Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Re: But... what's *wrong* then?

352 views
Skip to first unread message
Message has been deleted

LP

unread,
Dec 13, 2012, 9:18:33 AM12/13/12
to
On 2012-12-12, E.P.Sporgersi <E.P.Sporgersi> wrote:
> Once in a while, you come across a piece of scripting that is so
> horrible, so atrocious, that the question "What's wrong then" becomes
> utterly meaningless.

What is this "once in a while" of which you speak? This is a daily
event for me.

Todays example includes a loop which spins round an array called
@hosts copying each element into a string called $host, then creating
an object called $Host. $host is then concatenated with the result of
$Host->host() into a string called $Host (without destroying the
object first, which is a nice touch) with a : as a separator.

Finally, $Host is pushed onto the end of an array called @Hosts.

Which is hardly confusing at all.

-Paul
--
http://paulseward.com

Alan J Rosenthal

unread,
Dec 14, 2012, 8:49:40 PM12/14/12
to
LP <use...@lpbk.net> writes:
>Todays example includes a loop which spins round an array called
>@hosts copying each element into a string called $host, then creating
>an object called $Host. ...
>Finally, $Host is pushed onto the end of an array called @Hosts.

When I was young and foolish (these days I'm old and foolish), I was writing
a business/sales program in compiled MS-BASIC and things had to be in
different colours, and some of this was configurable so the colour numbers
needed to be stored in an array. And "color" was a reserved word. But
"colour" wasn't. So I stored the colour numbers in an array named "colour",
and there would be statements like "COLOR COLOUR(3)" (cf "COLOR 5").

As Monty Python says, I got better.

Peter Corlett

unread,
Dec 14, 2012, 9:24:16 PM12/14/12
to
Alan J Rosenthal <fl...@dgp.toronto.edu> wrote:
[...]
> When I was young and foolish (these days I'm old and foolish), I was writing
> a business/sales program in compiled MS-BASIC and things had to be in
> different colours, and some of this was configurable so the colour numbers
> needed to be stored in an array. And "color" was a reserved word. But
> "colour" wasn't. So I stored the colour numbers in an array named "colour",
> and there would be statements like "COLOR COLOUR(3)" (cf "COLOR 5").

A story I heard that was in a similar vein was where an organisation had two
otherwise similar boxes on opposite sides of the Atlantic. So of course the
stateside one had a hostname of "aluminum" and the one in Britain was
"aluminium". This apparently made for some interesting support calls.

> As Monty Python says, I got better.

Although if you were a Python programmer---or at least anything like the ones
I've met---you would assert that the language is perfect and therefore programs
in it are by definition elegant and contain no bugs.

Wojciech Derechowski

unread,
Dec 15, 2012, 1:16:08 AM12/15/12
to
On Sat, 15 Dec 2012 02:24:16 +0000, Peter Corlett wrote:
>
> A story I heard that was in a similar vein was where an organisation had two
> otherwise similar boxes on opposite sides of the Atlantic. So of course the
> stateside one had a hostname of "aluminum" and the one in Britain was
> "aluminium". This apparently made for some interesting support calls.
>

Or, dare I say, some interesting ideas for the new Sherlock episode. The
singular affair of the aluminum crutch mentioned en passant in The Musgrave
Ritual was later described in The Day of the Jackal with Sherlock disguised
as commissaire Lebel.

There is probably some money in it if we can sell the treatment to the BBC
which can contribute, for instance, to the ASR Pipe Tabacco Fund.

WD
--
Who is Entscheidungs and what is his problem?

Message has been deleted
Message has been deleted

Hans Klager

unread,
Dec 15, 2012, 11:03:41 AM12/15/12
to
On Sat, 15 Dec 2012 14:59:34 +0000 (UTC), Satya
<sat...@satyaonline.cjb.net> wrote:
>
> For example, yesterday I wished to *nix-script a dump out of the Redmond
> database, to be later replayed by another *nix script into a Redmond database.
> For Reasons. Apparently, this is impossible and undesirable; I should use a
> GUI or something else instead.

I worked for a Wintard, who like most Wintards could hardly
manage his PC he said to me: "Your skills are obselete".

What he meant was you keep using this Unix stuff with
arcane utilities, often of course to rescue his fuckups.



--
It is through disobedience that progress has been made, through
disobedience and through rebellion. - Oscar Wilde

Peter Corlett

unread,
Dec 15, 2012, 12:54:53 PM12/15/12
to
<E.P.Sporgersi> wrote:
> On Sat, 15 Dec 2012 02:24:16 +0000, Peter Corlett wrote:
>> Although if you were a Python programmer---or at least anything like the
>> ones I've met---you would assert that the language is perfect and therefore
>> programs in it are by definition elegant and contain no bugs.
> Being forced to learn that at the moment, for more or less precisely that
> reason. I wish that some of the Pythonites would come to recognise that if
> they feel moved to write the words "But you shouldn't want to do that
> anyway", it means that they probably didn't understand the question.

I find that Python is much improved if you ignore its community. You are likely
to get a more sensible answer if you asked a veteran Perl hacker. Although the
answer is just as likely to recommend a *third* language more appropriate to
your problem domain than to suggest a particular CPAN module.

> Like the one on FgnpxBiresybj who, when asked how one might pass an object to
> a function by reference, started a long treatise on why references are evil
> and should be avoided at all costs.

That's two people who don't understand Python's calling mechanism. Did somebody
point out that it's call-by-value, but what one thinks as an object in a
variable is actually a reference to an anonymous object, and so it looks and
works like call-by-reference to the uninitiated? And if so, how large was the
resulting mess due to exploded brains?

I can definitely get behind some of the sentiment after having seen some rather
exciting action-at-a-distance bugs caused by a rookie programmer in Java
reusing a POD object passed to a library rather than allocating a new one such
that the "old" object kept changing in sympathy with the "new" one, and also
see the point-of-view of those weirdoes who advocate functional programming
because it makes a whole class of pointer aliasing problems go away. But since
the day job usually involves writing Perl, I prefer to write defensive code
that isn't utter drivel rather than blindly follow arbitrary rules as to which
language features I should use.

I think my favourite daft rule was one site that wanted me to iterate over
*huge* datasets, which were best modelled as (tied) hashes, but I was not
permitted to use each() because they were once bitten by a misunderstanding as
to how it worked. Yes, each() is fundamentally broken. No, that doesn't mean it
should never be used.

Shmuel Metz

unread,
Dec 15, 2012, 8:00:41 PM12/15/12
to
In <kagn0g$m2l$1...@mooli.org.uk>, on 12/15/2012
at 02:24 AM, ab...@mooli.org.uk (Peter Corlett) said:

>Although if you were a Python programmer---or at least anything like
>the ones I've met---you would assert that the language is perfect

It is my firm conviction that nobody unable to cite flaws in a
language actually understands it.

--
Shmuel (Seymour J.) Metz <http://patriot.net/~shmuel> ISO position
Reply to domain Patriot dot net user shmuel+bspfh to contact me.
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

Alan J Rosenthal

unread,
Dec 15, 2012, 9:35:37 PM12/15/12
to
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> writes:
>It is my firm conviction that nobody unable to cite flaws in a
>language actually understands it.

Ah, but that's only if the language does have flaws. I think the
ASR mottos preclude anyone here's thinking that any programming language
might be flawless. But apparently some people out there think some certain
programming languages are indeed flawless.

-- aj "scientists are baffled!!!" r

mrob...@att.net

unread,
Dec 16, 2012, 12:44:00 AM12/16/12
to
Peter Corlett <ab...@mooli.org.uk> wrote:
> <E.P.Sporgersi> wrote:
>> On Sat, 15 Dec 2012 02:24:16 +0000, Peter Corlett wrote:
>>
>>> Although if you were a Python programmer---or at least anything like
>>> the ones I've met---you would assert that the language is perfect
>>> and therefore programs in it are by definition elegant and contain
>>> no bugs.
>>
>> I wish that some of the Pythonites would come to recognise that if
>> they feel moved to write the words "But you shouldn't want to do that
>> anyway", it means that they probably didn't understand the question.
>
> I find that Python is much improved if you ignore its community.

Something I find interesting is that if you ask the original author of
Perl about his religious beliefs, he will cheerfully tell you what they
are. The Perl community as a whole, though, seems to be fairly low on
evangelism. If you ask the original author of Python about his
religious beliefs, he will cheerfully tell you that he doesn't have
any. The Python community as a whole, though... yeah.

Matt Roberds

Peter Corlett

unread,
Dec 16, 2012, 2:11:33 AM12/16/12
to
Alan J Rosenthal <fl...@dgp.toronto.edu> wrote:
> Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> writes:
>> It is my firm conviction that nobody unable to cite flaws in a language
>> actually understands it.
> Ah, but that's only if the language does have flaws.

Dear sir, may I respectfully invite you to go and fuck yourself. All Software
Sucks, and therefore All Languages Suck.

