Microsoft And The Environmentalism Of Cyberspace
Mark S. Bilk
clarify things:
Let's define "cyberspace" as the totality of the environ-
ment(s) in which programs run and data exists, on networks
and in individual computers. A very important aspect of it
is the commonly-used protocols by which programs and systems
communicate.
The relationship of cyberspace to programs and data (and
thus to the people who need to use them) is analogous to the
relationship of the planetary biological environment to the
plants and animals (including us) that live here. These
living things can only function by exchanging oxygen, CO2,
nitrogenous waste/fertilizer, minerals, water, etc., with
each other through the environment. Each has to be able to
take in and process the substances that certain others emit.
Likewise, in cyberspace, the common protocols are what allow
programs and systems to function, by enabling them to trans-
fer various types of data (or programs) among themselves.
And, as in the biosphere, it is also necessary to have a wide
variety of software available, so that a rich collection of
methods and facilities can evolve (through engineering) for
our use.
In the biological world, some people, to benefit themselves,
have created corporations that emit pollutants that poison
and kill plants and animals, and that destroy resources which
are needed by these living things.
In cyberspace, Microsoft has consistently done its best to
destroy various common protocols that the human race (in
its software) relies upon in doing its work. It has also
destroyed resources, in the form of alternative types of
software, that were (or could have become) very beneficial
to people. Microsoft's owners have harmed people by doing
all this, in order to enrich themselves.
We have learned that greedy wealthy people and their corpor-
ations *cannot* be allowed to destroy the biological environ-
ment, even though such action is permitted under laissez-faire
(Conservative/Libertarian) Capitalism, whose *de facto* rule
is that it's OK to harm other people in any way, in order to
benefit oneself, as long as you can get away with it. The
human race has found it necessary to work together, through
government, to regulate companies in order to prevent this.
We are only now learning that pollution and destruction of
the *cyberspace* environment -- its protocols and resources
-- can also have quite harmful effects on people. We must
therefore take steps to prevent such predatory activities,
either by undertaking governmental action, or by shunning
the products and business of the companies that do these
things, of which Microsoft is the worst.
Chad Myers
the paranoia, and the delusion to think up these things
and actually belive them!
A more accurate view on things is that several small
elitest computer engineers have created these cyberspace
protocols and functions that are near cryptic and unusable
to the average consumer in the market.
Microsoft has taken on the task of trying to bring these
difficult to use, poorly implemented, poorly documented,
and poorly designed protocols and trying to retrofit them
to make them usable by everyone instead of a few people
on a few systems.
Granted, sometimes these protocols and standards loose
some flexibility and functionality, but which is better, a
protocol or standard that a few computer science students
use, or that everyone uses?
Microsoft isn't right 100% of the time, but they do a pretty
damn good job with what they try. The problem is, is that stuck
up arrogant know-it-alls like you would rather bitch that MS is
"corrupting" these "standards" rather than helping them or beating
them by making these "standards" usable to everyone instead of
the elite few.
Linux is a prime example of this. Sure, linux does some things
really good and supports many of these standards, but is
functionally useless to the average consumer. So who benefits?
The elite few? How is that good for anyone, including cyberspace?
It seems people like you are on a crash course to ruining cyberspace
yourself by attempting to destroy Microsoft, the only entity _trying_
to make cyberspace easy enough for the layperson to use.
Chad
Mark S. Bilk <m...@netcom.com> wrote in message
news:7qmamm$5...@dfw-ixnews17.ix.netcom.com...
Tim Kelley
>
> What impresses me more in this is that you have the time,
> the paranoia, and the delusion to think up these things
> and actually belive them!
Actually, the analogy was rather "right-on". What is delusional
about what Mark wrote? Point out specifics.
> Microsoft has taken on the task of trying to bring these
> difficult to use, poorly implemented, poorly documented,
> and poorly designed protocols and trying to retrofit them
> to make them usable by everyone instead of a few people
> on a few systems.
Oh, the humanity! You are a fool. MS - and all other companies
- are out to do one thing: make money. While there is nothing
intrinsically wrong with that, it is pretty abominable to admire
those that make that their only goal in life, or to admire
corporations, which are nothing more than legal constructs. There
is an overwhelming case for abuse WRT Micros~1. If you actually
believe MS is on some holy mission to bring computing to the
masses, you are just buying into the corporate propaganda. Wake
up.
> Microsoft isn't right 100% of the time, but they do a pretty
> damn good job with what they try. The problem is, is that stuck
> up arrogant know-it-alls like you would rather bitch that MS is
> "corrupting" these "standards" rather than helping them or beating
> them by making these "standards" usable to everyone instead of
> the elite few.
If micros~1 is on a mission to bring computing to non-geeks, what
does that make the KDE and GNOME developers? What do you think
of them? Certainly, if you admire Micros~1 for their "mission",
you must love the KDE guys, right?
I don't see them trying to break any standards.
--
Tim Kelley
tpke...@winkinc.com
"If evolution is outlawed only outlaws will evolve"
Joseph T. Adams
: We have learned that greedy wealthy people and their corpor-
: ations *cannot* be allowed to destroy the biological environ-
: ment, even though such action is permitted under laissez-faire
: (Conservative/Libertarian) Capitalism, whose *de facto* rule
: is that it's OK to harm other people in any way, in order to
: benefit oneself, as long as you can get away with it.
I agree with much of the rest of your post.
You are incorrect regarding the libertarian position however (as are
most people who are not libertarians). Libertarians consider it wrong
to pollute anything that does not belong to you, or to initiate force
or fraud or otherwise harm any individual or entity without their
permission.
In a libertarian society you would *not* be allowed to pollute common
property, such as air, or any water that is ever in contact with
streams, groundwater, etc.
While libertarians call for the repeal of most existing
"environmental" laws on the ground that they exist to enrich some
entities rather than to protect the environment, we would replace them
with something considerably more draconian: a blanket prohibition
against polluting ANYTHING that you have not bought and paid for and
kept entirely separate from anything that is not yours. We understand
that this would have detrimental economic effects in the short term,
but, like others who believe that government should protect life,
liberty and property rather than threaten them, we know the long-term
benefits far outweigh the short-term cost.
Similarly, in a libertarian society, Microsoft would NOT be allowed to
steal the intellectual property of others, nor make false claims about
the reliability and interoperability of its products. It would not be
allowed to call something "HTML" which in fact was not, or "Java" if
it was not Java as defined by the owner of that trademark. It would
be able to make clean-room reimplementations of other people's
software, but anyone else would be allowed to make clean-room
reimplementations of its software as well.
While libertarians differ amongst themselves as to how "intellectual
property" should be handled, this libertarian would respectfully
suggest that the existing laws regarding torts, contracts, injunctions
to prevent irreparable harm, and other time-tested legal constructs
would be very adequate to protect what little intellectual property
actually *should* be protected. I also would suggest that unlike
physical property, which cannot be enjoyed by numerous individuals
simultaneously without a loss of utility, knowledge and information
are most valuable when shared, not hoarded, and at least arguably
ought not to be considered "property" at all if they are publicly
disclosed.
Joe
Greg Yantz
> A more accurate view on things is that several small
> elitest computer engineers have created these cyberspace
> protocols and functions that are near cryptic and unusable
> to the average consumer in the market.
No.
> Microsoft has taken on the task of trying to bring these
> difficult to use, poorly implemented, poorly documented,
> and poorly designed protocols and trying to retrofit them
> to make them usable by everyone instead of a few people
> on a few systems.
There is a tiny grain of truth here but reality is far more
complicated.
> Granted, sometimes these protocols and standards loose
> some flexibility and functionality, but which is better, a
> protocol or standard that a few computer science students
> use, or that everyone uses?
Neither. Try an open, documented protocol that can be
implemented by anyone, that fosters interoperability.
> Microsoft isn't right 100% of the time, but they do a pretty
> damn good job with what they try. The problem is, is that stuck
> up arrogant know-it-alls like you would rather bitch that MS is
> "corrupting" these "standards" rather than helping them or beating
> them by making these "standards" usable to everyone instead of
> the elite few.
Again, no.
> Linux is a prime example of this. Sure, linux does some things
> really good and supports many of these standards, but is
> functionally useless to the average consumer. So who benefits?
> The elite few? How is that good for anyone, including cyberspace?
Anyone who wants a system that is open, reliable, versatile,
powerful...
> It seems people like you are on a crash course to ruining cyberspace
> yourself by attempting to destroy Microsoft, the only entity _trying_
> to make cyberspace easy enough for the layperson to use.
This one is just laughably wrong.
> Chad
This is just so bad, so detached from reality, it reads like the
product of a marketing department. Do you really *believe* that
yourself? Amazing.
-Greg
Stephen Edwards
: Chad Myers wrote:
: > What impresses me more in this is that you have the time,
: > the paranoia, and the delusion to think up these things
: > and actually belive them!
: Actually, the analogy was rather "right-on". What is delusional
: about what Mark wrote? Point out specifics.
Gladly.
--
> In cyberspace, Microsoft has consistently done its best to
> destroy various common protocols that the human race (in
> its software) relies upon in doing its work. It has also
> destroyed resources, in the form of alternative types of
> software, that were (or could have become) very beneficial
> to people. Microsoft's owners have harmed people by doing
> all this, in order to enrich themselves.
--
"Harmed people"... give me a small break. If anything,
Microsoft has made people more productive... tell me, how
long has UNIX been around... since 1969, IIRC. Tell me,
exactly how productive were people with computers prior
to Microsoft's existence?... not very productive compared
to today. Chad is correct, for the most part... Microsoft
has brought the power of computers to the layperson. Why
they did it doesn't mean squat. The fact that they did it
means everything in the context of his reply to Mark's
paranoid blithering.
: > Microsoft has taken on the task of trying to bring these
: > difficult to use, poorly implemented, poorly documented,
: > and poorly designed protocols and trying to retrofit them
: > to make them usable by everyone instead of a few people
: > on a few systems.
: Oh, the humanity! You are a fool. MS - and all other companies
Chad's a fool for thinking that Microsoft has made computers
usable by the layperson? Wow... you must really have some
sharp insight into the matter then. </SARCASM>
: - are out to do one thing: make money. While there is nothing
And exactly where did Chad say "Microsoft did what they did for
the betterment of mankind"?
: intrinsically wrong with that, it is pretty abominable to admire
: those that make that their only goal in life, or to admire
: corporations, which are nothing more than legal constructs. There
Where did he say that he admired Microsoft? All he stated was
that Microsoft's presence had a positive effect on the ability
for average Joes to use computers.
: is an overwhelming case for abuse WRT Micros~1. If you actually
: believe MS is on some holy mission to bring computing to the
: masses, you are just buying into the corporate propaganda. Wake
: up.
Perhaps you need to read his posts a little more carefully.
: > Microsoft isn't right 100% of the time, but they do a pretty
: > damn good job with what they try. The problem is, is that stuck
: > up arrogant know-it-alls like you would rather bitch that MS is
: > "corrupting" these "standards" rather than helping them or beating
: > them by making these "standards" usable to everyone instead of
: > the elite few.
: If micros~1 is on a mission to bring computing to non-geeks, what
: does that make the KDE and GNOME developers? What do you think
: of them? Certainly, if you admire Micros~1 for their "mission",
: you must love the KDE guys, right?
I for one highly respect anyone who would write OpenSource software. But,
GNOME and KDE aren't fair examples, as they are merely "interfaces".
: I don't see them trying to break any standards.
Why should they? But if I want to make a proprietary standard, sure I'm
going to try to beat out the norm... I want my product to succeed.
--
.-----.
|[_] :| Stephen S. Edwards II | http://www.primenet.com/~rakmount
| = :| "Well, next time I'm in dire need of an opinion of absolutely
| | no value, you'll be the first to know." -- Chad Mulligan
|_..._|
void
>
>A more accurate view on things is that several small
>elitest computer engineers have created these cyberspace
>protocols and functions that are near cryptic and unusable
>to the average consumer in the market.
I don't see how the size of any given software engineer is relevant to
this debate.
Why is it that you have contempt for the people who invented modern
networks? The TCP/IP protocol suite has been around for some time, and
has proven remarkably adaptable to different usages. Compare this to
say NetBEUI, which is non-routable and therefore unsuitable for use on
anything besides a small LAN.
