgetting something????? */
> }
> ==================================================================
>
> > It shouldn't be used in an applications program. It should only be
> > used in an OO runtime to support higher-level capabilities.
>
> Just because you can't use your toys effectively, don't think for a
> moment we all can't.
Just because you won't grow up, don't assume I play with toys in the
first place.
Programs aren't mechanisms, 'equivalent under the covers'. Programs
are texts which express a computation. As such, it matters more than
you think how they are expressed. The CS community has been working
on this for some time, only to be retarded by people who might be
happier working at Jiffy Lube, for what they REALLY want is to
manipulate physical objects, but were driven into programming by
middle class delusion.
Edward G. Nilges
Of course not. But applying them to the wrong problem is.
>
> >
> > Using malloc/free as a general purpose memory management methodology
> > is A BAD THING TO DO,
>
> No, it's not.
>
> > bad because the methodology facilitates memory
> > management errors.
>
> By the same token, iteration is a bad thing to use because it
> facilitates off-by-one errors.
It can blind the programmer to those cases where recursion is better.
Many recursive problems are misaddressed by programs that never work
because they use iteration instead. The classic example is the
compiler written by iterating through the source text, maintaining
'tables' which soon grow out of control (I am familiar with this
blunder because I made it in 1970.)
Edward G. Nilges
Al, I pointed out a "real contradiction" between this (corporate) push
for "data quality" from corporations that don't take equivalent
responsibility for investor information, as at Enron. It would
therefore be surprising if dialectically there was not another
contradiction.
The apparent contradiction is easily resolved by a commonly accepted
set of principles which if violated would be occasion for redress.
For example, if a data system is proved to violate fair dealing (by,
let's say, inappropriate rounding) then of course, its programmers
would have to be, if you please, "micromanaged." Would this be a bad
thing?
Programmers are ALREADY micro-managed by the owners of the firm in
many cases.
The illusion is regarding any kind of oversight as additive. The
working person, reflecting on the already amount of control he
experiences on the job, considers the possibility of government
regulation as one more thing he has to deal with on an already full
plate.
This neglects the complexity of the actual reality, for if the
government regulation is (for example) a mandated limit to the length
of the working day, then the added regulation lowers the control of
the boss and lessens the sum total of "micromanaging."
And, it a situation where the employee is responsible to several
independent agencies, an interesting phenomenon discovered by a guy I
worked with at Princeton (John Nash) kicks in.
This is that a game of more than two players can have multiple
winners.
Note that in recent years and in the increasingly corporate workplace,
the employee faces only the employer in the form of his boss, and in
recent years, increasing levels of control (only cosmetically
addressed by relaxation of dress codes) mean that there are two
"players", and as John von Neumann showed, in such as game, there will
be a winner, and a loser.
Guess who gets to be the winner in the corporation. Most of the time
it is management, but I'm sure that if the employees, in this "two
person game" get the upper hand, the corporation is the loser.
Certainly, in the upper echelons (where American, and devil a doubt
Canadian, CEOs were paid the big bucks to inflate the stock price),
the upper-division employees trash the abstract company to become Easy
Winners. For example, Carly Fiorina is laying off good engineers and
trumpeting something like "HP TV" to a market sated with bad
bandwidth...for obviously only stupid people want a successor to the
HP 48.
Whereas if the employee has an independent professional association he
can appeal his treatment to that association. Or, when the
association's demands become too burdensome, the employee can use his
employer to get the association to moderate its demands, as when a
journeyman engineer is able to work on a project without all the
necessary diplomas.
Corporate propaganda accentuates the bad features of unions such as
dues (and I am well aware of their problems including British
bloody-mindedness and American criminality.) But unionized employees
can grieve mistreatment and thus form a game-theoretic coalition
against the employer, or get the employer to pay the union dues when
in high demand.
And, John Nash's result means that the union (or professional
association) and company can also "win" when the number of players is
3 or greater. During the heyday of American unions, companies found
that collective bargaining saved time and did not need the massive
"human resources" staffs we have today.
In fine, there's micromanaging, some of which is in the mind of the
beholder.
>
> /Al
Richard Heathfield
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFFCD1C...@eton.powernet.co.uk>...
> > "Edward G. Nilges" wrote:
> > >
> > > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
> >
> > <snip>
> > > >
> > > > tell me what language do you think I'm talking about here....
> > >
> > > Probably, C or a similar out-of-date language.
> >
> > And C would be ou- of-date compared to... what, exactly? What language
> > of comparable power, portability and flexibility is more up-to-date than
> > C?
>
> C++, of course: its power and flexibility are proved by the fact that
> bozos write C code using C++ compilers, and its portability is proved
> by the fact that its original implementation used the C preprocessor.
C++ was most recently standardised in 1998. C was most recently
standardised in 1999. C is thus more up to date than C++, by the only
objective measure I can think of. Words such as "power", "flexibility"
etc are of little value when trying to decide out-of-dateness; I can
argue, and have argued successfully in the past, that C is more powerful
than C++, and it's trivial to argue that C++ is more powerful than C.
Conclusion: the word "power" can be interpreted in several different
ways.
BTW merely using the C preprocessor is insufficient to make a language
portable.
>
> You are pitching slow balls indeed (hopefully the metaphor is portable
> to countries playing cricket as well as baseball.)
More of an off-spinner really. (Does that translate to "curve ball"? I
don't know.)
Richard Heathfield
>
<snip>
> >
> > in a given situation
>
> Richard Heathfield is true to type: for he infers in another post that
> because at this point I use VB I am an *untermensch*.
No, *you* inferred it. Please consult a dictionary.
<snip>
verulam
> bozos write C code using C++ compilers,
I'm not sure how peoples ignorence proves power...
> and its portability is proved
> by the fact that its original implementation used the C preprocessor.
oh I though it's portablity was due to the fact that it's a widely
accepted standards based language.
> You are pitching slow balls indeed (hopefully the metaphor is portable
> to countries playing cricket as well as baseball.)
just about, we get the idea
verulam
> > pipe down, I don't want to hear it (most of you don't even use OO any
> > way so why not use C :->))</troll>
>
> Good point.
> >
> > There are even situations where Assembly language is the best choice
> > in a given situation
>
> Richard Heathfield is true to type: for he infers in another post that
> because at this point I use VB I am an *untermensch*.
no I think it was I that me, he mearly remarked that it 'explained a
lot'
>
> Common to this Orientalist confusion (where by "Orientalism" I mean
> the racial prejudice that codes external cues and uses them in
> preference to understanding) and Richard's sage advice that I avoid
> big words is a refusal, of course, to engage conceptual reality.
I can't help agree with him though, nothing taints an opinion more
than
when you get the feeling that they are ravaging their thesaurus mearly
to
appear clever.
I'm sorry but it's not racial prejudice, race has never come into it,
where did you get that from?
Might be simple prejudice though, I can however honestly say that the
only really really good VB programmers I have met have used other
languages
well before VB.
> This is the reason why the phenomenon you name above is so common.
> The head-hunters and the managers look at labels and no further.
> Therefore they refuse to hire the skilled OO VB.Net coder and hire
> instead some bozo who writes C in C++. Because in America we think it
> somehow impolite to claim knowledge above and beyond that which is
> assessed by the crudest, most demotic measures, the C++ clown is given
> a free pass.
Well you people are very silly. We have interviews and second
interviews
and technical tests. And I think that when a project has been spec'ed
to use
VB.Net and a manager hires a C++ programmer he'd be raddished.
> Unfortunately, this was called "getting your ticket punched" in
> Vietnam and we know what that led to.
you know you only have to mention Hitler now to get your troll
certificate.
verulam
> > to countries playing cricket as well as baseball.)
>
> More of an off-spinner really. (Does that translate to "curve ball"? I
> don't know.)
No no that's swing as in 'Right arm medium pace in-swinger'
although if I remember correctly from school cricket, swing is
produced by polishing on side of a rough ball. Apparently you can't
swing a cricket ball with spin like you can swing a football.
Er Soccer ball to Ed
verulam
> > Husserl?
>
> No, but a reading of Husserl shows that English and American
> philosophy tends to want to destroy the link between a textual
> activity and its referent.
>
> For example, symbolic logic is unconcerned as to the referents of
> symbols
aha Predicate Logic were back onto programming nearly, a very good
example of not concernong yourself with too much detail to solve a
problem.
> Unfortunately, this tends to blind programmers to the "forensic"
> nature of their work, where such activities as copying a legal data
Your quite right the spec says calculate the angle between the normal
of the centre of said polygon and you do it, low and behold that
polygon
is used to guide a missile on it's 3d map.
Oh please I've worked on defense projects and you know exactly what
your
doing. You just have to choose if you agree with it or not.
> base are speech acts with a reference that APPEAR to be purely
> mechanical and purely syntactical. In the case of copying a data
> base, the making of the copy creates a potential confusion after the
> first change to the copy as to which text (db) is accurate, therefore
yup they replication and databases, although I would suggest a
database
newsgroup for that sort of question.
> Husserl emphasizes that representation is always representation of x.
> I think this is at least corrective of the fallacy that we can just
> code and take no responsibility for what the code does.
least corrective of the fallacy? I'm not quite sure if I agree with
you
here, I thank that last scentence is encoded in some way can I have a
clue?
> This is a pseudo-romanticism which is at times apposite but not, I
> fear, in my case. For here your objection is not to the letter count
> of my words.
Your quite right I sneaked two objections in there did you spot them
both.
> For if it were, your rather idiotic list of sites in the other posts
> would fail your test!
Hey that wasn't me, apologise now
> Your objection to "big" words is actually an objection to novelty and
> density of ideas.
No that's insane, I object to the substitution of one long word where
two
little words are just as effective.