> I think the ASR mottos preclude anyone here's thinking that any programming
> language might be flawless. But apparently some people out there think some
> certain programming languages are indeed flawless.

I've been around the block in my pitifully-few years on this planet, and all of
the languages I've dealt with are all software, and therefore suck. There are
some that I have, through bitter experience, concluded that suck least in their
own particular problem domain. Elucidating on which those are is tantamount to
advocacy and therefore forbidden. So if you're a rookie programmer, fuck you,
do your own research.

I do however suggest y'all learn Perl. Deciding whether that recommendation is
useful career advice, or a feint from a bitter old fart is down to your
judgement. If you're worthy and competent, you'll make the correct decision.

--
M: +44 7867 556741 / T: +44 20 7381 5299
Message has been deleted

Joe Zeff

unread,
Dec 16, 2012, 3:14:03 AM12/16/12
to
On Sun, 16 Dec 2012 07:11:33 +0000, Peter Corlett wrote:

> I've been around the block in my pitifully-few years on this planet, and
> all of the languages I've dealt with are all software, and therefore
> suck. There are some that I have, through bitter experience, concluded
> that suck least in their own particular problem domain. Elucidating on
> which those are is tantamount to advocacy and therefore forbidden.

Not only would it be advocacy, it would also be UI, making it doubly
forbidden. And, I see that my sigmonster's been reading my mind again
about what I'm going to be posting.

--
Joe Zeff -- The Guy With The Sideburns:
http://www.zeff.us http://www.lasfs.info
All hardware sucks, all software sucks. But you have to have met
management to really know the meaning of 'suck'.

David Cameron Staples

unread,
Dec 16, 2012, 3:17:53 AM12/16/12
to
On Sun, 16 Dec 2012 08:07:56 +0000, Roger Burton West wrote:

> On 2012-12-16, Peter Corlett wrote:
>>I do however suggest y'all learn Perl. Deciding whether that
>>recommendation is useful career advice, or a feint from a bitter old
>>fart is down to your judgement. If you're worthy and competent, you'll
>>make the correct decision.
>
> I do not believe that it is possible to know too many programming
> languages.
>
> Still wondering what I should learn next. Clojure, Scala, Go?

Haskell for scripting, Cobol for games, InterCAL for everything else. Duh.

ppint. at pplay

unread,
Dec 15, 2012, 10:57:43 PM12/15/12
to
- hi; in article, <kagn0g$m2l$1...@mooli.org.uk>,
ab...@mooli.org.uk "Peter Corlett" observed:
> Alan J Rosenthal <fl...@dgp.toronto.edu> wrote:
>>When I was young and foolish (these days I'm old and foolish), I was writ-
>>ing a business/sales program in compiled MS-BASIC and things had to be in
>>different colours, and some of this was configurable so the colour numbers
>>needed to be stored in an array. And "color" was a reserved word. But
>>"colour" wasn't. So I stored the colour numbers in an array named "colour",
>>and there would be statements like "COLOR COLOUR(3)" (cf "COLOR 5").
>[]
>>As Monty Python says, I got better.
>
>Although if you were a Python programmer---or at least anything like the
>ones I've met---you would assert that the language is perfect and there-
>fore programs in it are by definition elegant and contain no bugs.

- it is at least theoretically possible to devise a language
in which no grammatically and syntactically correctly-written
statement can be logically incorrect unless it is meaningless;

- might it not be theoretically possible to devise a computer
language in which nothing syntactically correctly-written could
be non-functional unless trivially true, reducing to zero, or
providing self-contradictory output?

- in which case, programs written in it that were not productive
of self-contradictions, trivialities, or a string of zeroes would
necessarily work.

- love, a ppint. as trusts that deducing what such any programs
actually did could safely be left as exercises...

[drop the "v", and change the "f" to a "g", to email or cc.]
--
"but i believe the figure of one and one sixteenth
will be sufficiently accurate for poetry"
- charles babbage, writing to correct the second half of tennyson's line,
"every moment dies a man; every moment one is born"

Wojciech Derechowski

unread,
Dec 16, 2012, 9:41:53 AM12/16/12
to
On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>
> - it is at least theoretically possible to devise a language
> in which no grammatically and syntactically correctly-written
> statement can be logically incorrect unless it is meaningless;
>

Can you prove that (so I can undarstand what you mean)?

Alan J Rosenthal

unread,
Dec 16, 2012, 11:26:25 AM12/16/12
to
ab...@mooli.org.uk (Peter Corlett) writes:
>Alan J Rosenthal <fl...@dgp.toronto.edu> wrote:
>> Ah, but that's only if the language does have flaws.
>
>Dear sir, may I respectfully invite you to go and fuck yourself. All Software
>Sucks, and therefore All Languages Suck.

Um, I did go on to say just about exactly this in the next sentence.

-- aj "not that I see anything wrong with fucking oneself" r

Howard S Shubs

unread,
Dec 16, 2012, 3:02:27 PM12/16/12
to
In article <kak03h$dg5$1...@dont-email.me>,
Learn ACIDIC. It's the opposite of BASIC.

--
Don't bother with piddly crap like "gun control".
Life is 100% fatal. Ban it.

Ralph Wade Phillips

unread,
Dec 17, 2012, 12:54:25 AM12/17/12
to
On 12/16/2012 2:07 AM, Roger Burton West wrote:
> On 2012-12-16, Peter Corlett wrote:
>> I do however suggest y'all learn Perl. Deciding whether that recommendation is
>> useful career advice, or a feint from a bitter old fart is down to your
>> judgement. If you're worthy and competent, you'll make the correct decision.
>
> I do not believe that it is possible to know too many programming
> languages.
>
> Still wondering what I should learn next. Clojure, Scala, Go?
>

Malbolge.

RwP

John Burnham

unread,
Dec 17, 2012, 4:45:40 AM12/17/12
to
On Sun, 16 Dec 2012 08:07:56 +0000, Roger Burton West wrote:

>
> I do not believe that it is possible to know too many programming
> languages.
>
> Still wondering what I should learn next. Clojure, Scala, Go?

Epigram.

J

Zebee Johnstone

unread,
Dec 17, 2012, 4:54:55 AM12/17/12
to
In alt.sysadmin.recovery on 17 Dec 2012 09:45:40 +0000 (GMT)
Fortran. After all, it's an off year for predicting its death.

Zebee
Message has been deleted

David Cameron Staples

unread,
Dec 17, 2012, 6:42:09 AM12/17/12
to
On Mon, 17 Dec 2012 11:10:49 +0100, Michel wrote:

> On Sat, 15 Dec 2012 14:59:34 +0000 (UTC), Satya wrote:
>> On 15 Dec 2012 10:29:08 GMT, E.P.Sporgersi wrote:
>>> Being forced to learn that at the moment, for more or less precisely
>>> that reason. I wish that some of the Pythonites would come to
>>> recognise that if they feel moved to write the words "But you
>>> shouldn't want to do that anyway", it means that they probably didn't
>>> understand the question.
>>
>> This isn't limited to Python.
>
> Oh ghods no. This attitude has "Apple" written all over it as well.

*cough*Gnome*cough*

Shmuel Metz

unread,
Dec 16, 2012, 9:04:09 AM12/16/12
to
In <20121216080427....@firedrake.org>, on 12/16/2012
at 08:07 AM, Roger Burton West
<roger+a...@nospam.firedrake.org> said:

>I do not believe that it is possible to know too many programming
>languages.

As long as they are not all fundamentally the same, thus reinforcing
the prejudices acquired from your first programming language. If it
were up to me schools would require programming classes featuring at
least three radically different languages, instead of offering courses
on specific languages.

>Still wondering what I should learn next. Clojure, Scala, Go?

Icon?

SETL?

RUNCIBLE?

Shmuel Metz

unread,
Dec 16, 2012, 9:00:40 AM12/16/12
to
In <kajs75$u7d$1...@mooli.org.uk>, on 12/16/2012
at 07:11 AM, ab...@mooli.org.uk (Peter Corlett) said:

>Dear sir, may I respectfully invite you to go and fuck yourself.
>All Software Sucks, and therefore All Languages Suck.

Whoosh! That is what he was alluding to.

>I do however suggest y'all learn Perl. Deciding whether that
>recommendation is useful career advice, or a feint from a bitter
>old fart

Why not both?

Shmuel Metz

unread,
Dec 16, 2012, 8:57:25 AM12/16/12
to
In <20121216.0047...@i-m-t.demon.co.uk>, on 12/16/2012
at 03:57 AM, v$af$pp...@i-m-t.demon.co.uk ("ppint. at pplay") said:

> - it is at least theoretically possible to devise a language
> in which no grammatically and syntactically correctly-written
> statement can be logically incorrect unless it is meaningless;

I want programs that produce correct results; "logically correct"
isn't good enough.

> - might it not be theoretically possible to devise a computer
> language in which nothing syntactically correctly-written could
> be non-functional unless trivially true, reducing to zero, or
> providing self-contradictory output?