If TCP/IP is so cryptic and unusable, why and how has it been
implemented on every major platform? And since you're using it to post
here, are you some kind of elite uberhacker?
>Microsoft has taken on the task of trying to bring these
>difficult to use, poorly implemented, poorly documented,
>and poorly designed protocols and trying to retrofit them
>to make them usable by everyone instead of a few people
>on a few systems.
O the revisionism! Let's take on these claims one by one.
1) Difficult to use
TCP/IP isn't supposed to be user-friendly. It is not visible to the
user at all, generally. One build applications atop TCP/IP, which may
or may not be user-friendly.
2) Poorly implemented
That isn't the fault of the protocols nor their designers, now is it?
In any case, Microsoft's implementations have proven to be no less
buggy than anyone else's, and sometimes more. And they didn't even
have their own implementation until very late in the game.
3) Poorly designed
Would you like to substantiate that claim with some specific flaws that
you feel are present in the protocol suite? I believe that the
longevity and ubiquity of the TCP/IP protocols disprove this fairly
well, though there is certainly room for improvement.
4) Retrofitting
Microsoft has not to my knowledge proposed nor implemented any
extensions or changes to TCP, UDP, IP, ICMP, ARP, Ethernet, BGP, OSPF,
or any other protocol that makes the Internet run. They are now doing
some work with Dynamic DNS, but so are Cisco, DEC, the Internet
Software Consortium, and others.
>Microsoft isn't right 100% of the time, but they do a pretty
>damn good job with what they try. The problem is, is that stuck
>up arrogant know-it-alls like you would rather bitch that MS is
>"corrupting" these "standards" rather than helping them or beating
>them by making these "standards" usable to everyone instead of
>the elite few.
No, the problem is that Microsoft tries to leverage work done for free
or in the public interest by others, and then lock out other implementations
by making their extremely common software only work correctly with their
own implementation.
>Linux is a prime example of this. Sure, linux does some things
>really good and supports many of these standards, but is
>functionally useless to the average consumer. So who benefits?
>The elite few? How is that good for anyone, including cyberspace?
Linux, through KDE and GNOME, is well on its way to becoming a system that
can support a full range of home and office software. If you had any idea
what that entails, or any understanding of the kind and scope of
development that's happened in the free software world over the last
few years, you might have a little more respect for the software and
the people creating it.
Would I be incorrect if I guessed that you've never written a program
in your life?
--
Ben
[X] YES! I'm a brain-damaged lemur on crack, and I'd like to
order your software package for $459.95!
tony roth
Tim Kelley <tpke...@winkinc.com> wrote in message
news:37CE6DD4...@winkinc.com...
> Chad Myers wrote:
> >
> > What impresses me more in this is that you have the time,
> > the paranoia, and the delusion to think up these things
> > and actually belive them!
>
> Actually, the analogy was rather "right-on". What is delusional
> about what Mark wrote? Point out specifics.
how about this crap
In cyberspace, Microsoft has consistently done its best to
destroy various common protocols that the human race (in
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ name one protocol that ms has
destroyed
void
>
>I for one highly respect anyone who would write OpenSource software. But,
>GNOME and KDE aren't fair examples, as they are merely "interfaces".
Nothing "mere" about them. They are the glue that is necessary to
create a consistent look, feel and API to desktop apps. That
consistency is a genuine advantage that MacOS and Windows have over
unix at this point.
I predict adoption of KDE and/or GNOME by major commercial unix vendors
within three years. I bet Sun will be first, too, and perhaps IBM,
just because they've shown the most interest in open systems out of the
major unix vendors.
Tim Kelley
>
> Tim Kelley <tpke...@winkinc.com> wrote in message
> news:37CE6DD4...@winkinc.com...
> > Chad Myers wrote:
> > >
> > > What impresses me more in this is that you have the time,
> > > the paranoia, and the delusion to think up these things
> > > and actually belive them!
> >
> > Actually, the analogy was rather "right-on". What is delusional
> > about what Mark wrote? Point out specifics.
>
> how about this crap
>
> In cyberspace, Microsoft has consistently done its best to
> destroy various common protocols that the human race (in
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ name one protocol that ms has
> destroyed
He said "done their best" - read the halloween documents, it's
right there for you to see.
smtp/pop3 - would love to have everyone on exchange
html - Micros~1: "we're ?-ing the web"
Stephen Edwards
: On Thu, 2 Sep 1999 12:46:47 -0500, Chad Myers <cmy...@austin.rr.com> wrote:
: >A more accurate view on things is that several small
: >elitest computer engineers have created these cyberspace
: >protocols and functions that are near cryptic and unusable
: >to the average consumer in the market.
: I don't see how the size of any given software engineer is relevant to
: this debate.
It was for effect... NT users like drama.
: Why is it that you have contempt for the people who invented modern
: networks? The TCP/IP protocol suite has been around for some time, and
: has proven remarkably adaptable to different usages. Compare this to
: say NetBEUI, which is non-routable and therefore unsuitable for use on
: anything besides a small LAN.
: If TCP/IP is so cryptic and unusable, why and how has it been
: implemented on every major platform? And since you're using it to post
: here, are you some kind of elite uberhacker?
It has been implemented by "engineers" so that "users" can take advantage
of its usefulness. Chad is speaking from the user's perspective.
: >Microsoft has taken on the task of trying to bring these
: >difficult to use, poorly implemented, poorly documented,
: >and poorly designed protocols and trying to retrofit them
: >to make them usable by everyone instead of a few people
: >on a few systems.
: O the revisionism! Let's take on these claims one by one.
: 1) Difficult to use
: TCP/IP isn't supposed to be user-friendly. It is not visible to the
: user at all, generally. One build applications atop TCP/IP, which may
: or may not be user-friendly.
Yet, as Chad stated, Microsoft has made it easy for people to take
advantage of TCP/IP... your contention here means little, if anything.
: 2) Poorly implemented
: That isn't the fault of the protocols nor their designers, now is it?
: In any case, Microsoft's implementations have proven to be no less
: buggy than anyone else's, and sometimes more. And they didn't even
: have their own implementation until very late in the game.
This isn't a rebuttal. This is an excuse.
: 3) Poorly designed
: Would you like to substantiate that claim with some specific flaws that
: you feel are present in the protocol suite? I believe that the
: longevity and ubiquity of the TCP/IP protocols disprove this fairly
: well, though there is certainly room for improvement.
I'm not knowledgable to comment on this one.
: 4) Retrofitting
: Microsoft has not to my knowledge proposed nor implemented any
: extensions or changes to TCP, UDP, IP, ICMP, ARP, Ethernet, BGP, OSPF,
: or any other protocol that makes the Internet run. They are now doing
: some work with Dynamic DNS, but so are Cisco, DEC, the Internet
: Software Consortium, and others.
But I thought it was typical for Microsoft to munge a standard and then
make it their own... now they _don't_ do that? I'm confused...
: >Microsoft isn't right 100% of the time, but they do a pretty
: >damn good job with what they try. The problem is, is that stuck
: >up arrogant know-it-alls like you would rather bitch that MS is
: >"corrupting" these "standards" rather than helping them or beating
: >them by making these "standards" usable to everyone instead of
: >the elite few.
: No, the problem is that Microsoft tries to leverage work done for free
: or in the public interest by others, and then lock out other implementations
: by making their extremely common software only work correctly with their
: own implementation.
So they do it after all? Now I'm really confused.
: >Linux is a prime example of this. Sure, linux does some things
: >really good and supports many of these standards, but is
: >functionally useless to the average consumer. So who benefits?
: >The elite few? How is that good for anyone, including cyberspace?
: Linux, through KDE and GNOME, is well on its way to becoming a system that
: can support a full range of home and office software. If you had any idea
: what that entails, or any understanding of the kind and scope of
: development that's happened in the free software world over the last
: few years, you might have a little more respect for the software and
: the people creating it.
: Would I be incorrect if I guessed that you've never written a program
: in your life?
If he said no, would you believe him?
TNC
>
> Microsoft has taken on the task of trying to bring these
> difficult to use, poorly implemented, poorly documented,
> and poorly designed protocols and trying to retrofit them
> to make them usable by everyone instead of a few people
> on a few systems.
This is false. The original standards are always, by definition, well
documented and generally well designed. The docuemtation is almost
always available in the form of RFCs on the web for free. You have, in
fact, completely inverted the actual situation. It is *Microsoft's*
crocked-together, re-implimentations of the protocols which are usually
poorly documented and/or documented only in "developers kits" for which
you must pay MS hundreds of dollars.
Microsoft's "retrofitting" always follows their strategy of "embrace and
extend" which is basically to take something that everyone else has
agreed upon and and then modify it in subtle ways so that developers and
consumers are forced to choose between the MS implementation and that
used by the rest of the world. The canonical example of this is the way
they corrupted Java in their J++ product, which is now effectively a
Windows-only language.
I will grant you only this - the original poster's point about the
environment was a weak analogy.
--
"Gun control proposals are nothing more than a modern liberal
suggestion that government, which is unable to protect its citizens,
makes sure those citizens cannot defend themselves."
- Robert. H. Bork
Stephen Edwards
: Chad Myers wrote:
: > What impresses me more in this is that you have the time,
: > the paranoia, and the delusion to think up these things
: > and actually belive them!
: Actually, the analogy was rather "right-on". What is delusional
: about what Mark wrote? Point out specifics.
Gladly.
--
> In cyberspace, Microsoft has consistently done its best to
> destroy various common protocols that the human race (in
> its software) relies upon in doing its work. It has also
> destroyed resources, in the form of alternative types of
> software, that were (or could have become) very beneficial
> to people. Microsoft's owners have harmed people by doing
> all this, in order to enrich themselves.
--
"Harmed people"... give me a small break. If anything,
Microsoft has made people more productive... tell me, how
long has UNIX been around... since 1969, IIRC. Tell me,
exactly how productive were people with computers prior
to Microsoft's existence?... not very productive compared
to today. Chad is correct, for the most part... Microsoft
has brought the power of computers to the layperson. Why
they did it doesn't mean squat. The fact that they did it
means everything in the context of his reply to Mark's
paranoid blithering.
: > Microsoft has taken on the task of trying to bring these
: > difficult to use, poorly implemented, poorly documented,
: > and poorly designed protocols and trying to retrofit them
: > to make them usable by everyone instead of a few people
: > on a few systems.
: Oh, the humanity! You are a fool. MS - and all other companies
Chad's a fool for thinking that Microsoft has made computers
usable by the layperson? Wow... you must really have some
sharp insight into the matter then. </SARCASM>
: - are out to do one thing: make money. While there is nothing
And exactly where did Chad say "Microsoft did what they did for
the betterment of mankind"?
: intrinsically wrong with that, it is pretty abominable to admire
: those that make that their only goal in life, or to admire
: corporations, which are nothing more than legal constructs. There
Where did he say that he admired Microsoft? All he stated was
that Microsoft's presence had a positive effect on the ability
for average Joes to use computers.
: is an overwhelming case for abuse WRT Micros~1. If you actually
: believe MS is on some holy mission to bring computing to the
: masses, you are just buying into the corporate propaganda. Wake
: up.
Perhaps you need to read his posts a little more carefully.
: > Microsoft isn't right 100% of the time, but they do a pretty
: > damn good job with what they try. The problem is, is that stuck
: > up arrogant know-it-alls like you would rather bitch that MS is
: > "corrupting" these "standards" rather than helping them or beating
: > them by making these "standards" usable to everyone instead of
: > the elite few.
: If micros~1 is on a mission to bring computing to non-geeks, what
: does that make the KDE and GNOME developers? What do you think
: of them? Certainly, if you admire Micros~1 for their "mission",
: you must love the KDE guys, right?
I for one highly respect anyone who would write OpenSource software. But,
GNOME and KDE aren't fair examples, as they are merely "interfaces".
: I don't see them trying to break any standards.
Why should they? But if I want to make a proprietary standard, sure I'm
going to try to beat out the norm... I want my product to succeed.
void
>void <fl...@interport.net> wrote:
>
>: On Thu, 2 Sep 1999 12:46:47 -0500, Chad Myers <cmy...@austin.rr.com> wrote:
>
>: >A more accurate view on things is that several small
>: >elitest computer engineers have created these cyberspace
>: >protocols and functions that are near cryptic and unusable
>: >to the average consumer in the market.