> It is a lazy, pompous, and ultimately disingenuous objection, for its
> Romanticism sits ill with its use in corporate venues, where it is
> disciplinary and minatory.
Now I believe 88% of the usage panel agree that 'playfully insincere'
is an accepted usage of 'disingenuous' so can I take that to be the
meaning?
Yes I was lazy, there was little subtly in my poke of fun.
Now I know management jargon creeps in when you have managers involved
and
the odd bit or jargon or TLA saves breath when technical people are
talking
but they discipline you if you don't use it? wow
Al Dunbar
"Edward G. Nilges" <spino...@yahoo.com> wrote in message
news:f5dda427.02060...@posting.google.com...
news:<AmXL8.197142$GG6.15...@news3.calgary.shaw.ca>...> >
> > > > I PREDICT that Oppie would have done this and with equal certitude,
> > >
> > > Rejected. You've demonstrated no basis for this prediction.
> >
> > Actually, I disagree with you here. Given the bizarro world that Edward
has
> > supposed into existence, based on a 1948 internet, I find it quite
likely
> > that everything and everyone would behave completely illogically in it.
> >
> > /Al
>
> (Sigh) Al, you are focusing on exactly the wrong thing.
Not so much focussing as illustrating that the nature of some of the
scenarios you paint can be so bizarre that almost any resulting conclusion
could reasonably be reached, no matter how unlikely this would be in the
real world.
> I merely
> wanted to illustrate how, if a rather minor technical detail were
> altered as a counterfactual conditional, the Los Alamos scientists
> would have treated a toplevel discussion group.
You illustrated no such thing, merely speculated it.
> And, why would a 1948 Internet be "bizarro?"
I did not apply the word "bizarro" to a 1948 internet, but to the world that
you supposed into existence that simply happened to contain such a thing.
/Al
Al Dunbar
Perhaps "slider" or "sinker".
/Al
Al Dunbar
"Edward G. Nilges" <spino...@yahoo.com> wrote in message
news:f5dda427.02060...@posting.google.com...
Fortunately, one cannot be sued for what one infers (especially in
newsgroups), if one is careful :-)
> Common to this Orientalist confusion (where by "Orientalism" I mean
> the racial prejudice that codes external cues and uses them in
> preference to understanding) and Richard's sage advice that I avoid
> big words is a refusal, of course, to engage conceptual reality.
So what is the nature of *your* confusion, Edward, Occidentalist? If my
newsreader is correct, it was verulam, not Richard, that mentioned the size
of your words. Most of us with experience in previous Nilges threads realize
that complaining about *how* you say things is a waste of time and
off-topic, while at the same time getting in the way of disputing what we
disagree with in terms of *what* you say.
/Al
Richard Heathfield
>
<snip>
>
> although if I remember correctly from school cricket, swing is
> produced by polishing on side of a rough ball. Apparently you can't
> swing a cricket ball with spin like you can swing a football.
No, I don't think you can, *but* giving it spin as it leaves your hand
can make it *bounce* in a reasonably repeatable-to-you but fairly
unpredictable-to-batsman way. If done well, this can improve your
averages no end.
Anyway, the *point* is that Mr Nilges does seem to have made the mistake
of taking my question at face value, and answered it unthinkingly. In
the words of my schoolfriends of many years ago:
OWWWWWIIIZZZZZZZYYYYYYYY!!!!
(This translates approximately to "owzat!", which translates
approximately to "Did you see that, Mr Umpire? *I* thought he was out.
What is your opinion?")
Edward G. Nilges
> > Richard Heathfield is true to type: for he infers in another post that
> > because at this point I use VB I am an *untermensch*.
>
> Fortunately, one cannot be sued for what one infers (especially in
> newsgroups), if one is careful :-)
The only case where Richard would give cause of action would be where
he claimed that my professional capabilities were inadequate or
nonexistent, and snide remarks don't count. I have no plans to bring
Richard to law. If he were to make a bald statement, I should instead
have recourse first to Google. I did so in a similar situation last
winter in another group.
And, Richard is partly correct. Many Visual Basic programmers do not
know the academic basics of their profession and worse, don't care to
learn.
>
> > Common to this Orientalist confusion (where by "Orientalism" I mean
> > the racial prejudice that codes external cues and uses them in
> > preference to understanding) and Richard's sage advice that I avoid
> > big words is a refusal, of course, to engage conceptual reality.
>
> So what is the nature of *your* confusion, Edward, Occidentalist? If my
> newsreader is correct, it was verulam, not Richard, that mentioned the size
> of your words. Most of us with experience in previous Nilges threads realize
> that complaining about *how* you say things is a waste of time and
> off-topic, while at the same time getting in the way of disputing what we
> disagree with in terms of *what* you say.
>
> /Al
Yes, it was verulam, my error.
Al, the statistical frequency of these complaints about my "verbosity"
is an indicator of a problem...but it's only an indicator.
If a poster is verbose, there are at least two possibilities. His
words are in excess of the message or a smaller word count is
inadequate to the message.
In some cases I have used words in excess of a message. For example,
I wrote a verbose article last summer about how to "tag" controls on
Windows forms. Unfortunately, my technique relied on an inadequate
facility and as a result the overall value was too low. ANY words on
this subject would be excessive.
However, here, I believe that comp.programming needs to contain
discussions of the interface between programming and culture that is
accessible to real programmers and I believe that the discussion
belongs at the top level group precisely because this interface is
itself at the margins, of necessity. I point to posts which ask
random questions about C, appearing in comp.programming as examples of
true charter transgression.
The discussion is self-reflexive and "points to" issues external to
programming (as do issues in any computer program that is "about"
something else, viz., almost any applications program, such as one
that purports to show fine art), therefore introduction of in-depth
discussions of the external phenomena is not only called-for, it is
demanded, by intellectual honesty, as long as the discussion, like a
Romantic symphony, returns to the motif.
Edward G. Nilges
> > > There is no generic, good for all language <troll>(C++ programmers
> > > pipe down, I don't want to hear it (most of you don't even use OO any
> > > way so why not use C :->))</troll>
> >
> > Good point.
> > >
> > > There are even situations where Assembly language is the best choice
> > > in a given situation
> >
> > Richard Heathfield is true to type: for he infers in another post that
> > because at this point I use VB I am an *untermensch*.
>
> no I think it was I that me, he mearly remarked that it 'explained a
> lot'
Right.
>
> >
> > Common to this Orientalist confusion (where by "Orientalism" I mean
> > the racial prejudice that codes external cues and uses them in
> > preference to understanding) and Richard's sage advice that I avoid
> > big words is a refusal, of course, to engage conceptual reality.
>
> I can't help agree with him though, nothing taints an opinion more
> than
> when you get the feeling that they are ravaging their thesaurus mearly
> to
> appear clever.
>
Verulam, this inappropriately places public relations and sales over
content.
I don't use thesauri. Instead, I read goddamn books. Therein I
encounter words, or, as Hamlet told Polonius, "words, my lord: words,
words, words."
The encounter in a book as opposed to a dictionary or thesaurus
reveals more, for it discovers a new node in a network, like coming on
a useful Web site.
For example, to encounter "ethics" and morality in context is to
discover that while ethics comes from a Greek word meaning the ethos
or nature of a person or thing, whereas morality comes from a Roman
word meaning custom and law, one realizes that it's an ethical
question whether I should adhere to custom (and use simple words when
talking to the aggressively simple) or follow the inner content of
what I might want to say, which may have a complex ethos.
> I'm sorry but it's not racial prejudice, race has never come into it,
> where did you get that from?
I made a "structuralist" point. It is based on a second-order
generalization from observing modern-day anti-black and
anti-Palestinian racists in action, and listening to programmers talk
about languages.
Anti-black talk in America condemns the teaching of Ebonics or the
speech of American blacks and it makes the claim that ordinary "white"
speech is more expressive, especially about culture and technology.
Anti-Palestinian talk world-wide makes similar unwarranted
generalizations about the language and the culture of Arabs.
Talk in software that a given language is reflective of the taste of
the programmer is structurally similar in that it makes an inference
that may be statistically warranted (statistically, American blacks
cannot access the educational advantages of whites: statistically,
Arabs are poor because their countries are Third World) but is not
what it asserts itself to be: a necessary connection.
For example, a high-school buddy of mine wrote a chess program in
Cobol, and I've written an OO compiler in VB.Net.
To say prior to investigation that my buddy's chess program is
worthless is to make the "Orientalist" choice to infer from the
language alone. As it happened, he was working for a minicomputer
vendor in the 1970s who needed to show off the power of its flagship
compiler, which was Cobol because that's what the market wanted.
In a similar fashion, a security guard at my local Barnes and Noble
inferred from the color of Joffre Stewart, a Chicago author, that he
was a homeless person and ejected him. The problem was that Joffre
was the featured speaker at a signing of his book.
Note that I'm not claiming that harried head-hunters who reject a
skilled C++ applicant because the job order calls for C are racists.
Of course, if they are any good, they have taken classes on employment
discrimination, which does not prohibit discriminating in favor of C.
But I do feel that the business demand expressed by "no Irish need
apply" is structurally the same as "must know C" when applied to a C++
guy. In our society, we have always allowed businesses to use these
sorts of tests, and they have flaws INCLUDING the fact that they hurt
the business itself.
I hope this explains the relationship. I hope it is not too verbose,
for its prolixity is, I believe, an artifact of the absolute necessity
of disambiguating it from ANY claim that "discriminating against
programmers on the basis of language == racial discrimination."
>
> Might be simple prejudice though, I can however honestly say that the
> only really really good VB programmers I have met have used other
> languages
> well before VB.
That is probably correct. I was fortunate in that my first language
was machine language (not assembler) and my first encounter with Basic
was three years later, in the era of "My compute rloves me when I talk
Basic" and the People's Computer Company.