FSVO functional. I not only want a program to be functional, I want it
to actually perform the intended function.

> - in which case, programs written in it that were not productive
> of self-contradictions, trivialities, or a string of zeroes would
> necessarily work.

FSVO work.

Shmuel Metz

unread,
Dec 16, 2012, 8:52:07 AM12/16/12
to
In <2012Dec15.2...@jarvis.cs.toronto.edu>, on 12/16/2012
at 02:35 AM, fl...@dgp.toronto.edu (Alan J Rosenthal) said:

>Ah, but that's only if the language does have flaws. I think the
>ASR mottos preclude anyone here's thinking that any programming
>language might be flawless.

Just as the FAQ precludes anybody here from gratuitous ebg13 and from
posting UI.

>But apparently some people out there think

C 'think' 'believe'

>some certain programming languages are indeed flawless.

Indeed, and I'd rather not be the one that has to clean up their code,
TYVM.

Shmuel Metz

unread,
Dec 16, 2012, 8:47:21 AM12/16/12
to
In <kaidhd$b1u$1...@mooli.org.uk>, on 12/15/2012
at 05:54 PM, ab...@mooli.org.uk (Peter Corlett) said:

>I think my favourite daft rule

How about "90% of the lines in your code must have comments."?

L R9,7 Put seven in register 9.

Now, an explanation as to *why* you are putting 9 in register 7, e.g.,
as flags, as a loop limit, might be of some utility, but simply
echoing what the reader should already understand is worse than
useless; it draws the readers attention away from the comments that
actually tell him something.

I used to warn students that I would subtract points for comments that
didn't improve clarity.

Niklas Karlsson

unread,
Dec 17, 2012, 8:50:45 AM12/17/12
to
On 2012-12-16, Shmuel Metz <spam...@library.lspace.org.invalid> wrote:
> I used to warn students that I would subtract points for comments that
> didn't improve clarity.

Well done. Mind you, "you are not expected to understand this" might be
worth it nonetheless, depending.

Niklas
--
Keeping UUCP running is starting to seem a lot like keeping a 130-year-old
man who smokes 4 packs a day on life support because he's the last person
on Earth who knows how to do the cha-cha, but he won't tell anyone.
-- Ryan Tucker

Peter Corlett

unread,
Dec 17, 2012, 10:09:17 AM12/17/12
to
Roger Burton West <roger+a...@nospam.firedrake.org> wrote:
[...]
> Still wondering what I should learn next. Clojure, Scala, Go?

978-1-93435-659-3

Wojciech Derechowski

unread,
Dec 17, 2012, 11:51:46 AM12/17/12
to
On Mon, 17 Dec 2012 15:09:17 +0000, Peter Corlett wrote:
> Roger Burton West <roger+a...@nospam.firedrake.org> wrote:
> [...]
>> Still wondering what I should learn next. Clojure, Scala, Go?
>
> 978-1-93435-659-3
>

Or the new Emergency Services number:
0118 999 881 999 119 725... 3

Joe Zeff

unread,
Dec 17, 2012, 3:21:25 PM12/17/12
to
On Mon, 17 Dec 2012 13:50:45 +0000, Niklas Karlsson wrote:

> On 2012-12-16, Shmuel Metz <spam...@library.lspace.org.invalid> wrote:
>> I used to warn students that I would subtract points for comments that
>> didn't improve clarity.
>
> Well done. Mind you, "you are not expected to understand this" might be
> worth it nonetheless, depending.

Many, many years ago, I did some contract ork at a small company.
Judging by the code's quality, most of the "programmers" that'd orked on
it before were still in high school. This is why, just before the code
doing a ovanel frnepu through an array, I put this comment:

If you don't understand the following code, don't change it.

--
Joe Zeff -- The Guy With The Sideburns:
http://www.zeff.us http://www.lasfs.info
Accounts of Jesus’s miracles are rendered less credible
due to all the witnesses being fishermen.

Steve VanDevender

unread,
Dec 17, 2012, 3:39:13 PM12/17/12
to
Wojciech Derechowski <wdd...@um5000.mystora.com> writes:

> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>
>> - it is at least theoretically possible to devise a language
>> in which no grammatically and syntactically correctly-written
>> statement can be logically incorrect unless it is meaningless;
>
> Can you prove that (so I can undarstand what you mean)?

I was disappointed that someone at Uncyclopedia deleted my page on
"non-functional programming". In summary, it said that adherents of
non-functional programming believe that programs ultimately should
neither return results or produce side effects. While most languages
can be used for non-functional programming, they are generally still
capable of doing things. Purely non-functional languages have attracted
great interest in academia because of their extremely straightforward
semantics.

--
For I know what you don't know / And I see things you'll never see /
And I've a different way of living, you know / And I've such a different
frame of mind, and so ... / I'm on my way to the funnyfarm
-- Happy Rhodes, "To the Funnyfarm"

John F. Eldredge

unread,
Dec 17, 2012, 8:04:42 PM12/17/12
to
Enochian. Just be careful not to do too much programming in your head,
lest you end up invoking literal brain-eaters.

--
John F. Eldredge -- jo...@jfeldredge.com
"Reserve your right to think, for even to think wrongly
is better than not to think at all." -- Hypatia of Alexandria

Alan J Rosenthal

unread,
Dec 17, 2012, 7:57:46 PM12/17/12
to
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> writes:
> L R9,7 Put seven in register 9.
...
>I used to warn students that I would subtract points for comments that
>didn't improve clarity.

Of course, the above is the original comment in its pristine state.
Greater fun and games to be had when over time that becomes

L R9,8 Put seven in register 9.

(Is the comment trying to tell us that this really puts seven in R9 for
some bizarre reason? No, it's just out of date. We think. Let's waste
some time investigating further.)

Alan J Rosenthal

unread,
Dec 17, 2012, 7:59:26 PM12/17/12
to
Joe Zeff <the.guy.with....@lasfs.info> writes:
>Many, many years ago, I did some contract ork at a small company.
>Judging by the code's quality, most of the "programmers" that'd orked on
>it before were still in high school. This is why, just before the code
>doing a ovanel frnepu through an array, I put this comment:
>
>If you don't understand the following code, don't change it.

Thus implying that in all other cases in which one doesn't understand
the code, one is encouraged to change it.

Peter H. Coffin

unread,
Dec 17, 2012, 10:29:00 PM12/17/12
to
On Sun, 16 Dec 2012 09:04:09 -0500, Shmuel Metz wrote:
> In <20121216080427....@firedrake.org>, on 12/16/2012
> at 08:07 AM, Roger Burton West
><roger+a...@nospam.firedrake.org> said:
>
>>I do not believe that it is possible to know too many programming
>>languages.
>
> As long as they are not all fundamentally the same, thus reinforcing
> the prejudices acquired from your first programming language. If it
> were up to me schools would require programming classes featuring at
> least three radically different languages, instead of offering courses
> on specific languages.

Heh. 20 years ago, that would be easy: C, COBOL, RPG. All pretty
different and all going to be facing HEAVY use in the Y2K run-up. Now,
too many things end up looking like someone smashed c++ and put it back
together in the dark. I suppose SQL through MERGE and stored procedures
could be the second, but the third? Forth and the assorted stack-based
languages have rather lost their shine.

--
31. All naive, busty tavern wenches in my realm will be replaced with
surly, world-weary waitresses who will provide no unexpected
reinforcement and/or romantic subplot for the hero or his sidekick.
--Peter Anspach's list of things to do as an Evil Overlord

er...@rail.eu.org

unread,
Dec 18, 2012, 1:33:04 AM12/18/12
to
"Peter H. Coffin" <hel...@ninehells.com> disait le 12/18/12 que :

> On Sun, 16 Dec 2012 09:04:09 -0500, Shmuel Metz wrote:
>> In <20121216080427....@firedrake.org>, on 12/16/2012
>> at 08:07 AM, Roger Burton West
>><roger+a...@nospam.firedrake.org> said:
>>
>>>I do not believe that it is possible to know too many programming
>>>languages.
>>
>> As long as they are not all fundamentally the same, thus reinforcing
>> the prejudices acquired from your first programming language. If it
>> were up to me schools would require programming classes featuring at
>> least three radically different languages, instead of offering courses
>> on specific languages.
>
> Heh. 20 years ago, that would be easy: C, COBOL, RPG. All pretty
> different and all going to be facing HEAVY use in the Y2K run-up. Now,
> too many things end up looking like someone smashed c++ and put it back
> together in the dark. I suppose SQL through MERGE and stored procedures
> could be the second, but the third? Forth and the assorted stack-based
> languages have rather lost their shine.

I would say C, one flavour of ML (SML or Caml) and Prolog.

Perverts can look at cross breeds such as lambda prolog.