>
>: I don't see how the size of any given software engineer is relevant to
>: this debate.
>
>It was for effect... NT users like drama.
Why don't you let the man answer for his own grammatical mishaps? I
understood what he meant and was gently poking fun at him. Anyway,
you've written off a number of things to "dramatic effect" or "sarcasm"
over the last day or two. Maybe if you put a little more thought into
your posts, you wouldn't have to backpedal so much.
>: If TCP/IP is so cryptic and unusable, why and how has it been
>: implemented on every major platform? And since you're using it to post
>: here, are you some kind of elite uberhacker?
>
>It has been implemented by "engineers" so that "users" can take advantage
>of its usefulness. Chad is speaking from the user's perspective.
Chad is speaking from an ignoramus' perspective. The fact is, the
TCP/IP protocols are not user friendly nor user unfriendly. They are
simply not for users to mess with. It's the applications that use them
that can be friendly or unfriendly.
>: >Microsoft has taken on the task of trying to bring these
>: >difficult to use, poorly implemented, poorly documented,
>: >and poorly designed protocols and trying to retrofit them
>: >to make them usable by everyone instead of a few people
>: >on a few systems.
>
>: O the revisionism! Let's take on these claims one by one.
>
>: 1) Difficult to use
>
>: TCP/IP isn't supposed to be user-friendly. It is not visible to the
>: user at all, generally. One build applications atop TCP/IP, which may
>: or may not be user-friendly.
>
>Yet, as Chad stated, Microsoft has made it easy for people to take
>advantage of TCP/IP... your contention here means little, if anything.
Microsoft has done nothing for TCP/IP that others didn't do first.
Name a program from Microsoft that uses TCP/IP and I will show you a prior
implementation by someone else that was just as user-friendly.
>: 2) Poorly implemented
>
>: That isn't the fault of the protocols nor their designers, now is it?
>: In any case, Microsoft's implementations have proven to be no less
>: buggy than anyone else's, and sometimes more. And they didn't even
>: have their own implementation until very late in the game.
>
>This isn't a rebuttal. This is an excuse.
What is there to rebut? I might as well say that telnet is a bad idea
because Microsoft's implementation is so bad. It's nonsensical.
>: 3) Poorly designed
>
>: Would you like to substantiate that claim with some specific flaws that
>: you feel are present in the protocol suite? I believe that the
>: longevity and ubiquity of the TCP/IP protocols disprove this fairly
>: well, though there is certainly room for improvement.
>
>I'm not knowledgable to comment on this one.
That hasn't stopped you with the other ones.
>: 4) Retrofitting
>
>: Microsoft has not to my knowledge proposed nor implemented any
>: extensions or changes to TCP, UDP, IP, ICMP, ARP, Ethernet, BGP, OSPF,
>: or any other protocol that makes the Internet run. They are now doing
>: some work with Dynamic DNS, but so are Cisco, DEC, the Internet
>: Software Consortium, and others.
>
>But I thought it was typical for Microsoft to munge a standard and then
>make it their own... now they _don't_ do that? I'm confused...
Not with the protocols I named. They've done their best to pervert
SMTP, Java, HTML, etc., etc. ... see http://www.opensource.org/halloween.
They're also trying to impose Dynamic DNS on the public before it
(DDNS) is ready.
>: No, the problem is that Microsoft tries to leverage work done for free
>: or in the public interest by others, and then lock out other implementations
>: by making their extremely common software only work correctly with their
>: own implementation.
>
>So they do it after all? Now I'm really confused.
You certainly are. See above.
>: Would I be incorrect if I guessed that you've never written a program
>: in your life?
>
>If he said no, would you believe him?
No, but I wouldn't run his code, either. Why don't you work on prying
your own foot out of your mouth before you go rushing around to assist
others with similar tasks?
Brent Davies
void <fl...@interport.net> wrote in message
news:slrn7stmcj...@interport.net...
| On 2 Sep 1999 19:27:39 GMT, Stephen Edwards <ja20...@primenet.com>
wrote:
| >
| >I for one highly respect anyone who would write OpenSource software.
But,
| >GNOME and KDE aren't fair examples, as they are merely "interfaces".
|
| create a consistent look, feel and API to desktop apps. That
| consistency is a genuine advantage that MacOS and Windows have over
| unix at this point.
|
| I predict adoption of KDE and/or GNOME by major commercial unix
vendors
| within three years. I bet Sun will be first, too, and perhaps IBM,
| just because they've shown the most interest in open systems out of
the
| major unix vendors.
Why do you think that KDE will be adopted over CDE? I have not used KDE
yet, so this is a question for my knowledge, not a jab.
-B
Brent Davies
[snip]
|
| Microsoft has done nothing for TCP/IP that others didn't do first.
| Name a program from Microsoft that uses TCP/IP and I will show you a
prior
| implementation by someone else that was just as user-friendly.
The "user-friendly" part is not the most important aspect. Microsoft
didn't have to implement TCP/IP on Windows. They did mostly because of
the profit involved with higher sales. However, the fact that Windows
had the largest install base makes MS the single most successful
distributor of the TCP/IP protocol. Because of their market share, they
brought TCP/IP to more end users than any other company.
Did MS plan it this way? Probably
Did MS do this for noble reasons? Probably not
Why did they do it? Mostly for money.
Has it bettered the end users' computing experience? Most definitely.
I don't believe that the ends justify the means, but no one can say that
MS has not "bettered" computing in general.
[snip]
-B
Stephen Edwards
: Chad Myers wrote:
: > Microsoft has taken on the task of trying to bring these
: > difficult to use, poorly implemented, poorly documented,
: > and poorly designed protocols and trying to retrofit them
: > to make them usable by everyone instead of a few people
: > on a few systems.
: This is false. The original standards are always, by definition, well
: documented and generally well designed. The docuemtation is almost
: always available in the form of RFCs on the web for free. You have, in
And this differs from Microsoft's implementations _how_ exactly?
: fact, completely inverted the actual situation. It is *Microsoft's*
: crocked-together, re-implimentations of the protocols which are usually
: poorly documented and/or documented only in "developers kits" for which
: you must pay MS hundreds of dollars.
Hmmm... a search of "TCP/IP" at http://www.microsoft.com/ yields
http://www.microsoft.com/NTServer/nts/techdetails/compares/TCPIntrowp.asp
http://www.microsoft.com/SYSPRO/deploy/tcpintro/tcpintro14.htm
Is this the documentation that doesn't exist?
I guess I wasted a lot of money on those WindowsNT books, because I'm sure
there's no documentation whatsoever in them about TCP/IP.
And as for "developer kits"... many operating systems have them.
As I understand it, MS developer kits are meant for serious developers,
and not "hackers". OpenSource users and developers often have a difficult
time grasping the point of commercial software... to make something
commercially viable... therefore, to make money by providing a service or
product professionally.
: Microsoft's "retrofitting" always follows their strategy of "embrace and
: extend" which is basically to take something that everyone else has
: agreed upon and and then modify it in subtle ways so that developers and
: consumers are forced to choose between the MS implementation and that
: used by the rest of the world. The canonical example of this is the way
: they corrupted Java in their J++ product, which is now effectively a
: Windows-only language.
So, you're suggesting that the _only_ reason they made those modifications
was to screw up Sun's Java agenda? I guess it's too much to consider that
Microsoft may have actually been attempting to improve upon it for use
with Windows operating systems.
Even if Microsoft is responsible for hindering Java, I would sooner thank
them, than detest them.
: I will grant you only this - the original poster's point about the
: environment was a weak analogy.
It wasn't weak... it wasn't even an analogy. It was typical Mark
S. Bilk nonsense.
D'Arcy Smith
> : The canonical example of this is the way
> : they corrupted Java in their J++ product, which is now effectively a
> : Windows-only language.
> So, you're suggesting that the _only_ reason they made those modifications
> was to screw up Sun's Java agenda? I guess it's too much to consider that
> Microsoft may have actually been attempting to improve upon it for use
> with Windows operating systems.
Ah here we go (yipeeee!)...
Here is a perfect example of MS not being able to follow a
"standard". In the case the Java Language Specification.
The MS extensions explicitly break the JLS. Now they could
have gone off and provided a set of Win32 JNI libs to do what
they did... but that would be portable to other VMs on Win32.
So they took the co-opting way of doing things... and in doing
so (lets say it again) brok the Java Language Specification.
Thanks for providing a perfect example of MS extending a "standard".
Also read the "Halloween Documents"
(http://www.opensource.org/halloween)
"Linux's homebase is currently commodity network and server
infrastructure.
By folding extended functionality into today's commodity services and
create new protocols, we raise the bar & change the rules of the game."
The particular point of interest here is:
"By folding extended functionality into today's commodity services"
..darcy
Stephen Edwards
: On 2 Sep 1999 20:43:34 GMT, Stephen Edwards <ja20...@primenet.com> wrote:
: >void <fl...@interport.net> wrote:
: >
: >: On Thu, 2 Sep 1999 12:46:47 -0500, Chad Myers <cmy...@austin.rr.com> wrote:
: >
: >: >A more accurate view on things is that several small
: >: >elitest computer engineers have created these cyberspace
: >: >protocols and functions that are near cryptic and unusable
: >: >to the average consumer in the market.
: >
: >: I don't see how the size of any given software engineer is relevant to
: >: this debate.
: >
: >It was for effect... NT users like drama.
: Why don't you let the man answer for his own grammatical mishaps? I
: understood what he meant and was gently poking fun at him. Anyway,
I see... so my whimsical statement was uncalled for... so sorry if I
offended you... sheesh.
: you've written off a number of things to "dramatic effect" or "sarcasm"
: over the last day or two. Maybe if you put a little more thought into
: your posts, you wouldn't have to backpedal so much.
I would advise you to do the same.
: >: If TCP/IP is so cryptic and unusable, why and how has it been
: >: implemented on every major platform? And since you're using it to post
: >: here, are you some kind of elite uberhacker?
: >
: >It has been implemented by "engineers" so that "users" can take advantage
: >of its usefulness. Chad is speaking from the user's perspective.
: Chad is speaking from an ignoramus' perspective. The fact is, the
: TCP/IP protocols are not user friendly nor user unfriendly. They are
: simply not for users to mess with. It's the applications that use them
: that can be friendly or unfriendly.
And? _So_ _what_. His point is that before Microsoft came around, there
was no way for a regular user to configure a machine to connect to the
internet. Microsoft, among others, have made dealing with TCP/IP easier
so that more people could do it on their own. I believe that was his
point.
: >: >Microsoft has taken on the task of trying to bring these
: >: >difficult to use, poorly implemented, poorly documented,
: >: >and poorly designed protocols and trying to retrofit them
: >: >to make them usable by everyone instead of a few people
: >: >on a few systems.
: >
: >: O the revisionism! Let's take on these claims one by one.
: >
: >: 1) Difficult to use
: >
: >: TCP/IP isn't supposed to be user-friendly. It is not visible to the
: >: user at all, generally. One build applications atop TCP/IP, which may
: >: or may not be user-friendly.
: >
: >Yet, as Chad stated, Microsoft has made it easy for people to take
: >advantage of TCP/IP... your contention here means little, if anything.
: Microsoft has done nothing for TCP/IP that others didn't do first.
Sure they did... they made it easier for users to configure their systems
to use it... in most cases, the user doesn't even have to see anything
resembling numbers. Tell me, when were UNIX variants using dial-up
utilities like Microsoft's that were configured by end users?
: Name a program from Microsoft that uses TCP/IP and I will show you a prior
: implementation by someone else that was just as user-friendly.
Okay... the Dial-up Networking utility for one... editing /etc files, and
typing pppd isn't intuitive for most users, in case you haven't noticed.
Tell me, how would average joe user know how to connect to the internet?
He'd need an expert to explain it to him, that's how. With Microsoft
software, there are Wizards to guide a user though it.
: >: 2) Poorly implemented
: >
: >: That isn't the fault of the protocols nor their designers, now is it?
: >: In any case, Microsoft's implementations have proven to be no less
: >: buggy than anyone else's, and sometimes more. And they didn't even
: >: have their own implementation until very late in the game.
: >
: >This isn't a rebuttal. This is an excuse.