However, Brian Kernighan, the author of the C book, is using VB-6 to
teach a survey course in computer science to non-majors at Princeton
University. I think it would be possible to start people out with
Basic if you have his level of knowledge and integrity. I'd make the
kiddies write a simple simulator for a simple machine, and then make
them write code for their product.
>
> > This is the reason why the phenomenon you name above is so common.
> > The head-hunters and the managers look at labels and no further.
> > Therefore they refuse to hire the skilled OO VB.Net coder and hire
> > instead some bozo who writes C in C++. Because in America we think it
> > somehow impolite to claim knowledge above and beyond that which is
> > assessed by the crudest, most demotic measures, the C++ clown is given
> > a free pass.
>
> Well you people are very silly. We have interviews and second
> interviews
> and technical tests. And I think that when a project has been spec'ed
> to use
> VB.Net and a manager hires a C++ programmer he'd be raddished.
>
Why? My boss showed me an article by a CS prof (sorry, don't have the
link) in which the prof said that a GOOD C++ programmer can pick up
Java in hours.
Most good programmers learn new languages not by attending classes but
by programming fun and cool things in the new language. Just let them
alone on their first day.
What we need (and this was suggested to me by a member of the
Cincinatti Programmer's Guild recently) is a manufacturer and language
independent way of certifying programmer skill, for example, an
automated test that would allow the programmer to select C++, VB.Net,
or Java.
> > Unfortunately, this was called "getting your ticket punched" in
> > Vietnam and we know what that led to.
>
> you know you only have to mention Hitler now to get your troll
> certificate.
Hmph? Many veteran programmers are military veterans and many of them
observed the military BS which reappears on the job. The American
military cleaned up its act but the officer class went on to become
upper management. Some of these clowns still have a ticket-punching
mentality in which they leave at five while their employees work "in
country" until 2 AM. The common theme: complete failure of
responsibility and compassion.
Edward G. Nilges
> "Edward G. Nilges" wrote:
> >
> > Ian Woods <new...@wuggy.org> wrote in message news:<Xns9223F21B1298...@217.32.252.50>...> > > news:f5dda427.02060...@posting.google.com:
> > >
> <snip>
> > > >
> > > > The political problem of malloc is that it FAILS to discriminate and
> > > > mallocs mere bytes, which can then hold any kind of pernicious text.
> > > > I prefer to create objects for this reason.
> > >
> > > allocate storage for an object the size you requested. Importantly, it
> > > doesn't specify that the allocated space for your object "holds any kind of
> > > pernicious text" - merely that it's value is indeterminate. Here is the
> > > relavent section about mallocs behaviour from ISO/IEC 9899:1999...
> > >
> > > 7.20.3.3 The malloc function
> > >
> > > Synopsis
> > >
> > > 1 #include <stdlib.h>
> > > void *malloc(size_t size);
> >
> > I'm well aware of what malloc does. Actual experience shows that
> > malloc and free cause memory leaks,
>
> Nonsense! Bad programming causes memory leaks. The malloc function
> allocates memory, and the free function deallocates it. It is the
> programmer's responsibility to use them correctly. It's a poor workman
> who blames his tools.
It's also our responsibility, when able to select the language, to
choose the right tool, and if objects avoid malloc then we need to use
objects.
Of course, if you don't free an object or worse don't free contained
objects in the object (through a Finalize, Terminate or Dispose) then
a form of leakage reoccurs. The point is that malloc with sizeof
requires you to remember more details about the size and shape of the
UDT which are encapsulated in an object.
>
> > and that objects, which don't
> > allow you to malloc, don't cause as many problems.
>
> Objects cannot stop me from using malloc. Nor do they.
>
I'm sure they don't.
> >
> > malloc was an early 1990s solution to problems whose complexity
> > outruns its capabilities.
>
> The malloc function dates at least as far back as 1978 (K&R1). Are you
> always this careless with your "facts"?
>
It was in the early 1990s that actual applications systems ceased to
be "green screen" and started to use both visual interfaces and
databases in preference to flat text files, as a statistical
observation based on the literature.
The typical 1989 application was MS-DOS based and command line. C was
a good choice for such an application.
If it needed to build a complex data base, the choice forced on the
programmer was to build a structure of interdependent ordinary files.
In many cases, it was unpredictable how many elements would be needed
in reading an interdependent set of records from these files. For
example, an invoice might include 1..n products.
Where each product was a UDT in the code, it made sense to malloc with
sizeof rather than decide apriori and on behalf of the user that the
company shall never invoice more than (say) 255 line items.
The first viable Windows (3) was released *circa* 1990 and it used
simple screens to access data bases including the original Access and
various flavors of dBase.
Good workmen immediately realized that green screen and malloc with
sizeof were phasing out.
Are you always this ignorant of the history of software?
> >
> > It shouldn't be used in an applications program.
>
> Why not? It's fine if it's used *properly*.
This is Platonic idealization by means of which 90% of programmers
stand condemned. How convenient as an academic meal ticket this must
be: for thereby the ivory tower can rise above it all. My suspicion
is that there is anxiety to do so, the real world being so, well,
unpleasant.
>
> > It should only be
> > used in an OO runtime to support higher-level capabilities.
>
> Stuff and nonsense. Still, I suppose at least we're back on topic.
Actually it was in the early 1990s that I realized, in working for an
investment venture guy, that the problem with C was that EVERY project
that either had complex visuals, or complex data, needed to contain an
explicit phase in which I would get to develop an abstract tool for
both problems.
Other, less skilled, programmers, were willing to be silent about this
need in order to get work. On the job, they would either create an
invisible, undocumented toolset that, like homosexual praxis in Oscar
Wilde's time, "dared not to speak its name", or they would build
enormous, unmaintainable programs that, at random places in the text,
would do things completely unrelated to the problem.
Those more skilled programmers who realized the need for an explicit
project phase to either "build the graphical user interface" or "build
the database" were accused of wasting time in signal cases. They were
also confronted with an old chestnut of the programming business,
which is the claim that such support "already exists and was written
by far more skilled people than thou."
It is NOT off-topic to suggest that the genesis of this last claim was
the economist's "law" of comparative advantage. Indeed, an
understanding of the linkage between poor software tools and
globalization is the only solution to this miserable problem.
I left C and started with Visual Basic 3 PRECISELY because I did not
want to waste my client's time. And as it happens VB-3 was superior
in respect of creating GUIs than its contemporary C++. This is no
longer the case.
Edward G. Nilges
> In article <f5dda427.02060...@posting.google.com>,
> spino...@yahoo.com (Edward G. Nilges) writes:
> >
> > I'm well aware of what malloc does. Actual experience shows that
> > malloc and free cause memory leaks,
>
>
> Manual store management is prone to memory leaks, and if you're using
> C's malloc-and-free combination, then they will "cause" the leaks, as
> in, some store allocated by malloc won't be free'd.
> > and that objects, which don't allow you to malloc, don't cause as many
> > problems.
>
> definition) and the moral equivalent of malloc/free - for example, C++
> with new/delete.
>
> I think you mean that languages with automatic store allocation (Lisp,
> Simula, Pop11, Java), which also support "objects" well, are less
> prone to memory leaks. But it's not the "objects" that are the cause.
Objects simply help you to free far more accurately, but this
assistance is critical in the real world.
My claim is that complexity of GUIs and data bases is enough of a jump
to warrent rejecting malloc and free, except insofar as you build a
completely separate tool to do the graphics or file management, based
ultimately (as are objects) on malloc and on free.
For example, for an onboard system for the marine industry, I wrote a
comprehensive toolkit for drawing on the screen as a completely
separate project. For me, a "critical success factor" was that this
toolset could be used outside the system in another system. That's
the only way of assuring that the system would function well inside
the onboard system.
I'd rather not have to do this.
>
> > malloc was an early 1990s solution
>
> Early 1990s? Is this a typo? Because malloc predates the 1990s. C code
> in that style predates the 1980s - K&R I has a free-store-management
> example using alloc/free, starting on p173, with a publication date
> of 1978.
(Sigh.) I guess I am at fault for not using enough words. I realize
that C was developed by Dennis Ritchie in the early 1970s. I took my
first class on C in 1978. But in real-world programming, C wasn't
heavily used until the 1980s and started to give way to C++ in the
early 1990s, and to Java in mid-decade.
>
> > to problems whose complexity outruns its capabilities.
>
> That's not clear. One can over-reach oneself in any language; malloc/free
> are probably about right for C's objectives. It's hard to see how to do
> better *and* stay small *and* stay general.
>
My buddy uses C in embedded systems. I'd love to be able to use it
because VB.Net is so clunky, and I am learning C Sharp for this
reason. But ultimately, this is all typing (as in typing on a goddamn
keyboard.) Even if I move to C Sharp, the important issues (correct
and on time) remain the same. I prefer curly braces to End If,
myself, but I can type End If if you pay me.
> > It shouldn't be used in an applications program. It should only be
> > used in an OO runtime to support higher-level capabilities.
>
> Not all programming problems call for OO runtimes, and those that do
> don't all call for implementations using malloc/free. I think the term
> "applications program" is too fuzzy to be decisive here.
It is a myth that using OO entails a huge interpretive runtime...a
myth that is disproved by the fact that Bjarne Stroustroup used the C
PREPROCESSOR to build C++.
I have to say that I have never written a compiler for an OO language
(I've developed compilers for non-OO languages.) But I don't think
there would be any extra mojo in such a product.
If the runtime needs to create an object, all it has to do is malloc
the sizeof the object's "state" (its collection of variables that are
not local) in a heap structure, right? If the object is created and
destroyed many times then I suppose this could be a performance hit,
but it's the fault of the programmer who probably hasn't scoped the
object correctly.