--
Les simplifications c'est trop compliqué

Garrett Wollman

unread,
Dec 18, 2012, 2:31:11 AM12/18/12
to
In article <slrnkcvons....@nibelheim.ninehells.com>,
Peter H. Coffin <hel...@ninehells.com> wrote:
>Heh. 20 years ago, that would be easy: C, COBOL, RPG. All pretty
>different and all going to be facing HEAVY use in the Y2K run-up. Now,
>too many things end up looking like someone smashed c++ and put it back
>together in the dark. I suppose SQL through MERGE and stored procedures
>could be the second, but the third? Forth and the assorted stack-based
>languages have rather lost their shine.

These days, the contenders appear to be Python, Java, and
$FUNCTIONAL_LANG_OF_CHOICE (Scheme if you go to Northeastern,
otherwise something like ML, Haskell, or Clojure). Of course, the
place I work at[0] intends to teach programming, not any particular
programming language. (I know that Sussman's classes still use
Scheme, although since 6.001 ended there's no subject where you would
actually learn-by-doing. Architecture people tend to use Haskell or
some ML-family thing, and/or a specification language of some sort;
software engineering people are completely Java; AI people use either
Lisp or MATLAB depending on whether they're old-school symbolic AI or
new-school statistical AI.[1] Robotics people[2] tend to be very
Python-centric, and networking and systems people write in whatever
works, whether traditional C, Perl, or Java. You can guess where my
sympathies lie.)

-GAWollman

[0] ObDisclaimer: Of course I don't speak for them. The person who
does makes at least twice what I do.

[1] We don't have any of the old-school logic-programming AI types who
used Prolog -- any of those types as survived here would have
implemented the semantics of Prolog as a Lisp package.

[2] Technically part of AI here. At least until they announce the
next org chart.
--
Garrett A. Wollman | What intellectual phenomenon can be older, or more oft
wol...@bimajority.org| repeated, than the story of a large research program
Opinions not shared by| that impaled itself upon a false central assumption
my employers. | accepted by all practitioners? - S.J. Gould, 1993

Andrew

unread,
Dec 18, 2012, 11:36:46 AM12/18/12
to
On Mon, 17 Dec 2012 21:29:00 -0600, Peter H. Coffin wrote:

> On Sun, 16 Dec 2012 09:04:09 -0500, Shmuel Metz wrote:
>> In <20121216080427....@firedrake.org>, on 12/16/2012
>> at 08:07 AM, Roger Burton West
>><roger+a...@nospam.firedrake.org> said:
>>
>>>I do not believe that it is possible to know too many programming
>>>languages.
>>
>> As long as they are not all fundamentally the same, thus reinforcing
>> the prejudices acquired from your first programming language. If it
>> were up to me schools would require programming classes featuring at
>> least three radically different languages, instead of offering courses
>> on specific languages.
>
> Heh. 20 years ago, that would be easy: C, COBOL, RPG. All pretty
> different and all going to be facing HEAVY use in the Y2K run-up. Now,
> too many things end up looking like someone smashed c++ and put it back
> together in the dark. I suppose SQL through MERGE and stored procedures
> could be the second, but the third? Forth and the assorted stack-based
> languages have rather lost their shine.

I'd probably go with C for an understanding of bare-metal suffering, Python
for an understanding of modern bells-and-doohickeys suffering, and...maybe
some functional language (no experience there). Or perhaps shell scripting.
For an understanding of the importance of backup tapes.

--
Andrew

IT is a filter. It accepts masochists on stdin and emits misanthropes on
stdout.

Shmuel Metz

unread,
Dec 18, 2012, 2:43:32 PM12/18/12
to
In <slrnkcvons....@nibelheim.ninehells.com>, on 12/17/2012
at 09:29 PM, "Peter H. Coffin" <hel...@ninehells.com> said:

>Heh. 20 years ago, that would be easy: C, COBOL, RPG.

No, those are to close to each other. Contrast with, e.g., Prolog.

>All pretty different

FSVO different. Pick one of them and throw in, e.g., Icon, LISP,
Prolog, SETL for contrast. Maybe Chinese BASIC[1].

>Forth and the assorted stack-based
>languages have rather lost their shine.

A programming course should not be a popularity contest, it should
teach concepts.

[1] AKA The Abomination of Desolation[2], APL

[2] Yes, there are things that I like about it, but I understand
the reason for calling it that.

Juancho

unread,
Dec 18, 2012, 6:30:15 PM12/18/12
to
Peter Corlett wrote:
> I do however suggest y'all learn Perl. Deciding whether that recommendation is
> useful career advice, or a feint from a bitter old fart is down to your
> judgement.

Judgment is already done. You may guess, and you'll be right.

Shmuel Metz

unread,
Dec 18, 2012, 8:50:02 PM12/18/12
to
In <nct9w3uthq8$.tx2h702r9bk2$.d...@40tude.net>, on 12/18/2012
at 11:36 AM, Andrew <and...@invalid.invalid> said:

>I'd probably go with C for an understanding of bare-metal

C may have some relevance as an example of procedural language
suffering, but it has no relevance to bare metal suffering. I'd
probably include assemblers in an architecture class rather than in
the generic programming class, but that's what you need for bare
metal.

Howard S Shubs

unread,
Dec 18, 2012, 8:55:09 PM12/18/12
to
In article <kancit$7ig$1...@mooli.org.uk>,
ab...@mooli.org.uk (Peter Corlett) wrote:

> Roger Burton West <roger+a...@nospam.firedrake.org> wrote:
> [...]
> > Still wondering what I should learn next. Clojure, Scala, Go?
>
> 978-1-93435-659-3

Published on paper in 2010? It's OBsolete!

--
Don't bother with piddly crap like "gun control".
Life is 100% fatal. Ban it.

ppint. at pplay

unread,
Dec 18, 2012, 10:37:38 PM12/18/12
to
- hi; in article,
<50cdd345$15$fuzhry+tra$mr2...@news.patriot.net>,
spam...@library.lspace.org.invalid "Seymour J. Shmuel Metz" wrote:
> "ppint. at pplay" said:
>> - it is at least theoretically possible to devise a language
>> in which no grammatically and syntactically correctly-written
>> statement can be logically incorrect unless it is meaningless;
>
>I want programs that produce correct results; "logically correct" isn't
>good enough.

- not even as a first filter?

(it's not as though "logically correct" were incompatible with
"productive of correct result" - is it ?)

>> - might it not be theoretically possible to devise a computer
>> language in which nothing syntactically correctly-written could
>> be non-functional unless trivially true, reducing to zero, or
>> providing self-contradictory output?
>
>FSVO functional. I not only want a program to be functional, I want it
>to actually perform the intended function.

- now that's being downright picky.

- one could generate functional programs first, and then select
as desired from amongst those that perform the intended function;
those that do not may be discarded or, perhaps (& better?), stored
and indexed by the functions performed, for potential future use.
>
>> - in which case, programs written in it that were not productive
>> of self-contradictions, trivialities, or a string of zeroes would
>> necessarily work.
>
>FSVO work.

- itvo "reliably productive of results that are meaningful" ?

- love ppint.
[drop the "v", and change the "f" to a "g", to email or cc.]
--
"if you marry your mistress, you create a job vacancy"
- james goldsmith

David Cameron Staples

unread,
Dec 18, 2012, 11:21:36 PM12/18/12
to
On 17/12/12 1:41 AM, Wojciech Derechowski wrote:
> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>
>> - it is at least theoretically possible to devise a language
>> in which no grammatically and syntactically correctly-written
>> statement can be logically incorrect unless it is meaningless;
>>
>
> Can you prove that (so I can undarstand what you mean)?

There is a Professor Gödel on line four for you, and a Doctor Turing on
line seven.


--
David Cameron Staples | staples AT unimelb DOT edu DOT au
Melbourne University | ITS | Hosting | Unix Operations
all the technology in the world doesn't matter if the majority of your
population is mind numbingly stupid. -- bash.org/?453511

David DeLaney

unread,
Dec 19, 2012, 12:14:00 AM12/19/12
to
David Cameron Staples <cats...@gmail.com> wrote:
>On 17/12/12 1:41 AM, Wojciech Derechowski wrote:
>> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>> - it is at least theoretically possible to devise a language
>>> in which no grammatically and syntactically correctly-written
>>> statement can be logically incorrect unless it is meaningless;
>>
>> Can you prove that (so I can undarstand what you mean)?
>
>There is a Professor Gödel on line four for you, and a Doctor Turing on
>line seven.

But both of the calls are encrypted.

Dave, here, diagonally!
--
\/David DeLaney posting from d...@vic.com "It's not the pot that grows the flower
It's not the clock that slows the hour The definition's plain for anyone to see
Love is all it takes to make a family" - R&P. VISUALIZE HAPPYNET VRbeable<BLINK>
http://www.vic.com/~dbd/ - net.legends FAQ & Magic / I WUV you in all CAPS! --K.