: What is there to rebut? I might as well say that telnet is a bad idea
: because Microsoft's implementation is so bad. It's nonsensical.
The burden for debunking is yours... if you can't, then don't give a
half-assed attempt to blur the issue... just let it go.
: >: 3) Poorly designed
: >
: >: Would you like to substantiate that claim with some specific flaws that
: >: you feel are present in the protocol suite? I believe that the
: >: longevity and ubiquity of the TCP/IP protocols disprove this fairly
: >: well, though there is certainly room for improvement.
: >
: >I'm not knowledgable to comment on this one.
: That hasn't stopped you with the other ones.
Cute. Uncalled for, and trite... how fitting of you.
: >: 4) Retrofitting
: >
: >: Microsoft has not to my knowledge proposed nor implemented any
: >: extensions or changes to TCP, UDP, IP, ICMP, ARP, Ethernet, BGP, OSPF,
: >: or any other protocol that makes the Internet run. They are now doing
: >: some work with Dynamic DNS, but so are Cisco, DEC, the Internet
: >: Software Consortium, and others.
: >
: >But I thought it was typical for Microsoft to munge a standard and then
: >make it their own... now they _don't_ do that? I'm confused...
: Not with the protocols I named. They've done their best to pervert
: SMTP, Java, HTML, etc., etc. ... see http://www.opensource.org/halloween.
: They're also trying to impose Dynamic DNS on the public before it
: (DDNS) is ready.
: >: No, the problem is that Microsoft tries to leverage work done for free
: >: or in the public interest by others, and then lock out other implementations
: >: by making their extremely common software only work correctly with their
: >: own implementation.
: >
: >So they do it after all? Now I'm really confused.
: You certainly are. See above.
I don't need to... I can see all of this just fine.
: >: Would I be incorrect if I guessed that you've never written a program
: >: in your life?
: >
: >If he said no, would you believe him?
: No, but I wouldn't run his code, either. Why don't you work on prying
: your own foot out of your mouth before you go rushing around to assist
: others with similar tasks?
Ben, you can't make an issue of something, and then duck into a corner,
waving your fist out hoping to drive someone off... arguing doesn't work
that way.
Chad's point was that Microsoft has made computing easier for the
layperson... his point was accurate. Picking grammatical details
apart is useless, and proves nothing to counter his general assertion.
Jack Richards
Mark S. Bilk wrote in message <7qmamm$5...@dfw-ixnews17.ix.netcom.com>...
Gee, I was under the impression that the technologies that major
corporations generate is the best hope for the enviroment. Basically, you
seem to think that your hot air is actually doing something.
Please forgo using any such products for a month. Then report back.
You have the indelible ability to generate cant. \
>
>In cyberspace, Microsoft has consistently done its best to
>destroy various common protocols that the human race (in
>its software) relies upon in doing its work. It has also
>destroyed resources, in the form of alternative types of
>software, that were (or could have become) very beneficial
>to people. Microsoft's owners have harmed people by doing
>all this, in order to enrich themselves.
Actually MS has been associated with the personal computer revolution
empowering users throughout the world. OTOH, there are elements against MS
that are very powerfully funded, who have a history of espousing ideals that
shield their very real self interest and fear of competition that may
diminish their skillsets.
They have, for the lack of courage, participated in the dumbing down of the
university, by following every piece of work that has come down the pike.
They use non-profit organizations to mock real debate. They are now an
establishment who fear being overthrown.
They fear the user friendly desktop may take over
>
>We have learned that greedy wealthy people and their corpor-
>ations *cannot* be allowed to destroy the biological environ-
>ment, even though such action is permitted under laissez-faire
>(Conservative/Libertarian) Capitalism, whose *de facto* rule
>is that it's OK to harm other people in any way, in order to
>benefit oneself, as long as you can get away with it. The
>human race has found it necessary to work together, through
>government, to regulate companies in order to prevent this.
No doubt corporations, who of course require government regulation, have
supplied every article that you use, for head to toe, at a price affordable
to the vast bulk of humanity, especially computers. Behind the pompous
rhetoric is a jealousy that the world does not follow your precious self,
athough legions like you do their best to impose their views.
>We are only now learning that pollution and destruction of
>the *cyberspace* environment -- its protocols and resources
>-- can also have quite harmful effects on people. We must
>therefore take steps to prevent such predatory activities,
>either by undertaking governmental action, or by shunning
>the products and business of the companies that do these
>things, of which Microsoft is the worst.
Well cliche-worded, is this your best spin? What emission generated you?
>
>
Donovan Rebbechi
>So, you're suggesting that the _only_ reason they made those modifications
>was to screw up Sun's Java agenda? I guess it's too much to consider that
>Microsoft may have actually been attempting to improve upon it for use
>with Windows operating systems.
Shipping OS specific functions under java.* instead of win32.* ?
I would call that willful sabotage.
>Even if Microsoft is responsible for hindering Java, I would sooner thank
>them, than detest them.
Why is that ? What kind of market will we have if everyone works
at sabotaging their competitors products, rather than developing
their own ?
--
Donovan
Erik Funkenbusch
> tony roth wrote:
> >
> > Tim Kelley <tpke...@winkinc.com> wrote in message
> > news:37CE6DD4...@winkinc.com...
> > > >
> > > > What impresses me more in this is that you have the time,
> > > > the paranoia, and the delusion to think up these things
> > > > and actually belive them!
> > >
> > > Actually, the analogy was rather "right-on". What is delusional
> > > about what Mark wrote? Point out specifics.
> >
> >
> > In cyberspace, Microsoft has consistently done its best to
> > destroy various common protocols that the human race (in
has
> > destroyed
>
> He said "done their best" - read the halloween documents, it's
> right there for you to see.
>
> smtp/pop3 - would love to have everyone on exchange
> html - Micros~1: "we're ?-ing the web"
To my knowledge, MS has never tried to "destroy" a protocol. Favored other
protocols perhaps, but destroy?
Microsoft understands the value of SMTP/POP/IMAP. Exchange does not work
well over a slow interface like a modem dialup (hell, it doesn't work that
great over a T1 if it's got heavy useage). I don't think MS wants to kill
SMTP or POP3, they simply have a better way to do this for their own
software on an internal network.
Hobbyist ©
> Chad's point was that Microsoft has made computing easier for the
> layperson... his point was accurate. Picking grammatical details
> apart is useless, and proves nothing to counter his general assertion.
That's the only thing that can be done in defence of the point.
Kinda change the issue and tear apart inaccuracies stated in
effort to bring across the point.
I fall victim to that sorta thing all the time on usenet.
--
-=Ali=-
Chad Mulligan
wrote:
>>void <fl...@interport.net> wrote:
>>
<snip>
>
>Not exactly. Microsoft picked up TCP/IP because other people were
selling
>it for their platform and making money. TCP/IP had already "made it",
>so Microsoft decided to make money off of it. *Ordinary end-users were
>already using TCP/IP on home computers before Microsoft started selling
>a TCP/IP implementation.*
>
Only after the PPP RFC was published, at about that same time, MS along
with Intel were heavily funding the Winsock 2 spec. MS switched to
TCP/IP because IT managers were demanding single protocol environments,
that's the same reason Novell switched too.
>
>Trumpet did that first. Leave unix out of it. Point is: no innovation
>from Microsoft's quarter here. All they did was recognize someone
>else's good idea, copy it and integrate it into their OS. Poorly, as
>usual.
>
MS's native TCP/IP isn't really that bad, the NetBios extensions are
somewhat troublesome primarily because no one is really sure what it's
doing.
>>: Name a program from Microsoft that uses TCP/IP and I will show you a
prior
>>: implementation by someone else that was just as user-friendly.
>>
>>Okay... the Dial-up Networking utility for one... editing /etc files,
and
>>typing pppd isn't intuitive for most users, in case you haven't
noticed.
>
>Not a Microsoft first, sorry.
>
SMB transport over TCP/IP. Don't give me any Samba stuff either, MS
wrote Lan Manager in the early eighties as an OEM implementation and
AT&T served it over TCP/IP.
>>: >: 2) Poorly implemented
>>: >
>>: >: That isn't the fault of the protocols nor their designers, now is
it?
>>: >: In any case, Microsoft's implementations have proven to be no
less
>>: >: buggy than anyone else's, and sometimes more. And they didn't
even
>>: >: have their own implementation until very late in the game.
>>: >
>>: >This isn't a rebuttal. This is an excuse.
>>
>>: What is there to rebut? I might as well say that telnet is a bad
idea
>>: because Microsoft's implementation is so bad. It's nonsensical.
>>
>>The burden for debunking is yours... if you can't, then don't give a
>>half-assed attempt to blur the issue... just let it go.
>
>There is no issue to blur. The "issue" arises from the original
>poster's ignorance of history (and of logical types, but that's a more
>abstract argument, not worth pursuing, since I have a much better one).
>He criticizes TCP/IP and the TCP/IP designers for the so-called poor
>implementations of TCP/IP. He's probably thinking of teardrop and
boink
>and friends when he says that.
>
>When TCP/IP was originally written, there was no such thing as network
>security. It simply wasn't an issue; anyone on the network was
trusted.
>Nobody knew that in the late 90s, it would suddenly become necessary to
>have bulletproof networking stacks.
>
<snipped>
Stuart Fox
> smtp/pop3 - would love to have everyone on exchange
Well don't just include MS in that, I'm sure IBM would like everyone to use
Notes. And btw, Exchange supports POP3 and SMTP. The fact is, these
protocols aren't really that good for use in a large corporate.
J.T.Kay
>So, you're suggesting that the _only_ reason they made those modifications
>was to screw up Sun's Java agenda?
That is exactly what the Microsoft executives were saying to each
other in email. I would think the MS executives know their intent
quite a bit better than you do.
Their email was subpoena'd and produced as evidence in the Microsoft
anti-trust trial, which enabled the public to read it too.
"Do we have a clear plan on what we want Apple to do to
undermine Sun?" -- email from Bill Gates.
Microsoft has been leading the entire computer industry around by the nose
for more than a decade. It even succeeded in cowing IBM's PC division,
and Compaq, by threatening to impose unfavorable licensing terms for
its monopoly product Windows.
But this situation will not last for ever. The groundswell of support
behind Linux, behind Java, and soon behind StarOffice will be a tidal
wave that Redmond doesn't understand, and can do little except run from.
Look for a steep drop in MSFT stock price in the coming months.
void
wrote:
>
>The "user-friendly" part is not the most important aspect.
Now that's something I haven't heard from your camp often.
>Microsoft
>didn't have to implement TCP/IP on Windows. They did mostly because of
>the profit involved with higher sales.
They did because TCP/IP was more popular than their proprietary
protocols, because it had flexibility and power and was free, and people
were buying third-party implementations.
void
>
> Are you kidding, Stephen? Furrfu. People were connecting machines to
>(pitiful forerunners to) the Net back when Microsoft was selling Typing
>Tutor. Microsoft was _always_ _around_ and did not 'come around'. And if
>Microsoft was instrumental in bringing TCP/IP support to (surprise!) THEIR
>OWN FSCKING OPERATING SYSTEMS, well, color _me_ astonished ;P
Only after Trumpet proved it was profitable, mind you.
void
>void <fl...@interport.net> wrote:
>
>: Chad is speaking from an ignoramus' perspective. The fact is, the
>: TCP/IP protocols are not user friendly nor user unfriendly. They are
>: simply not for users to mess with. It's the applications that use them
>: that can be friendly or unfriendly.
>
>And? _So_ _what_. His point is that before Microsoft came around, there
>was no way for a regular user to configure a machine to connect to the
>internet. Microsoft, among others, have made dealing with TCP/IP easier
>so that more people could do it on their own. I believe that was his
>point.
Not exactly. Microsoft picked up TCP/IP because other people were selling
it for their platform and making money. TCP/IP had already "made it",
so Microsoft decided to make money off of it. *Ordinary end-users were
already using TCP/IP on home computers before Microsoft started selling
a TCP/IP implementation.*
>: Microsoft has done nothing for TCP/IP that others didn't do first.
>
>Sure they did... they made it easier for users to configure their systems
>to use it... in most cases, the user doesn't even have to see anything
>resembling numbers. Tell me, when were UNIX variants using dial-up
>utilities like Microsoft's that were configured by end users?