Languages (like PL/I) and methodologies (like OO) acquire a
reputation, in which perception is reality, for code and runtime bloat
which is then deconstructed by newer compilers. In the early 1970s,
the talk around the campfire was that PL/I, as opposed to Cobol and
BAL, was a pig on IBM 370 mainframes. In 1973, IBM released new
compilers for PL/I which were actually quite efficient, but the damage
had been done by the perception.
Java similarly proved the viability of the concept of OO which
Microsoft is following up on with .Net. My belief is that using
non-OO tools for modern problems is inhumane because you have to
invent so much UDT garbage that you end up marrying the cleaning lady,
she being the only female you ever see :-).
Edward G. Nilges
> "Edward G. Nilges" wrote:
> >
> > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFFCD1C...@eton.powernet.co.uk>...
> > > "Edward G. Nilges" wrote:
> > > >
> > > > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
> > >
> > > <snip>
> > > > >
> > > > > tell me what language do you think I'm talking about here....
> > > >
> > > > Probably, C or a similar out-of-date language.
> > >
> > > And C would be ou- of-date compared to... what, exactly? What language
> > > of comparable power, portability and flexibility is more up-to-date than
> > > C?
> >
> > C++, of course: its power and flexibility are proved by the fact that
> > bozos write C code using C++ compilers, and its portability is proved
> > by the fact that its original implementation used the C preprocessor.
>
> C++ was most recently standardised in 1998. C was most recently
> standardised in 1999. C is thus more up to date than C++, by the only
> objective measure I can think of. Words such as "power", "flexibility"
> etc are of little value when trying to decide out-of-dateness; I can
> argue, and have argued successfully in the past, that C is more powerful
> than C++, and it's trivial to argue that C++ is more powerful than C.
> Conclusion: the word "power" can be interpreted in several different
> ways.
That's silly: a one year difference trumping the genuine difficulty of
constantly having to "reinvent the wheel" using UDTs. You may have
management potential, Mr. Heathfield.
Edward G. Nilges
> > unregulated. ANYONE can claim to be a programmer, and many lack
> > the most basic skills.
>
> absolutely, I have a degree in Computer Science but I wouldn't trust a
> great proportion of my fellow graduates that year with some
> programming.
>
> I'm looking at some code now that has been written by some VB idiot (I
> know it's easy to bash them so I shoudn't) no consistent coding style,
> global and local variables with the same name, Goto's littered all
> over the place, an insane approach to logic, huge arrays of predefined
> length, poorly named functions, little or no error handling and to top
> it off there's 6 and a half thousand lines of the rubbish (although
> you can take 1 thousand away as commented out without explanation)
That's pretty representative of the skill level of VB programmers,
unfortunately.
You may be looking at the product of a serial collective of coders,
for many business programs are written by several people. Often, one
person starts, is fired or has a nervous collapse, and another person
is directed to take over the code.
IF you understand the problem independent of the code, I suggest a
complete rewrite. I often did this early in my career with success.
If this is not possible, use VB's excellent editor to systematically
rename variables using "Hungarian" notation. Change all the arrays,
if possible, to Collections: this may be a performance hit but it will
save your sanity.
Add On Error Go To lbl to the beginning of each Sub and Function (and
Property procedure if applicable.) For each procedure, add Exit to
its bottom and under that define lbl as a simple error handler. In
place of announcing the error using MsgBox, consider writing it to a
Windows log or to a text file.
Best of luck.
Eric Swanson
Please remove alt.global-warming from your posts.
You have drifted away from the original subject.
--
Eric Swanson --- E-mail address: e_sw...@skybest.com :-)
--------------------------------------------------------------
Eric Swanson
Please remove your replies from alt.global-warming
Hey, I used to be a programmer, now I'm a carpenter, thanks to Macro Sloth.
Eric Swanson
Eric Swanson
Eric Swanson
Eric Swanson
Eric Swanson
Edward G. Nilges
> [Followups set to comp.programming only - I see no reason why they
> should be dragged through a topicality debate unless they so choose.
> settings) suggest that he disagrees. I don't know why.]
>
> "Edward G. Nilges" wrote:
> >
> > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFDA74B...@eton.powernet.co.uk>...
> > >
> > > "Edward G. Nilges" wrote:
> <snip>
> > > > This very psychology is one of the reason for the decline in prestige
> > > > of "mere" programming.
> > >
> > > Maybe it is, and maybe it isn't. I don't do "mere" programming, so I
> > > don't know. Psychology isn't really topical for comp.programming.
> >
> > Tell that to Gerald Weinberg.
>
> You tell him. I'm busy.
>
> > It was PRECISELY his lack of focus, in
> > his 1972 book The Psychology of Computer Programming, that was so
> > valuable and useful, and it is why so much "focused" writing on UML
> > and use-cases is less valuable.
>
> It doesn't take a 30-year-old book to tell me UML is less valuable. But
> at least the discussion of the utility of UML for program design is
> topical here. Do you want to talk about the utility of UML? If so, I'm
> all ears.
Actually, I need to know more about UML. What I have seen seems not
to be terribly useful to the extent that its overall rhetorical tone
seems to be "now that we have x we can kill all the programmers."
Such a tone is rather offputting.
I did attend a presentation by Rational at Visual Studio 2002 and its
pitch was something I've seen repeatedly: "this is the magic bullet
that will allow the CEO's deepest desires instant realization, using
dogsbodies to handle the details." I realize that Rational has a
great product line (some of which I've used) but as long as this is
the pitch, it denies scientific reality as described in Fred Brooks'
1985 paper, "No Magic Bullet."
In my own work, I am developing a sort of core methodology which
dictates that each object shall have a Name property, an object2XML
method, and an inspect method. The Name will allow the object to have
a friendly name, the object2XML method will convert each state
variable to XML, and the inspect method will test a series of asserts
about the state, and tag the object as unusable when the test is
passed.
In addition, many objects seem to cry out for stacklike push and pop
methods, and the Holy Grail would be for each object to support a
macro language in which procedures could be written using its
properties and its methods.
But I see nothing of this level of craft in UML. Instead what I see
are pretty Visio diagrams which cloud the mind because they are never
tested. Perhaps I just need to get religion.
I do know that in 1960, an Air Force general said "now that we have
Cobol, we can get rid of all of those beatnik programmers." I do know
that I was pressured by my boss on my first job in 1971 to (1) get a
haircut and (2) become a manager and not a programmer. And, at
countless computer conferences, I have seen countless demos (on IBM
3270 CRTs, MS-DOS PCs, and Windows machines) of some moron drawing
pictures which then become code.
I think that commencing with John von Neumann, the elite of the NATO
countries (and, as far as I know, of the Warsaw pact) was resistant to
the very idea of the creation of any new trade or guild because prior
to WWI its experience was that skilled artisans were troublesome
chaps. I think that today's elite retains this resistance, while at
the same time civilisation as a whole desparately needs men and women
who have the level of committment that artisanship demands.
>
> > >
> > > <off-topic stuff about the USA snipped>
> > >
> > > > For example, Excel makes it easy to distort trend-lines in a
> > > > completely irresponsible fashion.
> > >
> > > So does a pencil and paper. This is not a programming issue. The issues
> >
> > Nope. Basically, it is IMPOSSIBLE to use David Stockman's "magic
> > asterisk" which he used at Reagan's OMB to get Reagan's targets, using
> > pencil and paper.
>
> If David Stockman's magic asterisk is a programming technique,
> discussion of its intent, implementation, and usage are topical here. Is
> /that/ what you want to talk about?
David Stockman's magic asterisk was created by programmers at
Visicalc. It is epiphenomenal as regards a deeper problem and this is
that actual code inside artifacts and bureaucracies is not properly
reviewed by independent bodies.
It appears, for example, that corporate accounting systems are rarely
audited by auditors with the skills to read Cobol.
To me, the activity of end users (like Stockman) is actually a form of
programming and there is a fuzzy boundary shown by the availability of
macro facilities. Stockman in ordinary parlance "programmed" the
budget spreadsheets when his asterisk caused out of balance totals
(which indicated that Reagan's massive tax cut would underfund social
programs, that the Republicans did not have the Congressional power to
repeal) to go in balance, because the asterisk represented the missing
funds.
Had Stockman worked in 1973 and not 1983, he might have directed a
Cobol programmer at OMB to add a feature to allow him to do this, but
Visicalc "empowered" him to perpetrate a fraud which he admitted in a
later book, and which Mario Cuomo named and assailed in a speech in
San Francisco in 1984.
Stockman in effect "programmed" OMB's workstations to work magic on
behalf of the gipper and this is a subject matter for this thread.
>
> >
> > > here are the knowledge and honesty of the person generating the
> > > "information", and the gullibility of the person receiving the
> > > "information". None of these issues have much, if anything, to do with
> > > programming (or indeed global warming!).
> >
> > Untrue.
>
> I strongly disagree, but let's see what you have to say...
>
> > Consider monetary rounding alone. What was an individual
> > fraud in the 1960s (where programmers were charged with sending
> > fractional pennies to their bank accounts) becomes a CORPORATE fraud
> > whenever a "business rule" in a telecom company dictates (as actually
> > happens) that calls of less than 60 seconds shall be rounded up.
>
> This still looks like a matter of honesty (or at least its corporate
> shadow, "compliance"), not a matter of programming. The programming here
> is not difficult to do either way. The difficulty, if there is any, is
> in the mind of the person writing the specification. Is he honest, or
> not? This is not a technical issue, but a moral one.
I support "separation of concerns", to an extraordinary depth. For
example, I switched from C to VB, despite VB's limitations, because on
the job, C required me to separate concerns, and build toolkits not
specifically authorized by the client.
However, you divide technology from morality in a completely different
way. Morality, unlike a graphics toolkit, is not a scientific concern
and is instead immanent in all concerns.