Wojciech Derechowski

unread,
Dec 19, 2012, 12:49:02 AM12/19/12
to
On Wed, 19 Dec 2012 05:14:00 +0000, David DeLaney wrote:
> David Cameron Staples <cats...@gmail.com> wrote:
>>On 17/12/12 1:41 AM, Wojciech Derechowski wrote:
>>> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>>> - it is at least theoretically possible to devise a language
>>>> in which no grammatically and syntactically correctly-written
>>>> statement can be logically incorrect unless it is meaningless;
>>>
>>> Can you prove that (so I can undarstand what you mean)?
>>
>>There is a Professor Gödel on line four for you, and a Doctor Turing on
>>line seven.
>
> But both of the calls are encrypted.
>
> Dave, here, diagonally!

So? And don't forget a Doctor Emil Post and his Post-consistency which really
is the only reason why I'm asking; "logical correctness" doesn't bother me
at all.

Peter Corlett

unread,
Dec 19, 2012, 10:14:17 AM12/19/12
to
Howard S Shubs <how...@shubs.net> wrote:
> ab...@mooli.org.uk (Peter Corlett) wrote:
[...]
>> 978-1-93435-659-3
> Published on paper in 2010? It's OBsolete!

Until somebody invents an ebook reader that doesn't suck - ASS AHS - then paper
will never become completely obsolete.

I've tried a few such devices and they mainly fail in that the common units
come from people who also want to sell you content. So the software focusses on
a slick process to part the user from their hard-earned, with the ability to
read and manage a library very much added as an afterthought.

Howard S Shubs

unread,
Dec 19, 2012, 6:15:43 PM12/19/12
to
In article <kaslk9$gbl$1...@mooli.org.uk>,
ab...@mooli.org.uk (Peter Corlett) wrote:

> I've tried a few such devices and they mainly fail in that the common units
> come from people who also want to sell you content. So the software focusses
> on
> a slick process to part the user from their hard-earned, with the ability to
> read and manage a library very much added as an afterthought.

I like my Nook quite a bit. It fails, if it does, with books with a
complex layout, like a JavaScript book I'm reading now. That might be a
problem with how that particular book was converted, though.

Most of the books on my Nook came from sources other than B&N, note.

Alexander Schreiber

unread,
Dec 19, 2012, 4:40:10 PM12/19/12
to
I aquired an vevire Fgbel UQ about a year ago and I'm quite happy with it.
Yes, it is tied into an online bookstore, but in a rather loose and strictly
optional way - I tried to use that online bookstore and it didn't like my
unamerican credit card so I didn't use it again. I'm feeding the device with
epub files on SD card and there are plenty of places on that are happy to sell
one epub files that are _not_ crippled with DRM, not to mention sources of
free[0] epub files.

The high resolution display of the device is quite nice, battery life is
good and appears to be reasonably robust (usually lives in a leg pocket
of my cargo pants without further protection when not in use).

As for library management ... well, I just use the directory tree structure
on the SD card for the files I currently have on the device and do the real
library management on the desktop.

Kind regards,
Alex.
[0] legitimate sources[1], not "free" as in "fell off the back of a truck"
[1] Project Gutenberg & co, but I assume those to be well known
--
"Opportunity is missed by most people because it is dressed in overalls and
looks like work." -- Thomas A. Edison

Joe Zeff

unread,
Dec 20, 2012, 12:14:43 AM12/20/12
to
On Wed, 19 Dec 2012 22:40:10 +0100, Alexander Schreiber wrote:

> I aquired an vevire Fgbel UQ about a year ago and I'm quite happy with
> it. Yes, it is tied into an online bookstore, but in a rather loose and
> strictly optional way

As is my Abbx. It doesn't care where I download the content from as long
as it's in a supported format and it doesn't seem to be too fussy about
the format. Granted, I've not tried downloading and reading straight
ASCII text, but I wouldn't be surprised if it handled it fairly
gracefully. Being hardware, and using software, of course it sucks; but
not more than I'm willing to put up with.

--
Joe Zeff -- The Guy With The Sideburns:
http://www.zeff.us http://www.lasfs.info
A watched pot never boils.
An unwatched pot always boils over.

Seth

unread,
Jan 17, 2013, 4:01:53 PM1/17/13
to
In article <50cf7ec4$1$27766$862e...@ngroups.net>,
Joe Zeff <the.guy.with....@lasfs.info> wrote:
>On Mon, 17 Dec 2012 13:50:45 +0000, Niklas Karlsson wrote:
>> On 2012-12-16, Shmuel Metz <spam...@library.lspace.org.invalid> wrote:
>>> I used to warn students that I would subtract points for comments that
>>> didn't improve clarity.
>>
>> Well done. Mind you, "you are not expected to understand this" might be
>> worth it nonetheless, depending.
>
>Many, many years ago, I did some contract ork at a small company.
>Judging by the code's quality, most of the "programmers" that'd orked on
>it before were still in high school. This is why, just before the code
>doing a ovanel frnepu through an array, I put this comment:
>
>If you don't understand the following code, don't change it.

My personal favorite is still "think hard" (on a line of APL code that
has an obvious improvement which fails in one really extreme edge
case, when the interval being calculated runs backwards and ends on
February 28 in a leap year, or something like that).

Seth

Seth

unread,
Jan 17, 2013, 4:10:02 PM1/17/13
to
In article <50d0c764$40$fuzhry+tra$mr2...@news.patriot.net>,
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> wrote:
>In <slrnkcvons....@nibelheim.ninehells.com>, on 12/17/2012
> at 09:29 PM, "Peter H. Coffin" <hel...@ninehells.com> said:
>
>>Heh. 20 years ago, that would be easy: C, COBOL, RPG.
>
>No, those are to close to each other. Contrast with, e.g., Prolog.

Agreed.

>>All pretty different
>
>FSVO different. Pick one of them and throw in, e.g., Icon, LISP,
>Prolog, SETL for contrast. Maybe Chinese BASIC[1].

Or one of its successors. I personally like k, for all sorts of
reasons.

I'd insist that students learn FORTRAN, if for no other reason than to
see all of the ways it's an improvement over its successors.

(E.g. Good luck doing a Bobby Tables SQL-injection attack against a
language that specifies strings as 5HHello.)

>A programming course should not be a popularity contest, it should
>teach concepts.

And a Computer Science course should teach them in a "pure" form, as
untainted by any particular religion as feasible.

Seth
--
[1] NMF, refers to APL

Seth

unread,
Jan 17, 2013, 4:15:15 PM1/17/13
to
In article <slrnkcrndg...@um5000.mystora.com>,
Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>
>> - it is at least theoretically possible to devise a language
>> in which no grammatically and syntactically correctly-written
>> statement can be logically incorrect unless it is meaningless;
>
>Can you prove that (so I can undarstand what you mean)?

There's the trivial example, of course.

But then, "There does not now, nor will there ever, exist a
programming language in which it is the least bit difficult to write
bad programs."

And (as those of us who taught back in the day know) there are plenty
of people who can write FORTRAN in any language.

Seth

Wojciech Derechowski

unread,
Jan 17, 2013, 6:05:25 PM1/17/13
to
On Thu, 17 Jan 2013 21:15:15 +0000, Seth wrote:
> In article <slrnkcrndg...@um5000.mystora.com>,
> Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>>On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>>>
>>> - it is at least theoretically possible to devise a language
>>> in which no grammatically and syntactically correctly-written
>>> statement can be logically incorrect unless it is meaningless;
>>
>>Can you prove that (so I can undarstand what you mean)?
>
> There's the trivial example, of course.

Which is?
Message has been deleted

Peter Corlett

unread,
Jan 18, 2013, 5:13:10 AM1/18/13
to
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> wrote:
[...]
> A programming course should not be a popularity contest, it should
> teach concepts.

Then it should teach Perl. The concept is fear.

I'm starting to get to quite like Scala. It's a bit like Perl but with the
legacy sharp edges that people complain about taken away and new scary stuff
they can't quite understand added.

Maarten Wiltink

unread,
Jan 18, 2013, 5:47:21 PM1/18/13
to
"Wojciech Derechowski" <wdd...@um5000.mystora.com> wrote in message
news:slrnkfh0tl...@um5000.mystora.com...
The null language. D'oh.

Although it doesn't sound too hard to construct a language that
expresses equalities on sums in unary numbers, with a grammar that
only allows true statements.

PQ is correct.
If xPyQz is correct, then so are IxPyQIz and xPIyQIz.
IPIIQIII.
Almost straight from G�del, Escher, Bach.