Trumpet did that first. Leave unix out of it. Point is: no innovation
from Microsoft's quarter here. All they did was recognize someone
else's good idea, copy it and integrate it into their OS. Poorly, as
usual.
>: Name a program from Microsoft that uses TCP/IP and I will show you a prior
>: implementation by someone else that was just as user-friendly.
>
>Okay... the Dial-up Networking utility for one... editing /etc files, and
>typing pppd isn't intuitive for most users, in case you haven't noticed.
Not a Microsoft first, sorry.
>: >: 2) Poorly implemented
>: >
>: >: That isn't the fault of the protocols nor their designers, now is it?
>: >: In any case, Microsoft's implementations have proven to be no less
>: >: buggy than anyone else's, and sometimes more. And they didn't even
>: >: have their own implementation until very late in the game.
>: >
>: >This isn't a rebuttal. This is an excuse.
>
>: What is there to rebut? I might as well say that telnet is a bad idea
>: because Microsoft's implementation is so bad. It's nonsensical.
>
>The burden for debunking is yours... if you can't, then don't give a
>half-assed attempt to blur the issue... just let it go.
There is no issue to blur. The "issue" arises from the original
poster's ignorance of history (and of logical types, but that's a more
abstract argument, not worth pursuing, since I have a much better one).
He criticizes TCP/IP and the TCP/IP designers for the so-called poor
implementations of TCP/IP. He's probably thinking of teardrop and boink
and friends when he says that.
When TCP/IP was originally written, there was no such thing as network
security. It simply wasn't an issue; anyone on the network was trusted.
Nobody knew that in the late 90s, it would suddenly become necessary to
have bulletproof networking stacks.
But when the networking stacks needed to be bulletproof, they were made
that way, without changes to the underlying protocols.
Over a quarter century of constantly expanding use, it's natural for a
protocol or piece of software to evolve, as new use cases arise and new
circumstances pertain. But I'll tell you what, I'll admit that TCP/IP
and its implementations are shoddy if you or Chad will just undertake
one simple task. All I want you to do is create some technology that's
used for 25 years, by millions of people, without any bugfixes or other
modifications.
No no, wait, that's unfair. I mean, I criticize Windows for being
crashy, but I haven't written an OS at all, let alone one that stays up
for more than 49 days. So here's a much easier project: just point me
to such a technology. Just name one protocol or piece of software that
fulfills the standards you demand of TCP/IP, and I will concede this
point.
You won't find one, because most software doesn't last very long at all,
because most software sucks quite a bit more than TCP/IP.
NetBEUI, anyone? Designed long after IP, and obsoleted by -- that's
right -- IP. Whaddaya know.
>Chad's point was that Microsoft has made computing easier for the
>layperson... his point was accurate.
Ladies and gentlemen, I give you: the dancing paperclip. A Microsoft
first!
Deep Blue
>
> void <fl...@interport.net> wrote:
>
> : On Thu, 2 Sep 1999 12:46:47 -0500, Chad Myers <cmy...@austin.rr.com> wrote:
> Yet, as Chad stated, Microsoft has made it easy for people to take
> advantage of TCP/IP... your contention here means little, if anything.
>Actually if I recall corectly the first TCP/IP stack for windows was written
by Trumpet software and NOT by Microsoft.
Brent Davies
Erik Funkenbusch <er...@visi.com> wrote in message
news:R2Gz3.1824$ok4.1...@ptah.visi.com...
[snip]
| I don't think MS wants to kill
| SMTP or POP3, they simply have a better way to do this for their own
| software on an internal network.
As does Lotus Notes. Exchange isn't trying to kill any protocols. It
works quite nicely with SMTP. But SMTP is extremely limited for
intra-office high-bandwidth networks. Why download all of your messages
to your station when bandwidth doesn't limit you from viewing them while
on the server? Not to mention that Sendmail and other SMTP-based apps
don't support shared anything! Viewing another user's schedule for the
day? Having an assistant or manager make additions/deletions/changes to
your calendar? Being able to manage a project involving multiple
people? All very nice features to boost productivity, no?
Brent Davies
| On Thu, 02 Sep 1999 22:39:58 GMT, Brent Davies
<brent...@spam-home.com>
| wrote:
| >
| >The "user-friendly" part is not the most important aspect.
|
| Now that's something I haven't heard from your camp often.
|
| >Microsoft
| >didn't have to implement TCP/IP on Windows. They did mostly because
of
| >the profit involved with higher sales.
|
| They did because TCP/IP was more popular than their proprietary
| protocols, because it had flexibility and power and was free, and
people
| were buying third-party implementations.
Well, even though I agree, I don't see how this makes MS look any worse
than Apple, Novell, or Digital, just to name a few. All had proprietary
protocols, and all those protocols failed. What's the point?
-B
Brent Davies
Chris Johnson <jinx...@sover.net> wrote in message
news:jinx6568-020...@arc1a208.bf.sover.net...
| In article <7qn0eo$2ps$3...@nnrp03.primenet.com>, Stephen Edwards
| <ja20...@primenet.com> wrote:
| >And? _So_ _what_. His point is that before Microsoft came around,
there
| >was no way for a regular user to configure a machine to connect to
the
| >internet. Microsoft, among others, have made dealing with TCP/IP
easier
| >so that more people could do it on their own. I believe that was his
| >point.
|
to
| (pitiful forerunners to) the Net back when Microsoft was selling
Typing
| Tutor.
Oh, yeah. I forgot. My grandmother was typing away on a VT while I was
in diapers. She was doint all _kinds_ of research on the Net!
</sarcasm>
If you have even a little bit of a clue, you would know what Stephen and
I have been getting at (which I think you do. I think you're using the
same "dramatic license" that Stephen likes to call upon once in a
while).
Joe Schmoe and Jane Doe were not using the Net during the time period
you spoke about. Hell, at that point Apple had just figured out what a
modem was.
I can't see a sensible person claiming that the consumer push to the
Internet would have happened as quickly if not for the consumer
usability that Windows brought to the market.
[snip]
-B
Jack Richards
J.T.Kay wrote in message <7qn6ie$b53$1...@samba.rahul.net>...
>Stephen Edwards <ja20...@primenet.com> wrote:
>>So, you're suggesting that the _only_ reason they made those modifications
>>was to screw up Sun's Java agenda?
>
>That is exactly what the Microsoft executives were saying to each
>other in email. I would think the MS executives know their intent
>quite a bit better than you do.
>
>Their email was subpoena'd and produced as evidence in the Microsoft
>anti-trust trial, which enabled the public to read it too.
>
> "Do we have a clear plan on what we want Apple to do to
> undermine Sun?" -- email from Bill Gates.
Worse than that, Gates is reported to have said Java was a "hairball."
This alone justifies anything, IMHO. It shows that he didn't want to compete
fairly, but merely to undermine a competitor's product. How awful!
Jack
Matt Curtin
"Jack Richards" <jrxxx...@worldnet.com> said:
Jack> They fear the user friendly desktop may take over
The user friendly desktop invented at Xerox PARC?
This "fear" of the "user friendly" (as if there's one definition of
this) environment "taking over" is nonexistent.
I do wish you'd stop spewing this nonsense or at least find some means
of justifying it. You never bothered to answer my previous questions
about "user friendly" environments. I'm therefore left with the
impression that these are unanswerable from your perspective.
--
Matt Curtin cmcu...@interhack.net http://www.interhack.net/people/cmcurtin/
Stephen Edwards
: In article <7qn0eo$2ps$3...@nnrp03.primenet.com>, Stephen Edwards
: <ja20...@primenet.com> wrote:
: >And? _So_ _what_. His point is that before Microsoft came around, there
: >was no way for a regular user to configure a machine to connect to the
: >internet. Microsoft, among others, have made dealing with TCP/IP easier
: >so that more people could do it on their own. I believe that was his
: >point.
: Are you kidding, Stephen? Furrfu. People were connecting machines to
*sigh* "Furrfu"... that word always makes me nostalgic... *sniff*
I actually had someone reply with "yeah, f*** you too!" when I used that
word in a reply to one of their posts... I wish I could remember the
article subject... it was quite humorous.
: (pitiful forerunners to) the Net back when Microsoft was selling Typing
: Tutor. Microsoft was _always_ _around_ and did not 'come around'. And if
: Microsoft was instrumental in bringing TCP/IP support to (surprise!) THEIR
: OWN FSCKING OPERATING SYSTEMS, well, color _me_ astonished ;P
True. All I'm doing is asserting the contention that Microsoft
(regardless whether the act was virtuous or not) has made the Internet
accessible to the layperson... the "people" that you're referring to were
usually of the geek persuasion, and therefore, enjoyed the task of
configuring their systems... most average users do not.
Also, I'm not stating that this is a "good" thing necessarily... I'm just
agreeing with Chad's contention.
: Jinx_tigr, escaping from the monastery to bite Stephen's ankle again ;)
: (aka Chris Johnson)
Joke's on you... I have a wooden leg.[0]
[0] Yes, I'm kidding. My leg is real... the rest of me is wooden. :-)
Jerry
>
> Chris Johnson <jinx...@sover.net> wrote:
>
> : In article <7qn0eo$2ps$3...@nnrp03.primenet.com>, Stephen Edwards
> : <ja20...@primenet.com> wrote:
>
> : >was no way for a regular user to configure a machine to connect to the
> : >internet. Microsoft, among others, have made dealing with TCP/IP easier
> : >so that more people could do it on their own. I believe that was his
> : >point.
>
>
> *sigh* "Furrfu"... that word always makes me nostalgic... *sniff*
>
> I actually had someone reply with "yeah, f*** you too!" when I used that
> word in a reply to one of their posts... I wish I could remember the
> article subject... it was quite humorous.
>
> : (pitiful forerunners to) the Net back when Microsoft was selling Typing
> : Tutor. Microsoft was _always_ _around_ and did not 'come around'. And if
> : Microsoft was instrumental in bringing TCP/IP support to (surprise!) THEIR
> : OWN FSCKING OPERATING SYSTEMS, well, color _me_ astonished ;P
>
> True. All I'm doing is asserting the contention that Microsoft
> (regardless whether the act was virtuous or not) has made the Internet
> accessible to the layperson... the "people" that you're referring to were
> usually of the geek persuasion, and therefore, enjoyed the task of
> configuring their systems... most average users do not.
>
> Also, I'm not stating that this is a "good" thing necessarily... I'm just
> agreeing with Chad's contention.
>
> : Jinx_tigr, escaping from the monastery to bite Stephen's ankle again ;)
> : (aka Chris Johnson)
>
> Joke's on you... I have a wooden leg.[0]
A good match for your head.....
>
> [0] Yes, I'm kidding. My leg is real... the rest of me is wooden. :-)
o b h @wasatch.com Bob Hauck
> But SMTP is extremely limited for intra-office high-bandwidth
> networks. Why download all of your messages to your station when
> bandwidth doesn't limit you from viewing them while on the server?
Minor point...you don't use SMTP for viewing mail. You use POP. If you
want to view them on the server, you use IMAP. If you're using open
standards that is.
> Not to mention that Sendmail and other SMTP-based apps don't support
> shared anything! Viewing another user's schedule for the day? Having
> an assistant or manager make additions/deletions/changes to your
> calendar?
What do these things have to do with email? Just about nothing outside
of change notification. Sendmail does email, and only email. It is not
a calendar or a day planner or a "discussion" (aka News) server.
Wouldn't it be better to adopt a more modular approach and have separate
apps that communicate using standard protocols? It isn't rocket science
to have the calendar program look up a user's email address in a common
database and send mail to him when somone schedules an appointment. You
could even build a front-end, perhaps web-based, to provide a nice
integrated UI to the whole thing.
The main reason why MS and Lotus put all the stuff into one package with
the email is to *tie you to a proprietary mail standard*. This makes it
harder for customers to change from one brand to the other. While this
may be great for MS or IBM, it doesn't exactly help the customers.
--
-| Bob Hauck
-| Wasatch Communications Group
-| http://www.wasatch.com/~bobh
Brent Davies
Bob Hauck <b o b h @ w a s a t c h . c o m> wrote in message
news:m3wvu7s...@lab.slc.codem.com...
| "Brent Davies" <brent...@spam-home.com> writes:
|
| > But SMTP is extremely limited for intra-office high-bandwidth
| > networks. Why download all of your messages to your station
when
| > bandwidth doesn't limit you from viewing them while on the
server?
|
| Minor point...you don't use SMTP for viewing mail. You use POP.