The Nuremburg laws alone might direct a "grunt" to memorize the
Universal Code of Military Justice in basic training (as, I believe,
American GIs and Canadian Tommies have to do) and not claim that he's
only a "grunt."
That is, the claim "I am only a grunt and I do what I am told" is
actually belied by its provenance, in which the common soldier has to
learn Geneva rules for the treatment of prisoners.
I realize that I am talking about the ideal and that Hollywood purveys
Rambo's exciting violation of the protocols. This is part of the
problem in programming: programmers want to be Rambo instead of
learning a mere trade, in which the ethics is part of the basic 101.
>
> > The
> > ability to do so electronically REMOVES a common practice of
> > long-distance systems prior to complete automation, where the operator
> > was empowered to forgive charges if the caller rang her and made the
> > case that the call was interrupted, or the person he wanted to speak
> > to was not home.
>
> Same applies. This isn't a programming issue but a corporate customer
> service issue. If the phone company wants to make this feature available
> to its customers, then it is free so to do. That decision, however, is
> not a matter of programming. It's up to the phone company.
Richard, I am well acquainted with this bureaucratic talk. "It's not
my department." It is ultimately a denial of human freedom.
>
> >
> > What you completely fail to see is an aspect of software "empowerment"
> > that is enhanced corporate control in disguise. This is ABOUT
> > programming.
>
> Show us. Write a computer program that disguises enhanced corporate
> control as an aspect of software "empowerment", and post the source here
> in a relatively well-known programming language. Many of us will then
> cheerfully comment upon it for you.
If customerBalance < 20 Then
declineTransaction
End If
or how about
corporateAccountBalance += customerBal - Math.Round(customerBal,
2)
and (derived from a health care system described in the New York
Times)
If prescriptionQuantity < standardDose Then
billStandardDose
End If
Also, please see Stephen Skiena's THE ALGORITHM DESIGN MANUAL
(Springer-Verlag 1999): he describes a sentencing spreadsheet that is
used by judges. It increases the length of the sentence
arithmetically while in the "amount stolen" column it increases the
amount geometrically, thereby making law to the effect that "poor,
black defendants, knocking over convenience stores, shall be subject
to severe sentences vis a vis rich, white men."
>
> >
> > >
> > > > Countless annual reports have been
> > > > deliberately distorted by its engine, which gives no normative advice
> > > > as to the mathematics of different trend-lines.
> > >
> > > This is an issue for people who prepare annual reports, not an issue for
> > > programmers.
> > >
> > This committs to me a legal error. If a programmer writes the
> > accounting software from which the annual report is derived, you
> > cannot seriously maintain that IF he's paid off by criminals to
> > systematically provide false reports, he's off the hook.
>
> Of course I don't maintain that. But this is still an honesty issue, not
> a programming issue.
>
> > This
> > exception proves the rule:
>
> No, it doesn't prove anything.
>
> > we interpret the programmer in the
> > exceptional case to be at fault because we know that the ordinary
> > programmer in the normal case is a full participant in the speech-act
> > that an annual report comprises.
>
> This is beginning to look like nonsense again.
I'd suggest reading and not staring at texts from afar and with not a
little disdain, that is unwarranted by your actual position.
>
> > The production of a correct and honest annual report is a group effort
> > in which the CPAs who sign it have to depend on the honesty of the
> > accountants who specify the reporting system. These accountants, as
> > "users" have to depend on the honesty of the coders who write the
> > general ledger.
>
> Right. It's an honesty issue.
>
> > It is to me a fantasy, encouraged by the cruel downgrading of the
> > programming profession, that their is some magical break at the doors
> > of the data center, beyond which we can play Doom, write a line of
> > code, have a cup of coffee, and pretend that we are not responsible.
>
> Nobody is pretending, as far as I can see, that programmers are not
> responsible for the code they write. That's one reason why it's good to
> have a newsgroup where programmers can discuss technical issues.
>
Ultimately, the issue is nameless (only partly captured by the Greek
word, arete) and it enfolds both technology and morality.
Let's take a VERY humble example: driver training in high school.
In such classes, good teachers do NOT teach mere driving skills.
Instead, closely integrated with the TECHNOLOGY, they talk about
SAFETY and they show goddamn movies with the consequences of
neglecting same.
A driving instructor who said "men, this is the gas pedal, and under
it, the metal: they should meet at all times" would be fired.
In traditional shop classes, good practice is likewise integrated.
Note that it's only in computer classes that the teacher takes a
hands-off attitude, which has two results.
The first is that most such classes are very boring. There's no
movies of severed heads, for starters. Second, even though it might
be thought that in a high school or college computer class, the
students would want to hear how to put the pedal to the metal, my own
experience (DeVry, Princeton, and Roosevelt) is that students actually
are hungry for guidance, even as I was in 1971.
The second result is of course that the students are released without
any real skills.
It's like law school, which teach authoritarianism and abuse through
the "Socratic method" while neglecting how to write a brief or make an
argument in a court of law.
My ultimate suspicion is that much "computer training" is ideology in
drag, which teaches, unlike driver training or shop, a sort of
flaccid, middle-class attitude towards life in which the important
thing is not bending metal or writing a compiler, but consuming stuff,
kicking the weak, and sucking up to the boss.
Mine isn't, which makes me a great trainer.
> > > > With all due respect, I stand prepared to win Yet Another border war
> > > > with you should you choose to begin one, for the problem is precisely
> > > > narrowness of focus and indeed of spirit.
> > >
> > > You might think of it as a war, and you might think you're winning. But
> > > when you spout off-topic crap in this newsgroup, everyone loses.
> > > Narrowness of focus is *essential* to Usenet, and narrowness of spirit
> > > is, would you believe, off-topic here.
> >
> > Yes, now dat we gots de Internet we must become nasty little clerks.
>
> Not at all. Let us, instead, refrain from childish ad hominem attacks
> and become considerate Netizens who find appropriate newsgroups in which
> to post.
Edward G. Nilges
> > Using malloc/free as a general purpose memory management methodology
> > is A BAD THING TO DO, bad because the methodology facilitates memory
> > management errors.
>
> hmm
> #define new malloc
> #define free delete
> better?
>
> or VB
> Set adoRecord = Nothing
> (if any body cares to take the 'goes out of scope is deallocated'
> argument, get your memory profiler out and take a look at what happens
> to ActiveX Recordsets)
I am still waiting for Microsoft to clean up its ADO act, but this is
not an argument for using malloc.
>
> This is of course the same argument used to leave pointers out of
> programming languages, people have trouble with pointers so let's
> leave them out so people don't cut themselves on them. oh look no
> pointers right ok, a double linked list here we go erm ok let's use
> agregation, oh dear if you set up a link back to the parent it creates
> a circular reference, darn, erm ok let's use a variable that's the
> same size as a pointer would be and write the code to set and
> dereference it so we trick the object reference counter, haha it works
>
Why do you NEED a doubly linked list? It's a fun challenge (I've
implemented it several times) but if your language provides a
primitive such as the Collection in all flavors of VB, you are wasting
your employer's time if you implement a doubly-linked list.
You need to write at least one in school in order to understand what's
going on under the hood, but as of .Net I've stopped implementing
them. I did occasionally use them in VB 6 for performance reasons.
> "People point all the time, it saves wandering over and touching
> something"
>
> How did I get to pointers? what was my argument? I'm rambling again
I use pointers, for example when I pass the "AddressOf" a
parameterless method to a thread.
Actually, a pointer is just a mechanization of the abstraction that in
a stored program, ANYTHING can be treated as data, and a good language
allows this.
My current system (.Net) allows me nearly complete runtime access to
the externals and internals of any object. This provides what in
former eras was provided by pointers.
I disavow any need for a bald pointer to some silly bytes, for long
ago I lost the instinct to tie my code to its runtime so intimately
that any change to the runtime broke the code.
verulam
> > Common to this Orientalist confusion (where by "Orientalism" I mean
> > the racial prejudice that codes external cues and uses them in
> > preference to understanding) and Richard's sage advice that I avoid
> > big words is a refusal, of course, to engage conceptual reality.
>
> So what is the nature of *your* confusion, Edward, Occidentalist? If my
> newsreader is correct, it was verulam, not Richard, that mentioned the size
> of your words. Most of us with experience in previous Nilges threads realize
> that complaining about *how* you say things is a waste of time and
> off-topic, while at the same time getting in the way of disputing what we
> disagree with in terms of *what* you say.
Tis true, a flash of impatience on a Friday afternoon. I'm glad he
defined what he took Orientalism to mean though :->
Richard Heathfield
global warming as far as I can see.
"Edward G. Nilges" wrote:
>
> "Al Dunbar" <Luigi...@hotmail.com> wrote in message news:<RjqM8.204984> >
> > > Richard Heathfield is true to type: for he infers in another post that
> > > because at this point I use VB I am an *untermensch*.
> >
> > Fortunately, one cannot be sued for what one infers (especially in
> > newsgroups), if one is careful :-)
>
> The only case where Richard would give cause of action would be where
> he claimed that my professional capabilities were inadequate or
> nonexistent, and snide remarks don't count.
That would still not be successfully actionable if I could prove that
such comments were truthful.
> I have no plans to bring
> Richard to law. If he were to make a bald statement, I should instead
> have recourse first to Google.
Why Google? I'm curious. After all, I don't use it to read newsgroups,
nor to post to them.
> I did so in a similar situation last
> winter in another group.
>
> And, Richard is partly correct. Many Visual Basic programmers do not
> know the academic basics of their profession and worse, don't care to
> learn.
Sadly, this is also true of programmers in general. We might disagree
over what constitutes "academic basics", mind you.
<snip>
Richard Heathfield
alt.global-warming.]