Tebrgwrf,
Maarten Wiltink


Wojciech Derechowski

unread,
Jan 18, 2013, 9:29:34 PM1/18/13
to
On Fri, 18 Jan 2013 22:47:21 +0000, Maarten Wiltink wrote:
> "Wojciech Derechowski" <wdd...@um5000.mystora.com> wrote in message
> news:slrnkfh0tl...@um5000.mystora.com...
>> On Thu, 17 Jan 2013 21:15:15 +0000, Seth wrote:
>>> In article <slrnkcrndg...@um5000.mystora.com>,
>>> Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>>>> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>
>>>>> - it is at least theoretically possible to devise a language
>>>>> in which no grammatically and syntactically correctly-written
>>>>> statement can be logically incorrect unless it is meaningless;
> [...]
> Although it doesn't sound too hard to construct a language that
> expresses equalities on sums in unary numbers, with a grammar that
> only allows true statements.
>
> PQ is correct.
> If xPyQz is correct, then so are IxPyQIz and xPIyQIz.
> IPIIQIII.
> Almost straight from G�del, Escher, Bach.
>

Take x/II, y/I, z/ , where "/" is substitution and " ", zero.
Then IIPIQ is correct expression. If IPIIQIII is an axiom,
then, for instance, IIPIIQIIII is a theorem, and IIPIQ is not.

However, IIPIIQI is also a correct expression, since IIPIQ is.
Yet, only one of IIPIIQIIII, IIPIIQI can be true. Therefore
your grammar allows false statements, too.

I've had the book since 1989, but never read it, so I can only
guess that there is a restriction somwhere inthere on what can
be put in x, y, z by substitution.

Also the book probably deals with axiom schemas and possibly
even gets into a ploy of asserting all correct expressions as
axioms, but, again, I'm only guessing.

If what I say about IIPIIQIIII and IIPIIQI seems all right to
you I'll get to the question at hand: is IIPIIQI meaningless?

Steve VanDevender

unread,
Jan 19, 2013, 1:48:01 AM1/19/13
to
"Maarten Wiltink" <maa...@kittensandcats.net> writes:

> "Wojciech Derechowski" <wdd...@um5000.mystora.com> wrote in message
> news:slrnkfh0tl...@um5000.mystora.com...
>> On Thu, 17 Jan 2013 21:15:15 +0000, Seth wrote:
>>> In article <slrnkcrndg...@um5000.mystora.com>,
>>> Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>>>> On Sun, 16 Dec 2012 03:57:43 +0000, "ppint. at pplay" wrote:
>
>>>>> - it is at least theoretically possible to devise a language
>>>>> in which no grammatically and syntactically correctly-written
>>>>> statement can be logically incorrect unless it is meaningless;
>>>>
>>>> Can you prove that (so I can undarstand what you mean)?
>>>
>>> There's the trivial example, of course.
>>
>> Which is?
>
> The null language. D'oh.

"Although most languages can be used for nonfunctional programming, they
still have ways of doing things. Purely nonfunctional languages, which
have no facilities for doing anything, have attracted great interest in
academia for their extremely straightforward semantics."

--
Steve VanDevender "I ride the big iron" http://hexadecimal.uoregon.edu/
ste...@hexadecimal.uoregon.edu PGP keyprint 4AD7AF61F0B9DE87 522902969C0A7EE8
Little things break, circuitry burns / Time flies while my little world turns
Every day comes, every day goes / 100 years and nobody shows -- Happy Rhodes

Maarten Wiltink

unread,
Jan 19, 2013, 7:37:18 AM1/19/13
to
"Wojciech Derechowski" <wdd...@um5000.mystora.com> wrote in message
news:slrnkfk18e...@um5000.mystora.com...
> On Fri, 18 Jan 2013 22:47:21 +0000, Maarten Wiltink wrote:
[...]
>> PQ is correct.
>> If xPyQz is correct, then so are IxPyQIz and xPIyQIz.
>> IPIIQIII.

> Take x/II, y/I, z/ , where "/" is substitution and " ", zero.
> Then IIPIQ is correct expression.

Not by rule one, and not by rule two since both derivations
require at least one I after Q for reduction.

I'm sorry. I forgot to add "Any statement not assertably correct,
is not correct." Of change if to iff, which I almost did but in the
end considered unnecessary and distracting.


>
If IPIIQIII is an axiom,
> then, for instance, IIPIIQIIII is a theorem, and IIPIQ is not.

IPIIQIII was meant as a theorem, through construction from
axiom PQ by expanding through derivation rule 2a once and rule 2b
twice (in any order).


> However, IIPIIQI is also a correct expression, since IIPIQ is.
> Yet, only one of IIPIIQIIII, IIPIIQI can be true. Therefore
> your grammar allows false statements, too.

The unamended grammar simply remains silent on the validity of
IIPIIQI, which might be good enough.

The amended grammar (the _smallest_ language described by...)
can be shown to exclude it from the set of valid statements.


[...]
> If what I say about IIPIIQIIII and IIPIIQI seems all right to
> you I'll get to the question at hand: is IIPIIQI meaningless?

Do you mean 'correct'?. I've posited an axiom and two derivation
rules and introduced the name 'correct', I've not made any claims
about what system they describe or what xPyQz _means_. Even though
it's obvious.

So, going with obvious for the moment, 2+2=1 is not meaningless.
It _is_ demonstrably false (or rather IIPIIQI is not correct),
since it can't be reduced to the axiom.

In this language, IIIMIIQI and IITIIIQIIIIII is also incorrect.
That's the price you pay for using a toy language.

Tebrgwrf,
Maarten Wiltink


Wojciech Derechowski

unread,
Jan 19, 2013, 8:37:19 AM1/19/13
to
On Sat, 19 Jan 2013 12:37:18 +0000, Maarten Wiltink wrote:
> "Wojciech Derechowski" <wdd...@um5000.mystora.com> wrote in message
> news:slrnkfk18e...@um5000.mystora.com...
>> On Fri, 18 Jan 2013 22:47:21 +0000, Maarten Wiltink wrote:
> [...]
>>> PQ is correct.
>>> If xPyQz is correct, then so are IxPyQIz and xPIyQIz.
>>> IPIIQIII.
>
>> Take x/II, y/I, z/ , where "/" is substitution and " ", zero.
>> Then IIPIQ is correct expression.
>
> Not by rule one, and not by rule two since both derivations
> require at least one I after Q for reduction.
[...]
> IPIIQIII was meant as a theorem, through construction from
> axiom PQ by expanding through derivation rule 2a once and rule 2b
> twice (in any order).

If PQ is an axiom and not a rule, than what is "rule one"? (*)

> The unamended grammar simply remains silent on the validity of
> IIPIIQI, which might be good enough.
>
> The amended grammar (the _smallest_ language described by...)
> can be shown to exclude it from the set of valid statements.

Or is it "correct" statements? After all your axiom is written
thus: "PQ is correct". That's why it seems to be a rule, just
like the implication "If xPyQz is correct, then so are IxPyQIz
and xPIyQIz."

>> If what I say about IIPIIQIIII and IIPIIQI seems all right to
>> you I'll get to the question at hand: is IIPIIQI meaningless?
>
> Do you mean 'correct'?. I've posited an axiom and two derivation
> rules and introduced the name 'correct', I've not made any claims
> about what system they describe or what xPyQz _means_. Even though
> it's obvious.

No I don't mean correct. Again, (*) is a problem plus you do make
the claim:

"Although it doesn't sound too hard to construct a language that
expresses equalities on sums in unary numbers, with a grammar that
only allows true statements."

And please don't try to insult my intelligence speaking about the
claims you have not made. It annoys the pig, you know, because I
don't give a flying fuck for what the xPyQz means.

Shmuel Metz

unread,
Jan 19, 2013, 11:38:40 PM1/19/13
to
In <kdb77m$eev$1...@mooli.org.uk>, on 01/18/2013
at 10:13 AM, ab...@mooli.org.uk (Peter Corlett) said:

>Then it should teach Perl.


Only as one of many. The point is to teach the concepts of programming
languages, not to condition the students to use a particular language.

Wojciech Derechowski

unread,
Jan 20, 2013, 4:25:18 AM1/20/13
to
On Sun, 20 Jan 2013 04:38:40 +0000, Shmuel Metz wrote:
> In <kdb77m$eev$1...@mooli.org.uk>, on 01/18/2013
> at 10:13 AM, ab...@mooli.org.uk (Peter Corlett) said:
>
>>Then it should teach Perl.
>
>
> Only as one of many. The point is to teach the concepts of programming
> languages, not to condition the students to use a particular language.

Pidgin ALGOL is nearly enough to teach concepts of programming languages.
What's important in teaching, though, is to give people a language that
they don't have to master just to write anything useful.

I agree with Peter Corlett.

Shmuel Metz

unread,
Jan 20, 2013, 12:22:03 PM1/20/13
to
In <slrnkfndvt...@um5000.mystora.com>, on 01/20/2013
at 09:25 AM, Wojciech Derechowski <wdd...@um5000.mystora.com> said:

>Pidgin ALGOL is nearly enough to teach concepts of programming
>languages.

Not even close. It is enough to teach specific concepts for Algol-like
languages.

>What's important in teaching, though, is to give people a language
>that they don't have to master just to write anything useful.