If you
| want to view them on the server, you use IMAP. If you're using
open
| standards that is.
You're right. Minor point, and my error. However, you knew what I
was getting at.
|
|
| > Not to mention that Sendmail and other SMTP-based apps don't
support
| > shared anything! Viewing another user's schedule for the day?
Having
| > an assistant or manager make additions/deletions/changes to your
| > calendar?
|
| What do these things have to do with email?
They all require an MTA. Bring these capabilities into the same app
that implements the MTA does a lot for performance and ease of
management. It also places all of them in a central message store.
Ease of management
Ease of administration
Ease of backup/disaster recovery
Ease of implementation
Increased performance.
-B
void
>
>True. All I'm doing is asserting the contention that Microsoft
>(regardless whether the act was virtuous or not) has made the Internet
>accessible to the layperson... the "people" that you're referring to were
>usually of the geek persuasion, and therefore, enjoyed the task of
>configuring their systems... most average users do not.
Assert failed. Someone (Brett I think) mentioned LanMan, but I think IBM
came up with that first. So Microsoft has still brought nothing to
TCP/IP that others haven't done first -- in a user-friendly fashion no
less. Did they come up with graphical browsers? Nope. Graphical email
clients? Nope. Instant messaging? Nope. User-friendly dialup
software? Uh-uh. All of these were pioneered by other people and
implemented by Microsoft after their utility and popularity were proven.
Microsoft's role in bringing TCP/IP to the masses is comparable to that
of Tucows or shareware.com.
void
wrote:
>
>| > Not to mention that Sendmail and other SMTP-based apps don't
>support
>| > shared anything! Viewing another user's schedule for the day?
>Having
>| > an assistant or manager make additions/deletions/changes to your
>| > calendar?
>|
>| What do these things have to do with email?
>
>They all require an MTA. Bring these capabilities into the same app
>that implements the MTA does a lot for performance and ease of
>management. It also places all of them in a central message store.
>
>Ease of management
>Ease of administration
>Ease of backup/disaster recovery
>Ease of implementation
>Increased performance.
Not at all. Sendmail is one program, and dog-slow compared to the fully
modular postfix or qmail. Ease of implementation? Exchange makes even
MCSEs complain. Ease of recovery? Again, Exchange has generated some
of the worst war stories I've ever heard. It's much easier to recover
flat files than opaque huge databases.
void
wrote:
>
>Well, even though I agree, I don't see how this makes MS look any worse
>than Apple, Novell, or Digital, just to name a few. All had proprietary
>protocols, and all those protocols failed. What's the point?
What's up with the attention span around here? I was refuting Chad's
baseless assertion that Microsoft brought TCP/IP to the masses like
Prometheus making off with BBN's fire.
void
>
>Minor point...you don't use SMTP for viewing mail. You use POP.
*I* use SMTP. I have an SDSL connection, my mail server is in my
bedroom.
Brent Davies
|
| Not at all. Sendmail is one program, and dog-slow compared to the
fully
| modular postfix or qmail. Ease of implementation? Exchange makes
even
| MCSEs complain.
I'm an MCSE and in my 2+ years of Exchange implementations, I
haven't had cause to complain about it. Who might you be talking
about? Are these recent graduates of the multitudes of "Boot Camps"
around town?
| Ease of recovery? Again, Exchange has generated some
| of the worst war stories I've ever heard.
I've had my fair share of recoveries (mostly due to bad hardware, or
one of your "MCSEs" mucking up the system), but by following the
many disaster recovery whitepapers available, it isn't difficult at
all.
| It's much easier to recover
| flat files than opaque huge databases.
It may be faster, but I haven't found it to be any easier.
-B
Brent Davies
| On Fri, 03 Sep 1999 06:00:59 GMT, Brent Davies
<brent...@spam-home.com>
| wrote:
| >
| >Well, even though I agree, I don't see how this makes MS look any
worse
| >than Apple, Novell, or Digital, just to name a few. All had
proprietary
| >protocols, and all those protocols failed. What's the point?
|
| What's up with the attention span around here? I was refuting
Chad's
| baseless assertion that Microsoft brought TCP/IP to the masses
like
| Prometheus making off with BBN's fire.
I was replying to only one of your claims. You are famous for that
same behavior around here. Are you telling me that you can't
identify that behavior when someone else does it?
-B
Brent Davies
|
| Microsoft's role in bringing TCP/IP to the masses is comparable to
that
| of Tucows or shareware.com.
Who would have done it otherwise, and as quickly? Apple? They
_still_ don't have their IP stack right, IMO.
-B
Chad Mulligan
void wrote in message ...
>>(regardless whether the act was virtuous or not) has made the Internet
>>accessible to the layperson... the "people" that you're referring to were
>>usually of the geek persuasion, and therefore, enjoyed the task of
>>configuring their systems... most average users do not.
>
>Assert failed. Someone (Brett I think) mentioned LanMan, but I think IBM
>came up with that first. So Microsoft has still brought nothing to
>TCP/IP that others haven't done first -- in a user-friendly fashion no
>less. Did they come up with graphical browsers? Nope. Graphical email
>clients? Nope. Instant messaging? Nope. User-friendly dialup
>software? Uh-uh. All of these were pioneered by other people and
>implemented by Microsoft after their utility and popularity were proven.
Wrong answer. Lan Manager, while first released by IBM, followed by AT&T,
3COM, Banyan and a host of others was developed by Microsoft under contract
by IBM. This was as I stated an OEM only release as such noone knew MS was
involved. Source (other than my rememberance of the time): "Mastering
Windows NT Server, Sixth Edition," Mark Minasi, 1998.
>
>Microsoft's role in bringing TCP/IP to the masses is comparable to that
>of Tucows or shareware.com.
>
With the exception that MS provided the platform for the TCP/IP to run on,
and delivered that to the masses.
Erik Funkenbusch
news:m3wvu7s...@lab.slc.codem.com...
> apps that communicate using standard protocols? It isn't rocket science
> to have the calendar program look up a user's email address in a common
> database and send mail to him when somone schedules an appointment. You
> could even build a front-end, perhaps web-based, to provide a nice
> integrated UI to the whole thing.
Most people do not like running seperate programs for everything. They'd
rather have one program running all day with one interface rather than 5 or
10 programs that all give different interfaces and require different
training.
Not to mention that it's hard for these programs to communicate. For
instance, Exchange allows you to send meeting notices to people in email and
have it automatically added to their calendar if they choose to accept it.
This would require the user to do this manually in a multi-app environment.
Then when you consider the side effects (the meeting time was changed or was
canceled, everyone get's updated automatically) this is pure productivity
(to use the word loosely, since meetings are not all that productive..
however it let's the user do more productive things instead of managing
meeting schedules).
> The main reason why MS and Lotus put all the stuff into one package with
> the email is to *tie you to a proprietary mail standard*. This makes it
> harder for customers to change from one brand to the other. While this
> may be great for MS or IBM, it doesn't exactly help the customers.
While I don't deny that this is a nice side-effect for these companies, it's
simply impossible to do the things that Exchange or Notes do with only
public standard protocols.
Matt Curtin
"Brent Davies" <brentdavi...@home.com> said:
Brent> Who would have done it otherwise, and as quickly? Apple? They
Brent> _still_ don't have their IP stack right, IMO.
Lest you believe the sort of cruft that Microsoft spews about their
role in bringing IP to the masses, I feel it necessary to engage in
the office of reminding you that whilst the Trade Press was quoting
Bill Gates saying "the Internet is the frothiest thing I've ever
seen", products like Trumpet Winsock and Chameleon were providing IP
services to Windoze users.
The real revolution is the Internet, and it's happened completely
without Microsoft's help. One could argue that its value has become
diminished from the point at which they jumped aboard the Internet
bandwagon.
Their tools have produced zillions of pages of uncomplaint HTML
encoded in proprietary character sets. Their welding of the browser
to their operating system and giving it away completely free on other
platforms yanked Netscape's revenue stream out of its hands. Their
pressure on Netscape led to the Browser Wars, the result of which is
unwieldy browsers which, even with gargantuan memory and processor
requirements satisfied, aren't stable. Their lack of attention to
security and privacy requirements have led to millions of people being
exposed needlessly to all sorts of risks.
Given this sort of track record, I find it somewhat amazing that
people would dare advertise the MCSE "credentials". The producers of
such excrement are not the sort of "technology professionals" with
whom I'd care to associate myself.
Stephen Edwards
: >>>>> On Fri, 03 Sep 1999 19:14:52 GMT,
: "Brent Davies" <brentdavi...@home.com> said:
: Brent> Who would have done it otherwise, and as quickly? Apple? They
: Brent> _still_ don't have their IP stack right, IMO.
: Lest you believe the sort of cruft that Microsoft spews about their
: role in bringing IP to the masses, I feel it necessary to engage in
: the office of reminding you that whilst the Trade Press was quoting
: Bill Gates saying "the Internet is the frothiest thing I've ever
: seen", products like Trumpet Winsock and Chameleon were providing IP
: services to Windoze users.
Right... products which ran on Microsoft operating systems. Your point?
: Their tools have produced zillions of pages of uncomplaint HTML
: encoded in proprietary character sets. Their welding of the browser
Uh... huh? Exactly what non-compliant HTML tags are you referring to?
: requirements satisfied, aren't stable. Their lack of attention to
: security and privacy requirements have led to millions of people being
: exposed needlessly to all sorts of risks.
Most security breaches of Microsoft software have been directly related to
negligent administrators, in most of the stories I've read about.
: Given this sort of track record, I find it somewhat amazing that
: people would dare advertise the MCSE "credentials". The producers of
: such excrement are not the sort of "technology professionals" with
: whom I'd care to associate myself.
Hmmm... same FUD, different addy.
Chad Mulligan
Bob Hauck > wrote in message ...
>"Brent Davies" <brent...@spam-home.com> writes:
>
<trimmed>
>What do these things have to do with email? Just about nothing outside
>of change notification. Sendmail does email, and only email. It is not
>a calendar or a day planner or a "discussion" (aka News) server.
>
Groupware is groupware, the calendar will mail meeting requests and place
them on the receiver's and sender's calendars. Updating the calendar when
someone accepts an invite, without the need for the user to read an email
and remember to update his calendar. Ease of use thing.
>Wouldn't it be better to adopt a more modular approach and have separate
>apps that communicate using standard protocols? It isn't rocket science
>to have the calendar program look up a user's email address in a common
>database and send mail to him when somone schedules an appointment. You
>could even build a front-end, perhaps web-based, to provide a nice
>integrated UI to the whole thing.
>
BTW, MS and Exchange, and Active Directory, use LDAP an established standard
protocol. That's why Outlook can work like an exchange client without
needing exchange. Let's see Notes do that.
>The main reason why MS and Lotus put all the stuff into one package with
>the email is to *tie you to a proprietary mail standard*. This makes it
>harder for customers to change from one brand to the other. While this
>may be great for MS or IBM, it doesn't exactly help the customers.
>
Chad Mulligan
void wrote in message ...
>wrote:
>>
>>Well, even though I agree, I don't see how this makes MS look any worse
>>than Apple, Novell, or Digital, just to name a few. All had proprietary
>>protocols, and all those protocols failed. What's the point?
>
>What's up with the attention span around here? I was refuting Chad's
>baseless assertion that Microsoft brought TCP/IP to the masses like
>Prometheus making off with BBN's fire.
>
I didn't make that assertion, I think Stephen did. Does your reader thread
articles?
YeahRight
>But this situation will not last for ever. The groundswell of support
>behind Linux, behind Java, and soon behind StarOffice will be a tidal
>wave that Redmond doesn't understand, and can do little except run from.
>Look for a steep drop in MSFT stock price in the coming months.
AHAHAHAHAHAHA so how much have you shorted the stock?? put your
money where your mouth is moron
Matt Curtin
Brent> Who would have done it otherwise, and as quickly? Apple?
Matt> products like Trumpet Winsock and Chameleon were providing IP
Matt> services to Windoze users.
Stephen> Right... products which ran on Microsoft operating systems.