"Edward G. Nilges" wrote:
>
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3D01A200...@eton.powernet.co.uk>...
> > "Edward G. Nilges" wrote:
> > >
> > > Richard Heathfield <bin...@eton.powernet.co.uk> wrote in message news:<3CFFCD1C...@eton.powernet.co.uk>...
> > > > "Edward G. Nilges" wrote:
> > > > >
> > > > > verula...@hotmail.com (verulam) wrote in message news:<a78137c0.02060...@posting.google.com>...
> > > >
> > > > <snip>
> > > > > >
> > > > > > tell me what language do you think I'm talking about here....
> > > > >
> > > > > Probably, C or a similar out-of-date language.
> > > >
> > > > And C would be ou- of-date compared to... what, exactly? What language
> > > > of comparable power, portability and flexibility is more up-to-date than
> > > > C?
> > >
> > > C++, of course: its power and flexibility are proved by the fact that
> > > bozos write C code using C++ compilers, and its portability is proved
> > > by the fact that its original implementation used the C preprocessor.
> >
> > C++ was most recently standardised in 1998. C was most recently
> > standardised in 1999. C is thus more up to date than C++, by the only
> > objective measure I can think of. Words such as "power", "flexibility"
> > etc are of little value when trying to decide out-of-dateness; I can
> > argue, and have argued successfully in the past, that C is more powerful
> > than C++, and it's trivial to argue that C++ is more powerful than C.
> > Conclusion: the word "power" can be interpreted in several different
> > ways.
>
> That's silly: a one year difference
So it's sensible for you to say that C is out-dated compared to C++, but
silly for me to use objective facts to demonstrate that it is not?
Interesting debating style you have there.
> trumping the genuine difficulty of
> constantly having to "reinvent the wheel" using UDTs. You may have
> management potential, Mr. Heathfield.
Most kind, but I'm already a company director, so management would be a
step down. But what is this "genuine difficulty" you speak of? I have
found code re-use in C to be just as easy as, if not easier than, re-use
in C++.
Richard Heathfield
>
<snip>
>
> not an argument for using malloc.
No, it isn't. But neither is it an argument against using malloc.
<snip>
>
> Why do you NEED a doubly linked list?
Interesting question. I need a doubly linked list when I have a bunch of
things which need to be kept in the right order, when I don't know in
advance how many there will be, and when I am more likely to add stuff
at the beginning and end than somewhere in the middle, but where
"middle" operations might happen (including the addition or removal of
whole chunks of data in one go) so they have to be accounted for. I've
used linked lists for everything from text editors to ACL hackers.
> It's a fun challenge (I've
> implemented it several times)
Not a terribly difficult challenge, of course, but I agree that
programming is fun.
> but if your language provides a
> primitive such as the Collection in all flavors of VB, you are wasting
> your employer's time if you implement a doubly-linked list.
Not if your employer won't touch VB with a bargepole.
>
> You need to write at least one in school in order to understand what's
> going on under the hood, but as of .Net I've stopped implementing
> them.
Why?
> I did occasionally use them in VB 6 for performance reasons.
If performance was your criterion, you were using the wrong language.
>
> > "People point all the time, it saves wandering over and touching
> > something"
> >
> > How did I get to pointers? what was my argument? I'm rambling again
>
> I use pointers, for example when I pass the "AddressOf" a
> parameterless method to a thread.
Right.
> Actually, a pointer is just a mechanization of the abstraction that in
> a stored program, ANYTHING can be treated as data, and a good language
> allows this.
We could argue about this if you like. A pointer is:
* small and therefore cheap
* quick
* flexible
* reliable
It's not *just* a mechanisation of anything. It's a very handy
technique.
>
> My current system (.Net) allows me nearly complete runtime access to
> the externals and internals of any object. This provides what in
> former eras was provided by pointers.
Well, there you go. Take pointers out, and you end up with an
environment in which the vendor explicitly states in the licence
agreement that you are not allowed to publish benchmarks without their
explicit written permission.
> I disavow any need for a bald pointer to some silly bytes,
Clearly performance isn't an issue for you.
> for long
> ago I lost the instinct to tie my code to its runtime so intimately
> that any change to the runtime broke the code.
ROTFL - are you claiming your programs can survive the malicious
alteration of a carefully selected runtime byte?
Richard Heathfield
Agreed.
> and if objects avoid malloc then we need to use
> objects.
(a) You can't really avoid objects in C (or C++, come to that);
(b) You are begging the question. You say we should use objects because
they "avoid malloc", which is nonsense, but even if it were true that
they do, you are still assuming malloc is bad, which you haven't yet
proved.
> Of course, if you don't free an object or worse don't free contained
> objects in the object (through a Finalize, Terminate or Dispose) then
> a form of leakage reoccurs. The point is that malloc with sizeof
> requires you to remember more details about the size and shape of the
> UDT which are encapsulated in an object.
I think you're using the word "object" differently to me. Consequently,
we are in danger of misunderstanding each other. This is easily fixed,
though. If you think malloc is so bad, show me. Write me a C program
which illustrates what you think to be bad about malloc.
> > > and that objects, which don't
> > > allow you to malloc, don't cause as many problems.
> >
> > Objects cannot stop me from using malloc. Nor do they.
> >
> I'm sure they don't.
So why did you say "objects, which don't allow you to malloc"?
> > > malloc was an early 1990s solution to problems whose complexity
> > > outruns its capabilities.
> >
> > The malloc function dates at least as far back as 1978 (K&R1). Are you
> > always this careless with your "facts"?
> >
> It was in the early 1990s that actual applications systems ceased to
> be "green screen" and started to use both visual interfaces and
> databases in preference to flat text files, as a statistical
> observation based on the literature.
ROTFL, I'm afraid. Macs have been using visual interfaces since forever,
and mainframes have been using databases (ISAM, VSAM, SQL, etc) since
some time before that.
>
> The typical 1989 application was MS-DOS based and command line.
No, I don't think that's right; the typical 1989 application was written
in COBOL and ran on MVS or VM/CMS.
> C was
> a good choice for such an application.
Of course. C is a good choice for many classes of application.
> If it needed to build a complex data base, the choice forced on the
> programmer was to build a structure of interdependent ordinary files.
Not really. Even then, SQL was available quite easily on PCs.
> In many cases, it was unpredictable how many elements would be needed
> in reading an interdependent set of records from these files. For
> example, an invoice might include 1..n products.
>
> Where each product was a UDT in the code, it made sense to malloc with
> sizeof rather than decide apriori and on behalf of the user that the
> company shall never invoice more than (say) 255 line items.
It would have made even more sense to use SQL, but yes, in the absence
of a database engine, what you say makes sense. Hard limits make bad
code.
>
> The first viable Windows (3) was released *circa* 1990 and it used
> simple screens to access data bases including the original Access and
> various flavors of dBase.
<shrug>
>
> Good workmen immediately realized that green screen and malloc with
> sizeof were phasing out.
ROTFL!
(a) You've clearly never seen a mainframe. Green screen is here for a
long time yet.
(b) I can't quite see how you get from "malloc was useful in the past"
to "malloc is no longer useful", or even "malloc is *only* useful in
green screen apps", neither of which I accept.
>
> Are you always this ignorant of the history of software?
I think the ignorance is yours - you seem to think that "all the world's
a PC running Windows".
> > > It shouldn't be used in an applications program.
> >
> > Why not? It's fine if it's used *properly*.
>
> This is Platonic idealization by means of which 90% of programmers
> stand condemned. How convenient as an academic meal ticket this must
> be: for thereby the ivory tower can rise above it all. My suspicion
> is that there is anxiety to do so, the real world being so, well,
> unpleasant.
And /this/ is pseudoscientific balderdash which you spout whenever asked
for actual reasons for your opinions. I asked you why malloc should not
be used in an application program. Your answer, "Platonic idealization",
doesn't convince me. Can you come up with something more technically
applicable?
> > > It should only be
> > > used in an OO runtime to support higher-level capabilities.
> >
> > Stuff and nonsense. Still, I suppose at least we're back on topic.
>
> Actually it was in the early 1990s that I realized, in working for an
> investment venture guy, that the problem with C was that EVERY project
> that either had complex visuals, or complex data, needed to contain an
> explicit phase in which I would get to develop an abstract tool for
> both problems.
Please explain further. This actually sounds interesting.
<snip>
> Those more skilled programmers who realized the need for an explicit
> project phase to either "build the graphical user interface" or "build
> the database" were accused of wasting time in signal cases. They were
> also confronted with an old chestnut of the programming business,
> which is the claim that such support "already exists and was written
> by far more skilled people than thou."
This is certainly not uncommon, I agree. I have found that the answer
"you gots to show me" works wonders with project managers who think they
know what they're talking about.
<snip>
> I left C and started with Visual Basic 3 PRECISELY because I did not
> want to waste my client's time. And as it happens VB-3 was superior
> in respect of creating GUIs than its contemporary C++. This is no
> longer the case.
Depends on the platform. If I'm in a hurry, I'd choose Borland C++
Builder for Windows, or GTK (which is C-based) for Linux. Some would
choose Qt for both, but (a) the licensing is still a bit off-putting and
(b) I'm not entirely sure that I understand how to use it.
VB3 was indeed very good at what it did - it made Windows programming
*easy* so that you could get on with the complexities of the application
- but of course it has been surpassed easily by more powerful
implementations such as Delphi and BCB.
Richard Heathfield
"Edward G. Nilges" wrote:
> Richard Heathfield <bin...@eton.powernet.co.uk> wrote:
> > "Edward G. Nilges" wrote:
> > > Richard Heathfield <bin...@eton.powernet.co.uk> wrote:
> > > > "Edward G. Nilges" wrote:
> >
> > > It was PRECISELY his lack of focus, in
> > > his 1972 book The Psychology of Computer Programming, that was so
> > > valuable and useful, and it is why so much "focused" writing on UML
> > > and use-cases is less valuable.