No, what is important is to teach them concepts that they will be able
to apply in a general context. They can learn specific languages when
they need them, but they don't need to learn to write FORTRAN in any

Wojciech Derechowski

unread,
Jan 20, 2013, 1:29:19 PM1/20/13
to
On Sun, 20 Jan 2013 17:22:03 +0000, Shmuel Metz wrote:
> In <slrnkfndvt...@um5000.mystora.com>, on 01/20/2013
> at 09:25 AM, Wojciech Derechowski <wdd...@um5000.mystora.com> said:
>
>>Pidgin ALGOL is nearly enough to teach concepts of programming
>>languages.
>
> Not even close. It is enough to teach specific concepts for Algol-like
> languages.

I'm going to irritate the hell out of you:

@book{Aho:74,
AUTHOR = {A. V. Aho and J. E. Hopcroft and J. D. Ullman},
TITLE = {The Design and Analysis of Computer Algorithms},
YEAR = 1974,
PUBLISHER = {Addison-Wesley},
ADDRESS = {Reading, MA},
KEYWORDS = {}}

Brian Kantor

unread,
Jan 20, 2013, 5:28:57 PM1/20/13
to
Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>I'm going to irritate the hell out of you:
>
>@book{Aho:74,
> AUTHOR = {A. V. Aho and J. E. Hopcroft and J. D. Ullman},
> TITLE = {The Design and Analysis of Computer Algorithms},
> YEAR = 1974,


Oh ghod I feel old. I still have that book on my shelf from
the Analysis of Algorithms course I took back then. And we
did the programming for that course in Algol on a Burroughs 6500.

Ok, so I celebrated my 60th birthday a few days ago. But I
refuse to grow up.
- Brian

Wojciech Derechowski

unread,
Jan 20, 2013, 8:18:59 PM1/20/13
to
On Sun, 20 Jan 2013 22:28:57 +0000, Brian Kantor wrote:
> Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
>>I'm going to irritate the hell out of you:
>>
>>@book{Aho:74,
>> AUTHOR = {A. V. Aho and J. E. Hopcroft and J. D. Ullman},
>> TITLE = {The Design and Analysis of Computer Algorithms},
>> YEAR = 1974,
>
>
> Oh ghod I feel old. I still have that book on my shelf from
> the Analysis of Algorithms course I took back then. And we
> did the programming for that course in Algol on a Burroughs 6500.

Yeah, not exactly "concepts of programming languages", but what the
hell.

> Ok, so I celebrated my 60th birthday a few days ago. But I
> refuse to grow up.

So, don't do it. After all, it's up to you.
Message has been deleted
Message has been deleted

Peter Corlett

unread,
Jan 21, 2013, 6:03:13 AM1/21/13
to
Wojciech Derechowski <wdd...@um5000.mystora.com> wrote:
[...]
> @book{Aho:74,
> AUTHOR = {A. V. Aho and J. E. Hopcroft and J. D. Ullman},
> TITLE = {The Design and Analysis of Computer Algorithms},
> YEAR = 1974,
> PUBLISHER = {Addison-Wesley},
> ADDRESS = {Reading, MA},
> KEYWORDS = {}}

I have not read that specific book, but books on algorithms tend to assume
constant-time access to memory, i.e. that we're all still programming on a
Sinclair ZX81. A book from 1974 is unlikely to know much about the horrors of
x86 memory subsystems...

Shmuel Metz

unread,
Jan 21, 2013, 7:05:42 AM1/21/13
to
In <kdhr39$r7r$1...@karoshi.ucsd.edu>, on 01/20/2013
at 02:28 PM, br...@karoshi.ucsd.edu (Brian Kantor) said:

>Oh ghod I feel old. I still have that book on my shelf from the
>Analysis of Algorithms course I took back then.

Oh, come now, Shirl;ey I can't be the only one here who has a copy of
TAoCP. And, yes, it is written towards a specific language and, no, I
don't like that.

Wojciech Derechowski

unread,
Jan 21, 2013, 7:44:22 AM1/21/13
to
Yes, RAM complexities are measured by the uniform or a logarithmic cost
criterion:

\[
l(i)=
\begin{cases}
\lfloor\,\log\;\lvert\,i\,\rvert\,\rfloor + 1, & i \neq 0 \\[2pt]
1, & i = 0
\end{cases}
\]

where l(i) is logarithmic function on integers.

That, I suppose, would've made Sinclair programmer very unhappy.

Hans Klager

unread,
Jan 21, 2013, 11:24:18 AM1/21/13
to
On 20 Jan 2013 14:28:57 -0800, Brian Kantor <br...@karoshi.ucsd.edu> wrote:
>
> Ok, so I celebrated my 60th birthday a few days ago. But I
> refuse to grow up.

I didn't know you were that young.



--
“America is not at war. The Marine Corps is at war; America is at the mall.”
- Graffiti at a Marine Barracks, Ramadi, Iraq. April 2010

Peter Corlett

unread,
Jan 22, 2013, 5:33:33 AM1/22/13
to
Shmuel (Seymour J.) Metz <spam...@library.lspace.org.invalid> wrote:
> ab...@mooli.org.uk (Peter Corlett) said:
[...]
>> Then it should teach Perl.
> Only as one of many. The point is to teach the concepts of programming
> languages, not to condition the students to use a particular language.

That was very much implicit. I know that university CS departments are mostly
just diploma mills churning out Java developers, but that wouldn't be the case
in any CS course *I* were to design. Them buggers will be *proud* they got as
high as a third class degree.

What Perl offers is a hybrid language with both imperative and functional
style, with object-orientation not looking remotely like Java, and which can be
run as simple throwaway test or admin scripts with a shebang line. On all these
fronts, Scala also shines.

What Perl gives us in addition to Scala is some rather funky syntax, bizarre
sharp edges that only made sense in a long-forgotten historical context[1],
libraries built in an apparently ad hoc manner by weasels on crack[0], and an
installed base of some of the worst code you can imagine that will make
maintenance programmers plot their revenge on the original author. Only then
does the rookie gain true understanding of programming.


[0] I believe this is actually the case.

[1] Hands up anybody who has used Perl formats. No? "man perlform" and weep. I
particularly like the admonition to not put a lone dot on a line by itself,
presumably due to a long-fixed bug in Sendmail.

LP

unread,
Jan 22, 2013, 6:47:40 AM1/22/13
to
On 2013-01-22, Peter Corlett <ab...@mooli.org.uk> wrote:
>
> What Perl gives us in addition to Scala is some rather funky syntax, bizarre
> sharp edges that only made sense in a long-forgotten historical context[1],
> libraries built in an apparently ad hoc manner by weasels on crack[0], and an
> installed base of some of the worst code you can imagine that will make
> maintenance programmers plot their revenge on the original author. Only then
> does the rookie gain true understanding of programming.

I'm of the belief that you don't truly learn how to write maintainable code
until you've spent significant time trying to maintain unmaintainable code
in situations where a complete rewrite isn't cost effective.

Perl happens to provide that experience a little too readily.

> [0] I believe this is actually the case.

I like the ones which are merely written by weasels on crack.

It's those which are written by psychotic inbred weasels on crack with
attention spans insufficient to write any documentation beyond the object
constructor, which are then used by a second team of psychotic weasels
on some combination of meth and ketamine with an attitude problem and a
dislike of meaningful variable names.

I shouldn't have to read the code in conjunction with a packet capture of
what actually goes out on the wire when I run it, and the RFC describing the
thing the module is supposed to implement - just to find out how to unpack
the return data from a function.

crey::BZNCV - I'm looking at you.

> [1] Hands up anybody who has used Perl formats. No? "man perlform" and weep.

Oh, that's 'interesting'

I'll make a note to *never* look at that again.

-Paul
--
http://paulseward.com
Message has been deleted

Juergen Nickelsen

unread,
Jan 22, 2013, 8:46:21 AM1/22/13
to
Roger Burton West <roger+a...@nospam.firedrake.org> writes:

> I have said before that, if I ever design a programming course, the
> second-year project will involve being handed a random first-year
> project and being told to add features to it.

Excellent idea. Chapeau!

--
So, as the Podembitzer Rebbe told his congregation when he won the
lottery, "I will spend the rest of my life thanking HaShem -- and
chasing women in Las Vegas."

Shmuel Metz

unread,
Jan 22, 2013, 8:40:02 AM1/22/13
to
In <slrnkfsv2r...@lemon.resnet.bris.ac.uk>, on 01/22/2013
at 11:47 AM, LP <use...@lpbk.net> said:

>It's those which are written by psychotic inbred weasels on crack
>with attention spans insufficient to write any documentation beyond
>the object constructor, which are then used by a second team of
>psychotic weasels on some combination of meth and ketamine with an
>attitude problem and a dislike of meaningful variable names.

Is that another way of writing "industry norm:?