Stephen> Your point?
Brent asked who else would have brought the Internet to the masses.
Microsoft had *nothing* to do with the Internet or any of its
technology. They were trying to get people to buy into the MSN pipe
dream. Microsoft "embracing" the Internet has had nothing to do with
its success or the ability of Windoze users to get thusly connected.
I noticed that you're not using a Microsoft news mangler. (The first
clue was that I didn't have to tell Gnus to reformat the article so
that it would be readable. Your headers seem to verify my
observation.) Pray tell, why might that be?
Stephen> Uh... huh? Exactly what non-compliant HTML tags are you
Stephen> referring to?
Give me a break. You can't possibly be the least bit Internet
literate and not have some understanding of Microsoft tools' complete
ineptitude.
Let's see what kind of cruft Word produces. Here's an example:
http://www.jamesline.com/output/content/cainfo/leukhome/leukfaq.htm
Let's run that sucker through a validator, shall we?
Surprise! It fails miserably. There's no excuse for that crap. I
won't even get started on the "bgsound" attribute ...
Stephen> Most security breaches of Microsoft software have been
Stephen> directly related to negligent administrators, in most of the
Stephen> stories I've read about.
Sounds like you need to do more reading. Here's a start.
http://www.cryptonym.com/hottopics/msft-nsa.html
http://www.news.com/News/Item/0,4,41287,00.html
http://catless.ncl.ac.uk/Risks/20.32.html#subj3
http://catless.ncl.ac.uk/Risks/20.02.html#subj8.1
http://catless.ncl.ac.uk/Risks/19.58.html#subj5.1
Enjoy.
Nick Manka
"Brent Davies" <brentdavi...@home.com> writes:
>| Microsoft's role in bringing TCP/IP to the masses is comparable to
> that
>| of Tucows or shareware.com.
>
> Who would have done it otherwise, and as quickly? Apple? They
> _still_ don't have their IP stack right, IMO.
Neither does Microsoft, and who's got the bigger budget?
And who has realized that their crufty old software was worse than
what was available in BSD and switched?
Who brought TCP/IP to the masses? -- BSD did. Who wrote the TCP/IP
stack everyone associates with early Windows TCP/IP? -- Trumpet and
Chameleon.
Funny, none of the above are everyone's favorite software tyranny.
--
The sad thing about open source is that it was easier to convince
companies to give their code away than to give it to their customers.
Brent Davies
Chad Mulligan <cmul...@hipcrime.vocab.org> wrote in message
news:ysWz3.640$i03....@typhoon-sf.snfc21.pbi.net...
|
[snip]
| >
|
| BTW, MS and Exchange, and Active Directory, use LDAP an
established standard
| protocol. That's why Outlook can work like an exchange client
without
| needing exchange. Let's see Notes do that.
You want established standards? The Exchange Directory service
completely conforms with X.500; the internal Exchange Server message
delivery protocol is X.400 (try to delete your mailbox's X.400
address. All hell breaks loose!); Internet Mail Connector
completely conforms to SMTP/POP3 standards; etc.
How much more "standardized" do you want it to be? ("you" used in
general. Chad, I know that you're in agreement).
[snip]
-B
o b h @wasatch.com Bob Hauck
> Bob Hauck <b o b h @ w a s a t c h . c o m> wrote in message
>> Wouldn't it be better to adopt a more modular approach and have
>> separate apps that communicate using standard protocols?
> Most people do not like running seperate programs for everything.
> They'd rather have one program running all day with one interface
> rather than 5 or 10 programs that all give different interfaces and
> require different training.
First point is that you should rephrase that to "most people trained in
the Windows way don't like running separate programs". Secondly, why
couldn't the front-end be browser-based, and thus "one program"?
> Not to mention that it's hard for these programs to communicate. For
> instance, Exchange allows you to send meeting notices to people in
> email and have it automatically added to their calendar if they choose
> to accept it.
This is "hard"? I don't think so...
> This would require the user to do this manually in a multi-app
> environment.
I don't see why. Is there some reason the calendar program can't
receive mail? The user can just hit "reply". If there is a web-based
integrated frontend, it's not even that hard.
>> The main reason why MS and Lotus put all the stuff into one package
>> with the email is to *tie you to a proprietary mail standard*.
> While I don't deny that this is a nice side-effect for these
> companies, it's simply impossible to do the things that Exchange or
> Notes do with only public standard protocols.
I disagree. Maybe it would be harder in some respects, but not even
close to impossible.
Brent Davies
Matt Curtin <cmcu...@interhack.net> wrote in message
news:xlx1zcf...@gold.cis.ohio-state.edu...
| >>>>> On 3 Sep 1999 20:44:44 GMT, Stephen Edwards
<ja20...@primenet.com> said:
|
| Brent> Who would have done it otherwise, and as quickly? Apple?
|
| Matt> products like Trumpet Winsock and Chameleon were providing
IP
| Matt> services to Windoze users.
|
| Stephen> Right... products which ran on Microsoft operating
systems.
| Stephen> Your point?
|
| Brent asked who else would have brought the Internet to the
masses.
You misrepresenting SOB!!! That was not my complete statement. I
KNOW DAMN WELL that someone else would have done it. My question
was, could that company have done it AS QUICKLY AS Microsoft did?
Credibility = N - 1
| Microsoft had *nothing* to do with the Internet or any of its
| technology. They were trying to get people to buy into the MSN
pipe
| dream. Microsoft "embracing" the Internet has had nothing to do
with
| its success or the ability of Windoze users to get thusly
connected.
No, of course not. Just the same way that Linux would be just as
popular if it cost $300 per license. </sarcasm> Wake up! How far
do you thing laypeople would have gotten if they had to buy the IP
stack *in addition to* Windows?
MS had the install-base, so when they added the TCP/IP stack, they
became the single largest distributor of TCP/IP on the planet!
[snip, *yawn*]
-B
Brent Davies
Nick Manka <ni...@faust.sec.baileynm.com> wrote in message
news:7qpgq7$m...@web.nmti.com...
| In article <M6Vz3.19081$Rn.1...@news.rdc2.occa.home.com>,
| "Brent Davies" <brentdavi...@home.com> writes:
|
| >| Microsoft's role in bringing TCP/IP to the masses is comparable
to
| > that
| >| of Tucows or shareware.com.
| >
| > Who would have done it otherwise, and as quickly? Apple? They
| > _still_ don't have their IP stack right, IMO.
|
| Neither does Microsoft, and who's got the bigger budget?
Cite some examples of MS's so-called "crippled" IP stack. I have no
issues with it. I've seen it run in heterogeneous environments
without any issues whatsoever.
|
| Who brought TCP/IP to the masses? -- BSD did. Who wrote the TCP/IP
| stack everyone associates with early Windows TCP/IP? -- Trumpet
and
| Chameleon.
So you're telling me that BSD, Trumpet, and Chameleon enjoy an
install-base equal to at least a majority of current Internet users?
Interesting...
Sources, please.
-B
o b h @wasatch.com Bob Hauck
> Bob Hauck > wrote in message ...
>> What do these things have to do with email? Just about nothing
>> outside of change notification. Sendmail does email, and only email.
>> It is not a calendar or a day planner or a "discussion" (aka News)
>> server.
> Groupware is groupware, the calendar will mail meeting requests and
> place them on the receiver's and sender's calendars.
The calendar server sends email to the user. The email says "reply to
accept this invite", user hits reply, done. Other situations can be
handled similarly, perhaps supplemented with embedded links in the
email ("click here to see your calendar" and the like).
> BTW, MS and Exchange, and Active Directory, use LDAP an established
> standard protocol. That's why Outlook can work like an exchange
> client without needing exchange. Let's see Notes do that.
I don't like Notes either. Or CCMail or PROFS. In my view, proprietary
email non-standards represent the worst sort of vendorlock. They are
inherently bad for the customer, in all forms, from all vendors.
o b h @wasatch.com Bob Hauck
> They all require an MTA. Bring these capabilities into the same app
> that implements the MTA does a lot for performance and ease of
> management. It also places all of them in a central message store.
I don't understand how putting everything in one big pile would help
performance significantly as compared to having several services that
communicate with each other. It seems to me that the opposite is true.
Making each major component a separate process makes it easier to scale
the system by moving components onto separate machines.
> Ease of management
> Ease of administration
> Ease of backup/disaster recovery
> Ease of implementation
> Increased performance.
I disagree. By putting MTA, Calendar, Database, News, and whatever else
all into one mega-app, you now have a mega-app that is much more
susceptible to problems, not less. Now you have one database to fail,
you have created scalibility challenges, and you have locked your data
into a proprietary and probably not-well-documented format.
Brent Davies
Bob Hauck <b o b h @ w a s a t c h . c o m> wrote in message
| "Brent Davies" <brentdavi...@home.com> writes:
|
| > They all require an MTA. Bring these capabilities into the same
app
| > that implements the MTA does a lot for performance and ease of
| > management. It also places all of them in a central message
store.
|
| I don't understand how putting everything in one big pile would
help
| performance significantly as compared to having several services
that
| communicate with each other. It seems to me that the opposite is
true.
| Making each major component a separate process makes it easier to
scale
| the system by moving components onto separate machines.
Every major mail messaging system runs a queueing process.
Activities are not immediately written to the database. The only
options you have when running seperate processes is to use a
separate database engine that much field requests from each process,
or each process has its own database to maintain. The latter would
be a fault tolerance nightmare. The former is do-able, but much
harder for the seperate devices to get the kind of access they need
to the database engine.
With a monolithic process, the subprocesses can be queued,
transactions written to the logs, and they can move about their
merry business. Besides that, these subprocesses work in
conjunction with each other continuously. The interoperability
benefits from increased performance, when compared with implementing
that same interoperability utilizing many different seperate
processes.
Did you ever wonder why these "seperate processes" have not been
implemented in the same fashion with Sendmail? At least I've never
seen it done. Maybe I've been too long under my rock. =}
|
|
| > Ease of management
| > Ease of administration
| > Ease of backup/disaster recovery
| > Ease of implementation
| > Increased performance.
|
| I disagree. By putting MTA, Calendar, Database, News, and whatever
else
| all into one mega-app, you now have a mega-app that is much more
| susceptible to problems, not less.
Buggy code can be a problem, but IME I've noticed that these
problems in Exchange haven't caused the whole thing to come tumbling
down very often. It is modular enough that many failures only
affect a particular function and not others. I once had a failure
in the MTA itself, but the IMC continued queueing inbound Internet
mail in the transaction logs.
| Now you have one database to fail,
With the level of interoperability that we are talking about, the
failure of any one database under your proposed model would pretty
much end the groupware capabilities, anyway.
| you have created scalibility challenges,
How so? Have you seen the scalability that Exchange 5.5 is capable
of? If so, you probably wouldn't be saying that. There is also the
framework: Organization/Site/Server This provides a high-level of
scalability within the internetwork.
| and you have locked your data
| into a proprietary and probably not-well-documented format.
This is a point to you. The format is proprietary, but I'm not sure
if it's documented or not. I do know that many other groupware
product support connectors to Exchange that will migrate all of the
data over to the new product. So it's not like you've got to start
over if you want to move to another system. Migrations happen ever
day.
-B
Brent Davies
Bob Hauck <b o b h @ w a s a t c h . c o m> wrote in message
| "Chad Mulligan" <cmul...@hipcrime.vocab.org> writes:
|
| > Bob Hauck > wrote in message ...
|
| >> What do these things have to do with email? Just about nothing
| >> outside of change notification. Sendmail does email, and only
email.
| >> It is not a calendar or a day planner or a "discussion" (aka
News)
| >> server.
|
| > Groupware is groupware, the calendar will mail meeting requests
and
| > place them on the receiver's and sender's calendars.
|
| The calendar server sends email to the user. The email says
"reply to
| accept this invite", user hits reply, done. Other situations can
be
| handled similarly, perhaps supplemented with embedded links in the
| email ("click here to see your calendar" and the like).
Do you realize how many steps you are adding to the process by
seperating these functions into their own "daemons"? Then imagine a
10,000 node network where 100,000 messages calendar/meeting
requests, etc. pass through the system every day. These added steps
will severely impact performance.
|
|
| > BTW, MS and Exchange, and Active Directory, use LDAP an
established
| > standard protocol. That's why Outlook can work like an exchange
| > client without needing exchange. Let's see Notes do that.
|
| I don't like Notes either. Or CCMail or PROFS. In my view,
proprietary
| email non-standards represent the worst sort of vendorlock. They
are
| inherently bad for the customer, in all forms, from all vendors.