> >
> > It doesn't take a 30-year-old book to tell me UML is less valuable. But
> > at least the discussion of the utility of UML for program design is
> > topical here. Do you want to talk about the utility of UML? If so, I'm
> > all ears.
>
> Actually, I need to know more about UML. What I have seen seems not
> to be terribly useful to the extent that its overall rhetorical tone
> seems to be "now that we have x we can kill all the programmers."
> Such a tone is rather offputting.
It has been said for values of x from VB right back to assembly
language.
<snip>
> In addition, many objects seem to cry out for stacklike push and pop
> methods, and the Holy Grail would be for each object to support a
> macro language in which procedures could be written using its
> properties and its methods.
>
> But I see nothing of this level of craft in UML. Instead what I see
> are pretty Visio diagrams which cloud the mind because they are never
> tested. Perhaps I just need to get religion.
You are better off without the UML religion, IMHO.
<snip>
> > > > > For example, Excel makes it easy to distort trend-lines in a
> > > > > completely irresponsible fashion.
> > > >
> > > > So does a pencil and paper. This is not a programming issue. The issues
> > >
> > > Nope. Basically, it is IMPOSSIBLE to use David Stockman's "magic
> > > asterisk" which he used at Reagan's OMB to get Reagan's targets, using
> > > pencil and paper.
> >
> > If David Stockman's magic asterisk is a programming technique,
> > discussion of its intent, implementation, and usage are topical here. Is
> > /that/ what you want to talk about?
>
> David Stockman's magic asterisk was created by programmers at
> Visicalc. It is epiphenomenal as regards a deeper problem and this is
> that actual code inside artifacts and bureaucracies is not properly
> reviewed by independent bodies.
So is it a programming technique or isn't it?
<snip>
> To me, the activity of end users (like Stockman) is actually a form of
> programming and there is a fuzzy boundary shown by the availability of
> macro facilities. Stockman in ordinary parlance "programmed" the
> budget spreadsheets when his asterisk caused out of balance totals
> (which indicated that Reagan's massive tax cut would underfund social
> programs, that the Republicans did not have the Congressional power to
> repeal) to go in balance, because the asterisk represented the missing
> funds.
Aha. It's not a programming technique, but an accounting trick.
Therefore, it's not topical here.
>
> Had Stockman worked in 1973 and not 1983, he might have directed a
> Cobol programmer at OMB to add a feature to allow him to do this, but
> Visicalc "empowered" him to perpetrate a fraud which he admitted in a
> later book, and which Mario Cuomo named and assailed in a speech in
> San Francisco in 1984.
He could have achieved the same thing with paper and pencil. This is not
a programming issue, but an honesty issue.
> Stockman in effect "programmed" OMB's workstations to work magic on
> behalf of the gipper and this is a subject matter for this thread.
Honesty issue, not a programming issue.
<snip>
> > > Consider monetary rounding alone. What was an individual
> > > fraud in the 1960s (where programmers were charged with sending
> > > fractional pennies to their bank accounts) becomes a CORPORATE fraud
> > > whenever a "business rule" in a telecom company dictates (as actually
> > > happens) that calls of less than 60 seconds shall be rounded up.
> >
> > This still looks like a matter of honesty (or at least its corporate
> > shadow, "compliance"), not a matter of programming. The programming here
> > is not difficult to do either way. The difficulty, if there is any, is
> > in the mind of the person writing the specification. Is he honest, or
> > not? This is not a technical issue, but a moral one.
>
> I support "separation of concerns", to an extraordinary depth. For
> example, I switched from C to VB, despite VB's limitations, because on
> the job, C required me to separate concerns, and build toolkits not
> specifically authorized by the client.
The way I see it, the client gets what he asks for. If he asks for an
impossibility, that's what he gets. I will do my utmost to explain to
the client what will happen if he tries to skimp on early work, and have
on occasion asked a client to sign and date a document containing such
an explanation, but at the end of the day, if he wants an unfinished
application, I can't really stop him from having one.
<OT stuff snipped>
> > > The
> > > ability to do so electronically REMOVES a common practice of
> > > long-distance systems prior to complete automation, where the operator
> > > was empowered to forgive charges if the caller rang her and made the
> > > case that the call was interrupted, or the person he wanted to speak
> > > to was not home.
> >
> > Same applies. This isn't a programming issue but a corporate customer
> > service issue. If the phone company wants to make this feature available
> > to its customers, then it is free so to do. That decision, however, is
> > not a matter of programming. It's up to the phone company.
>
> Richard, I am well acquainted with this bureaucratic talk. "It's not
> my department." It is ultimately a denial of human freedom.
No, it's not. You are perfectly free to discuss whatever you like *in an
appropriate group*. If such a group does not exist, form one. If you
can't form one because of insufficient interest, think about that lack
of interest.
> > > What you completely fail to see is an aspect of software "empowerment"
> > > that is enhanced corporate control in disguise. This is ABOUT
> > > programming.
> >
> > Show us. Write a computer program that disguises enhanced corporate
> > control as an aspect of software "empowerment", and post the source here
> > in a relatively well-known programming language. Many of us will then
> > cheerfully comment upon it for you.
>
> If customerBalance < 20 Then
> declineTransaction
> End If
That's not a program. It's a bug. Declining all transactions because of
a low balance would mean turning away deposits.
Now, there is no difference between a program containing such code and a
set of rules on a shelf containing an instruction to all staff. This is
not a software issue. This is not a programming issue. This is a
corporate policy issue.
> or how about
>
> corporateAccountBalance += customerBal - Math.Round(customerBal, 2)
Fraud is fraud, whether done by a computer or by pencil and paper.
>
> and (derived from a health care system described in the New York
> Times)
>
> If prescriptionQuantity < standardDose Then
> billStandardDose
> End If
This, again, is a policy issue, not a programming issue.
>
> Also, please see Stephen Skiena's THE ALGORITHM DESIGN MANUAL
> (Springer-Verlag 1999): he describes a sentencing spreadsheet that is
> used by judges. It increases the length of the sentence
> arithmetically while in the "amount stolen" column it increases the
> amount geometrically, thereby making law to the effect that "poor,
> black defendants, knocking over convenience stores, shall be subject
> to severe sentences vis a vis rich, white men."
Again, this is a policy issue, not a programming issue.
<snip>
> > > we interpret the programmer in the
> > > exceptional case to be at fault because we know that the ordinary
> > > programmer in the normal case is a full participant in the speech-act
> > > that an annual report comprises.
> >
> > This is beginning to look like nonsense again.
>
> I'd suggest reading and not staring at texts from afar and with not a
> little disdain, that is unwarranted by your actual position.
My actual position is that what you are talking about is not
programming, but societal concerns which are not topical here.
<snip>
Edward G. Nilges
> [global warming crosspost removed]> > >
> > > If David Stockman's magic asterisk is a programming technique,
> > > discussion of its intent, implementation, and usage are topical here. Is
> > > /that/ what you want to talk about?
> >
> > David Stockman's magic asterisk was created by programmers at
> > Visicalc. It is epiphenomenal as regards a deeper problem and this is
> > that actual code inside artifacts and bureaucracies is not properly
> > reviewed by independent bodies.
>
> So is it a programming technique or isn't it?
It is a programming technique. Merely because David Stockman did not
have the job title "OMB Programmer" does NOT mean he did not program
when he instructed Visicalc to force the crossfoot, if we can define
programming as behavior that controls or directs a computer.
Alan Greenspan, the Chairman of the Federal Reserve started with the
Fed in 1968 with the job title "data analyst" but in an NPR interview
Greenspan stated that he was a Cobol programmer.
>
> <snip>
>
> > To me, the activity of end users (like Stockman) is actually a form of
> > programming and there is a fuzzy boundary shown by the availability of
> > macro facilities. Stockman in ordinary parlance "programmed" the
> > budget spreadsheets when his asterisk caused out of balance totals
> > (which indicated that Reagan's massive tax cut would underfund social
> > programs, that the Republicans did not have the Congressional power to
> > repeal) to go in balance, because the asterisk represented the missing
> > funds.
>
> Aha. It's not a programming technique, but an accounting trick.
> Therefore, it's not topical here.
The rule: programmers write code and are not responsible for ANYTHING.
This is rather self-protective. That way the "user" can be blamed.
I am not saying that Visicalc programmers (some of whom I knew when
they worked at Bell-Northern Research in 1981 prior to being recruited
by Visicalc) were responsible for "the magic asterisk." Any
spreadsheet needs this feature.
I am saying that its use is not mere data entry. Instead, it is akin
to writing an Excel macro in that it controls the operation of the
tool. It is programming.
>
> >
> > Had Stockman worked in 1973 and not 1983, he might have directed a
> > Cobol programmer at OMB to add a feature to allow him to do this, but
> > Visicalc "empowered" him to perpetrate a fraud which he admitted in a
> > later book, and which Mario Cuomo named and assailed in a speech in
> > San Francisco in 1984.
>
> He could have achieved the same thing with paper and pencil. This is not
> a programming issue, but an honesty issue.
Richard, you've made this claim before and with all due respect it is
false.
It is simply impossible to have a box on a pencil and paper
spreadsheet that changes, invisibly, to the needed balancing total
when the other numbers change.
Your ideological blinders tell you "technology is value-neutral." The
problem is that Stockman and Reagan were part of a social class that
arose in the 1970s and that had an interest in manipulating a public
opinion that was to sophisticated to accept the New Deal promises of
1960s liberalism, which had been proved hollow by Vietnam and by
inflation.
Both the new class and its audience had a self-conception that
imagined itself as above being misled by JFK-style "rhetoric" and
large promises and large demands such as "ask not what your country
can do for you."