Shmuel Metz

unread,
Jan 22, 2013, 8:38:08 AM1/22/13
to
In <kdlptt$cs5$1...@mooli.org.uk>, on 01/22/2013
at 10:33 AM, ab...@mooli.org.uk (Peter Corlett) said:

>What Perl offers is a hybrid language with both imperative and
>functional style, with object-orientation not looking remotely like
>Java, and which can be run as simple throwaway test or admin scripts
>with a shebang line. On all these fronts, Scala also shines.

If you're only going to teach a single language, Perl might be a good
choice, but it's missing a lot that is in other languages, e.g.,
backtracking. I'd rather prepare the students for a world in which all
languages suck, with a general enough understanding that they can
quickly pick up a new language dissimilar to those they already know.

David DeLaney

unread,
Jan 22, 2013, 10:02:32 AM1/22/13
to
LP <use...@lpbk.net> wrote:
>I like the ones which are merely written by weasels on crack.
>
>It's those which are written by psychotic inbred weasels on crack with
>attention spans insufficient to write any documentation beyond the object
>constructor, which are then used by a second team of psychotic weasels
>on some combination of meth and ketamine with an attitude problem and a
>dislike of meaningful variable names.

...surely ferrets were involved at SOME point? A maze of twisty little
undocumented footprints, all dirty and smelly, and a trail of mental
destruction.

Dave, otters are much too helpful and happy
--
\/David DeLaney posting from d...@vic.com "It's not the pot that grows the flower
It's not the clock that slows the hour The definition's plain for anyone to see
Love is all it takes to make a family" - R&P. VISUALIZE HAPPYNET VRbeable<BLINK>
http://www.vic.com/~dbd/ - net.legends FAQ & Magic / I WUV you in all CAPS! --K.
Message has been deleted
Message has been deleted
Message has been deleted

G. Paul Ziemba

unread,
Jan 22, 2013, 11:59:30 AM1/22/13
to
ab...@mooli.org.uk (Peter Corlett) writes:

>[1] Hands up anybody who has used Perl formats.

*cough*

It seemed like a good idea at the time.


--
G. Paul Ziemba
FreeBSD unix:
8:56AM up 50 days, 14:34, 24 users, load averages: 0.40, 0.46, 0.42

Garrett Wollman

unread,
Jan 22, 2013, 1:56:16 PM1/22/13
to
In article <kdlptt$cs5$1...@mooli.org.uk>,
Peter Corlett <ab...@mooli.org.uk> wrote:
>[1] Hands up anybody who has used Perl formats. No? "man perlform" and weep.

[raises hand]

I believe I may still have some of that code in production. Hopefully
I'll find out when I have to replace it, rather than when it
mysteriously breaks.

-GAWollman
--
Garrett A. Wollman | What intellectual phenomenon can be older, or more oft
wol...@bimajority.org| repeated, than the story of a large research program
Opinions not shared by| that impaled itself upon a false central assumption
my employers. | accepted by all practitioners? - S.J. Gould, 1993

David Scheidt

unread,
Jan 22, 2013, 2:25:37 PM1/22/13
to
LP <use...@lpbk.net> wrote:
:on some combination of meth and ketamine with an attitude problem and a

Is there ketamine that doesn't have an attitude problem?

--
sig 71

Willem

unread,
Jan 22, 2013, 3:53:13 PM1/22/13
to
Peter Corlett wrote:
) [1] Hands up anybody who has used Perl formats. No? "man perlform" and weep. I
) particularly like the admonition to not put a lone dot on a line by itself,
) presumably due to a long-fixed bug in Sendmail.

No (although I have maintained code that used Perl formats).

But I *have* used the Cobol equivalent, which:
a) Worked
b) Made the course instructor go 'what the fuck is this?'


SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT

Wojciech Derechowski

unread,
Jan 22, 2013, 7:20:31 PM1/22/13
to
On Tue, 22 Jan 2013 18:56:16 +0000, Garrett Wollman wrote:
> In article <kdlptt$cs5$1...@mooli.org.uk>,
> Peter Corlett <ab...@mooli.org.uk> wrote:
>>[1] Hands up anybody who has used Perl formats. No? "man perlform" and weep.
>
> [raises hand]
>
> I believe I may still have some of that code in production. Hopefully
> I'll find out when I have to replace it, rather than when it
> mysteriously breaks.

Same here. Recently there was some speculation along the lines of
"why don't we use utf8 and see what happens."

Peter H. Coffin

unread,
Jan 22, 2013, 7:58:52 PM1/22/13
to
On Tue, 22 Jan 2013 10:33:33 +0000 (UTC), Peter Corlett wrote:
> [1] Hands up anybody who has used Perl formats. No? "man perlform" and weep. I
> particularly like the admonition to not put a lone dot on a line by itself,
> presumably due to a long-fixed bug in Sendmail.

Yeah, have. For report generation even. And it wasn't even ... Okay, it
was 2004, but that's not really THAT old, is it?

--
I think it's a beautiful day to go to the zoo and feed the ducks.
To the lions.
-- Brian Kantor

Steve VanDevender

unread,
Jan 23, 2013, 12:33:33 AM1/23/13
to
Satya <sat...@satyaonline.cjb.net> writes:

> On Tue, 22 Jan 2013 10:33:33 +0000 (UTC), Peter Corlett wrote:
>> [1] Hands up anybody who has used Perl formats. No? "man perlform" and weep. I
>> particularly like the admonition to not put a lone dot on a line by itself,
>> presumably due to a long-fixed bug in Sendmail.
>
> *hand*
>
> And I'm not even that old.

A lone dot on a line by itself goes farther back, to 'ed'. That may
have been the inspiration for the SMTP end-of-DATA indicator.

--
Steve VanDevender "I ride the big iron" http://hexadecimal.uoregon.edu/
ste...@hexadecimal.uoregon.edu PGP keyprint 4AD7AF61F0B9DE87 522902969C0A7EE8
Little things break, circuitry burns / Time flies while my little world turns
Every day comes, every day goes / 100 years and nobody shows -- Happy Rhodes
Message has been deleted
Message has been deleted

Garrett Wollman

unread,
Jan 23, 2013, 1:09:33 AM1/23/13
to
In article <kdnsnu$310$1...@thames.novusordo.net>,
Steve VanDevender <ste...@hexadecimal.uoregon.edu> wrote:

>A lone dot on a line by itself goes farther back, to 'ed'. That may
>have been the inspiration for the SMTP end-of-DATA indicator.

Did any of the people responsible for SMTP actually use any Unix
systems?

The single-dot-on-a-line was very common.

Thompson's QED for CTSS (and presumably also Multics) used control-F
for the same purpose, according to the CTSS Programmers Guide, 2/e, of
December, 1969, assuming I'm interpreting the notation correctly. I
don't know if this was also used by the original QED on the SDS-940.
Saltzer's TYPSET and other early CTSS editors all used an empty line
(i.e., CR CR) to exit input mode.

Notably, QED already had the g/re/p metaphor, except that in QED the
command went after the "G", so it was "GP/re/". One wonders if
Thompson changed this just to make it pronounceable.

<http://www.bitsavers.org/pdf/mit/ctss/CTSS_ProgrammersGuide_Dec69.pdf>

Maarten Wiltink

unread,
Jan 23, 2013, 1:57:38 AM1/23/13
to
"Michel" <ab...@rubberchicken.nl> wrote in message
news:gnj2t9-...@rubberchicken.nocrap...

> [...] _gb_ zlfdy. (Qonfr qrfvta[0] yvsgrq arneyl 1:1 sebz gur bevtvany
> syng grkg svyr.) ...

I thought this was the default model for zlfdy anyway.

In a parallel move to being able to program Pascal in any language,
I've long since drifted from relational data modelling to programming
an ORM in any database. Not being able to describe a foreign key at
the database level was rather a shock.

Tebrgwrf,
Maarten Wiltink


Maarten Wiltink

unread,
Jan 23, 2013, 1:59:39 AM1/23/13
to
"Shmuel (Seymour J.) Metz" <spam...@library.lspace.org.invalid> wrote in
message news:50fe9640$29$fuzhry+tra$mr2...@news.patriot.net...
[...]
> If you're only going to teach a single language, Perl might be a good
> choice, but it's missing a lot that is in other languages, e.g.,
> backtracking. ...

So, you'd need a teaching language that's everything to all people,
badly?

This language already exists. I'm sure of it.

Tebrgwrf,
Maarten Wiltink


Wojciech Derechowski

unread,
Jan 23, 2013, 3:57:49 AM1/23/13
to
On Wed, 23 Jan 2013 05:57:56 +0000, Satya wrote:
> On Wed, 23 Jan 2013 00:20:31 -0000, Wojciech Derechowski wrote:
>> Same here. Recently there was some speculation along the lines of
>> "why don't we use utf8 and see what happens."
>
> Now you have ? problems.

Oh yes: the meaning of life and how to dress for eternity. Problems
answering even the simple ones like so:

Q: Are you the Holy Old Man?
A: Anything but old, you motherfucker.

That sort of thing.
It is loading more messages.
0 new messages