It replaced this "gullibility" with a numeracy that it learned during
the inflation of the 1970s in which microeconomic sophistication
became in vogue.
At the same time, digital technology was in the process of being
disseminated to first adopters in this class in the form of the Apple
II and 1981 IBM PC.
These early adopters happened to combine a cynicism about "just words"
and about poliiticians with a NAIVETE about the actual workings of
computers when they themselves did not program. They tended to be
rather overwhelmed by the rapidity in which Visicalc (and later Lotus
and Excel) recalculate to the extent that they failed to grasp the
underlying software, and, most important, the fact that this software
was not necessarily correct.
Reagan, as a member of this new class, had of course no understanding
of the magic asterisk, and, as a result, Stockman was able to meet his
OMB assigned targets and impress the gipper. This simply would not
have happened in the pencil and paper era, or even the mainframe era.
It would not have happened in the pencil and paper era (let us say,
the administration of Rutherford Hayes) for the simple reason that
anywhere along the paper trail, it is likely that at least one
clerical employee would have refused to defraud the President; indeed,
it was these sorts of refusals that brought down the bunkum artists
around Sam Grant.
Nor would it have happened in the mainframe era under Jerry Ford, for
in this era OMB was compelled to use outside specialists who weren't
in all cases on Stockman's "team": in a word, those infamous
"mainframe priests" who would have taken an effort by Stockman to
modify Cobol programs to the Washington Post.
The technology is not value-neutral, and ignorance of the above facts
is not adequate focus on the matter at hand. It is just ignorance.
>
> > Stockman in effect "programmed" OMB's workstations to work magic on
> > behalf of the gipper and this is a subject matter for this thread.
>
> Honesty issue, not a programming issue.
Tell that to hero computer scientist Edsger Dijkstra, whose writings
are consistently informed by an intellectual honesty that so shamed
American scientists like McCarthy that Dijkstra was nearly driven from
the field.
As I have said, you assume that there is a Department of Honesty,
alongside the Department of Programming. This is a profoundly
bureaucratic mindset, which enables the Department of Honesty to
become its orwellian reverse.
>
> <snip>
>
> > > > Consider monetary rounding alone. What was an individual
> > > > fraud in the 1960s (where programmers were charged with sending
> > > > fractional pennies to their bank accounts) becomes a CORPORATE fraud
> > > > whenever a "business rule" in a telecom company dictates (as actually
> > > > happens) that calls of less than 60 seconds shall be rounded up.
> > >
> > > This still looks like a matter of honesty (or at least its corporate
> > > shadow, "compliance"), not a matter of programming. The programming here
> > > is not difficult to do either way. The difficulty, if there is any, is
> > > in the mind of the person writing the specification. Is he honest, or
> > > not? This is not a technical issue, but a moral one.
> >
> > I support "separation of concerns", to an extraordinary depth. For
> > example, I switched from C to VB, despite VB's limitations, because on
> > the job, C required me to separate concerns, and build toolkits not
> > specifically authorized by the client.
>
> The way I see it, the client gets what he asks for. If he asks for an
> impossibility, that's what he gets. I will do my utmost to explain to
> the client what will happen if he tries to skimp on early work, and have
> on occasion asked a client to sign and date a document containing such
> an explanation, but at the end of the day, if he wants an unfinished
> application, I can't really stop him from having one.
Most programmers don't have this kind of relationship to the tools of
production. Instead, the actual tools of production (which are
networks and not individual computers) are kept out of their reach
save through either the employment contract, or the ersatz consulting
contract, which is really just day labor in drag.
As such, assuming that the programmer can ask enough questions about
the job content in the first or second interview, he is free to turn
down the work, and risk further unemployment. Of course, he'd be well
advised to have savings or unemployment.
However, companies don't hire programmers for specific projects but
instead feel free to reassign as needed.
This MEANS that in a company with poor and dishonest practice, the
programmer is always at risk for professional violation. As a result
most programmers cease regarding themselves as professional at all and
become indeed subhuman Morlocks.
>
> <OT stuff snipped>
>
> > > > The
> > > > ability to do so electronically REMOVES a common practice of
> > > > long-distance systems prior to complete automation, where the operator
> > > > was empowered to forgive charges if the caller rang her and made the
> > > > case that the call was interrupted, or the person he wanted to speak
> > > > to was not home.
> > >
> > > Same applies. This isn't a programming issue but a corporate customer
> > > service issue. If the phone company wants to make this feature available
> > > to its customers, then it is free so to do. That decision, however, is
> > > not a matter of programming. It's up to the phone company.
> >
> > Richard, I am well acquainted with this bureaucratic talk. "It's not
> > my department." It is ultimately a denial of human freedom.
>
> No, it's not. You are perfectly free to discuss whatever you like *in an
> appropriate group*. If such a group does not exist, form one. If you
> can't form one because of insufficient interest, think about that lack
> of interest.
Basically, this is *pro forma* for GFY.
>
>
> > > > What you completely fail to see is an aspect of software "empowerment"
> > > > that is enhanced corporate control in disguise. This is ABOUT
> > > > programming.
> > >
> > > Show us. Write a computer program that disguises enhanced corporate
> > > control as an aspect of software "empowerment", and post the source here
> > > in a relatively well-known programming language. Many of us will then
> > > cheerfully comment upon it for you.
> >
> > If customerBalance < 20 Then
> > declineTransaction
> > End If
>
> That's not a program. It's a bug. Declining all transactions because of
> a low balance would mean turning away deposits.
>
Oh you silly man. Banks happen to turn away this kind of business all
the time.
> Now, there is no difference between a program containing such code and a
> set of rules on a shelf containing an instruction to all staff. This is
> not a software issue. This is not a programming issue. This is a
> corporate policy issue.
No, it's a programming issue: for prior to programming, the bank's
personnel could, on an individual basis, service small accounts that
made no "economic sense" as defined by the (absentee) bank owners.
The programmability of the business rule MEANS that the lowest common
denominator processing applies IN PLACE of individual decisions by
tellers to service an old and valued customer.
And, the rigid definition of 'economic reality' expressed by business
rules that prohibit de minimis transactions does not even make
business sense as customers withdraw their funds from the uncaring
individual...or take their frustrations out by writing "putana" in
spray paint on the ATM, which happens in Argentina today: for in
Argentina, the business rules have been taken over by the IMF and they
dictate NO withdrawals.
I realize that in the 1930s, the bank could post police and National
Guard troops to prevent withdrawals, and, under FDR, a bank holiday
was enforced without computers. However, troops and National Guard
are people who could in Argentina and here be engaged in dialog and
persuaded, for example, to allow poor widows to make small
withdrawals. Whereas the business rules in force in the Argentine
situation simply don't care. This is a programming issue because it
shows the moral limits of programming and the programmer's mind-set,
and just because you say it isn't, does mean it isn't.
>
> > or how about
> >
> > corporateAccountBalance += customerBal - Math.Round(customerBal, 2)
>
> Fraud is fraud, whether done by a computer or by pencil and paper.
I agree that pencil and paper systems are subject to fraud. The
difference is critical, and it is that the fraud requires conscious
participation, dialog and intent. Whereas at Enron, only a small
number of insiders had to be brought in. The programmers and network
administrators were HONEST, hardworking people who did what they were
told.
Any fraud that uses the honesty and work-ethic of a third party, such
as a programmer, is worse because its victims include the programmer.
He can't get another job because Enron is on his resume, and his
honesty has been compromised.
>
> >
> > and (derived from a health care system described in the New York
> > Times)
> >
> > If prescriptionQuantity < standardDose Then
> > billStandardDose
> > End If
>
> This, again, is a policy issue, not a programming issue.
"We were only following orders."
>
> >
> > Also, please see Stephen Skiena's THE ALGORITHM DESIGN MANUAL
> > (Springer-Verlag 1999): he describes a sentencing spreadsheet that is
> > used by judges. It increases the length of the sentence
> > arithmetically while in the "amount stolen" column it increases the
> > amount geometrically, thereby making law to the effect that "poor,
> > black defendants, knocking over convenience stores, shall be subject
> > to severe sentences vis a vis rich, white men."
>
> Again, this is a policy issue, not a programming issue.
"We were only following orders."
>
> <snip>
>
> > > > we interpret the programmer in the
> > > > exceptional case to be at fault because we know that the ordinary
> > > > programmer in the normal case is a full participant in the speech-act
> > > > that an annual report comprises.
> > >
> > > This is beginning to look like nonsense again.
> >
> > I'd suggest reading and not staring at texts from afar and with not a
> > little disdain, that is unwarranted by your actual position.
>
> My actual position is that what you are talking about is not
> programming, but societal concerns which are not topical here.
"I needed a job in 1933 and the security services contacted me. I did
not agree wholeheartedly with the Party's aims but my family's needs
came first. I regarded myself as an expert technician in the use of
Xyklon-B and its use was not my department."
>
> <snip>
Programmers, it seems, would dearly love to be regarded as pure
scientists, above the sordid realities of business. The problem is
that almost no-one outside the business agrees with this self-image.
The elite does not share their awe of technology's potential. Bush's
father was shown to be completely ignorant of technology when he
naively praised bar-code technology that was old hat in 1992. Henry
Kissinger deigned to notice Xerography only insofar as that Kissinger
said (and I am NOT making this up) that it looked useful for
falsifying the record.
Ever since I started out in 1971, programmers and programming have
been regarded as less a value-neutral science and far more a nuisance
to be strictly subordinated to the firm's interests, even when those
interests happen to be criminal in nature.
This is on-topic, indeed it is the most important issue, I think,
comp.programming can discuss. Is the programmer just an exceptionally
subordinate member of management, in which case he should do EXACTLY
as he is told, or what?