So what do you use? Does Sendmail actually have groupware companion
apps now?
-B
Matt Curtin
"Brent Davies" <brentdavi...@home.com> said:
Brent> You misrepresenting SOB!!!
You are going to write something like this and suggest that my
credibility is to be decremented? Interesting.
If you expect to win an argument with me, you'd better do your
homework first and do yourself the favor of spending all of your
energy on your argument. Namecalling and excessive punctuation aren't
going to do much for your chances for success.
Brent> I KNOW DAMN WELL that someone else would have done it. My
Brent> question was, could that company have done it AS QUICKLY AS
Brent> Microsoft did?
My point, Brent, is that it was *already being done* at the point that
Microsoft jumped into the game. Can't get much quicker than that.
Brent> How far do you thing laypeople would have gotten if they had to
Brent> buy the IP stack *in addition to* Windows?
Thousands -- maybe millions -- of people seemed to do pretty well. Of
course, if they wanted an operating system with IP built in, such were
*already available*. Indeed, the Internet already existed for
something like two decades at this point. Mosaic, Netscape, and a
number of other browsers were already out there, running on a LOT of
different kinds of computers, getting a lot of people in touch with
one another.
Giving Microsoft credit for the proliferation of IP (as if every
installation of Windows '95 gave IP to someone who didn't already have
it) is, well, a misrepresentation. IP was already everywhere by the
time Windows '95 came on to the scene.
Matt Curtin
"Brent Davies" <brentdavi...@home.com> said:
Brent> Cite some examples of MS's so-called "crippled" IP stack.
Three examples are: winnuke, bonk, and teardrop. Another excellent
example is the inability of NT to be used as a reliable sniffer.
Observe the inability of NT to serve as a host platform for the
Network Flight Recorder (http://www.nfr.net/).
Brent> So you're telling me that BSD, Trumpet, and Chameleon enjoy an
Brent> install-base equal to at least a majority of current Internet
Brent> users? Interesting...
Your argument is fundamentally flawed because it assumes that the
current Internet users are using the same product that brought IP to
them in the first place. That's both ridiculous and irrelevant.
Erik Funkenbusch
> "Erik Funkenbusch" <er...@visi.com> writes:
> > Most people do not like running seperate programs for everything.
> > They'd rather have one program running all day with one interface
> > rather than 5 or 10 programs that all give different interfaces and
> > require different training.
>
> First point is that you should rephrase that to "most people trained in
> the Windows way don't like running separate programs". Secondly, why
> couldn't the front-end be browser-based, and thus "one program"?
No, most people. Period. Most unix based systems that "most people" use
are turnkey based with a single interface.
It's only extreme power users with hefty IQ's and the ability to remember
different interfaces that enjoy the challenge of keeping everything straight
in their minds. Most people just want to get their job done.
> > Not to mention that it's hard for these programs to communicate. For
> > instance, Exchange allows you to send meeting notices to people in
> > email and have it automatically added to their calendar if they choose
> > to accept it.
>
> This is "hard"? I don't think so...
Well, clearly the vast quantity of applications which do this using standard
protocols under unix backs you up. Right?
> > This would require the user to do this manually in a multi-app
> > environment.
>
> I don't see why. Is there some reason the calendar program can't
> receive mail? The user can just hit "reply". If there is a web-based
> integrated frontend, it's not even that hard.
So how exactly does the calendar program pick out mail messages which are
destined for it, and how does the email application not retrieve messages
that are for other applications? Even then, you have to format the messages
in some way to communicate the information or the application won't have a
clue as to how to parse it. That requires a standard protocol of some kind
which doesn't exist (otherwise someone sending a meeting request in pine
wouldn't be able to figure it out under elm.)
Then on top of everything else. What about things like return reciepts (so
that the originator knows when the message was read)? What about a host of
other features that Exchange or Notes provide that no standard protocol
exists for?
> > While I don't deny that this is a nice side-effect for these
> > companies, it's simply impossible to do the things that Exchange or
> > Notes do with only public standard protocols.
>
> I disagree. Maybe it would be harder in some respects, but not even
> close to impossible.
Note the requisite "standard". If the standard doesn't exist, it's
impossible to use it.
Jim Richardson
Brent Davies, in the persona of <brentdavi...@home.com>,
brought forth the following words...:
>
>Nick Manka <ni...@faust.sec.baileynm.com> wrote in message
>news:7qpnnu$m...@web.nmti.com...
>|
>| [ This thread has little relevance on comp.lang.java.advocacy]
>|
>[snip]
>|
>| Uh, the old FINWAIT problem where a Microsoft computer wouldn't
>| properly close TCP connections.
>|
>| The various fragmentation attacks, many of which affected
>Microsoft's
>| TCP/IP stacks but not others.
>|
>| Try the various "security bug tracking" sites on the net. They
>have
>| lists of dozens of Denial of Service attacks against MS TCP/IP
>that
>| I'm probably not even aware of, because I'm not silly enough to
>| leave Windows machines on the open internet and thus don't have to
>| care.
>
>I am currently under the impression that a great many of the
>problems you are describing here were problems within the TCP/IP
>spec itself. I have been lead to believe that most of the DoS
>problems and fragmentation problems affected most, if not all, of
>the machines running the protocol.
Just on example, the teardrop attack, it affected linux (and *bsd etc) and
windows, took linux what, 24hrs for a patch to fix it? and windows was, well
severaly weeks before M$ would admit that it even affected windows. (if
memory serves, whilst M$ was denying a vulnerability anyone with a compiler
could demonstrate, their servers went off line...)
>
>I currently have little time to run that kind of a research project,
>so I'll have to ask that this issue be tabled, at the least, for the
>time being. I don't have any information to refute what you've
>said, and I won't likely have time to gather such information in the
>near future.
>
>i
Why should it be tabled? you may be busy, but there are others actively
checking and finding the info now. In todays info saturated world, you
can't keep up to date on everything, but that's no reason for the
discussion to stop. Unless I am misunderstanding you?
>|
>| Then there is the inability of a single-processor WNT machine to
>| saturate a 100mbit network link. Rather sad. You have to buy a
>| N-way Xeon with special "server network cards" that offload most
>| of the IP processing from NT just to get half a gigabit, whilest
>| a single processor BSD machine can pull 700 mbit over gigabit
>| ethernet with the initial test release of it's driver. (and since
>| you'll whine about citing, I suggest you go look at all the
>"Server
>| ethernet cards," out there, which are interestingly enough only
>| useful on NT. Be sure to ask their vendors what they do. If you
>| want the FreeBSD numbers, they have a mailing list search web
>page.)
>
>Interesting that you think substantiating statistics or comments
>like those you made above is such an unnecessary chore. Are you
>someone that I should believe above all other things? Perhaps you
>ought to introduce yourself before we talk.
>
perhaps you could verify them yourself? after all, if you don't believe
that what he cites is true, why would you believe him if he said he'd
verified it without first checking out the info yourself?
--
Jim Richardson
Anarchist, pagan and proud of it
WWW.eskimo.com/~warlock
Linux, because life's too short for a buggy OS.
Nick Manka
[ This thread has little relevance on comp.lang.java.advocacy]
In article <GvYz3.19255$Rn.1...@news.rdc2.occa.home.com>,
"Brent Davies" <brentdavi...@home.com> writes:
>| >| Microsoft's role in bringing TCP/IP to the masses is comparable
> to
>| > that
>| >| of Tucows or shareware.com.
>| >
>| > Who would have done it otherwise, and as quickly? Apple? They
>| > _still_ don't have their IP stack right, IMO.
>|
>| Neither does Microsoft, and who's got the bigger budget?
>
> Cite some examples of MS's so-called "crippled" IP stack. I have no
> issues with it. I've seen it run in heterogeneous environments
> without any issues whatsoever.
Uh, the old FINWAIT problem where a Microsoft computer wouldn't
properly close TCP connections.
The various fragmentation attacks, many of which affected Microsoft's
TCP/IP stacks but not others.
Try the various "security bug tracking" sites on the net. They have
lists of dozens of Denial of Service attacks against MS TCP/IP that
I'm probably not even aware of, because I'm not silly enough to
leave Windows machines on the open internet and thus don't have to
care.
Then there is the inability of a single-processor WNT machine to
saturate a 100mbit network link. Rather sad. You have to buy a
N-way Xeon with special "server network cards" that offload most
of the IP processing from NT just to get half a gigabit, whilest
a single processor BSD machine can pull 700 mbit over gigabit
ethernet with the initial test release of it's driver. (and since
you'll whine about citing, I suggest you go look at all the "Server
ethernet cards," out there, which are interestingly enough only
useful on NT. Be sure to ask their vendors what they do. If you
want the FreeBSD numbers, they have a mailing list search web page.)
>| Who brought TCP/IP to the masses? -- BSD did. Who wrote the TCP/IP
>| stack everyone associates with early Windows TCP/IP? -- Trumpet
> and
>| Chameleon.
>
> So you're telling me that BSD, Trumpet, and Chameleon enjoy an
> install-base equal to at least a majority of current Internet users?
> Interesting...
Funny how you completely misunderstand the issue.
It isn't about who made the OS the stuff is installed on, it's about
who made it possible for the stuff to get there.
BSD contained the freely available reference implementation of
IPv4. Chameleon and espescially Trumpet were the companies that
actually shipped working IP suites for Windows, and who let the
Windows-masses get on the net in the first place. There wouldn't
have been a demand for Windows TCP/IP if Trumpet and Chameleon
hadn't created it. Microsoft was set up to use MSN to duke it out
with AOL and Compuserve for dialins into proprietary networks, not
to deal with the internet. Until, of course, all their users suddenly
said "Hey, we like the open Internet." And then it still took them
a year or so to notice.
The BSD TCP/IP code, or code based on it, is present on every single
Windows and most every Unix Internet-attached computer in the world.
That's a pretty big "install base," and easily musters most current
internet users. Chameleon and Trumpet had 99% of the Windows TCP/IP
install base until Windows 95, and still have all of the Windows
TCP/IP install base for everyone who has not yet upgraded past
Windows 3.11 and a substantial amount of the people still running
earlier versions of Windows 95 who kept using them instead of
Microsoft's implementation.
> Sources, please.
Try the 4.x BSD sources available at ftp.cdrom.com
Brent Davies
| >>>>> On Fri, 03 Sep 1999 23:03:28 GMT,
| "Brent Davies" <brentdavi...@home.com> said:
|
| Brent> You misrepresenting SOB!!!
|
| You are going to write something like this and suggest that my
| credibility is to be decremented? Interesting.
Everyone is entitled to a little angre when they are intentionally
misrepresented. Excuse me for being human!
|
| If you expect to win an argument with me, you'd better do your
| homework first and do yourself the favor of spending all of your
| energy on your argument. Namecalling and excessive punctuation
aren't
| going to do much for your chances for success.
|
| Brent> I KNOW DAMN WELL that someone else would have done it. My
| Brent> question was, could that company have done it AS QUICKLY AS
| Brent> Microsoft did?
|
| My point, Brent, is that it was *already being done* at the point
that
| Microsoft jumped into the game. Can't get much quicker than that.
Mac also had a GUI OS before MS came around. So? Who got it on to
millions of desktops? I'm not claiming that MS is some innovative
genious. I _know_ that others were doing it. I _know_ that the
capabilities were already there. Let's not mention the fact that
some of these companies were *selling free software* (TCP/IP). I am
simply stating that MS succeeded in bringing these capabilities to
the masses, where the other companies did not succeed in as wide a
distribution. Is that so wrong a thing to say? Is that heresy in
your book? I'm sorry that I believe MS may have done a couple of
good things during its existance, but I do.
|
| Brent> How far do you thing laypeople would have gotten if they
had to
| Brent> buy the IP stack *in addition to* Windows?
|
| Thousands -- maybe millions -- of people seemed to do pretty well.
Of
| course, if they wanted an operating system with IP built in, such
were
| *already available*.
Such that they could actually by used by laypeople? I love this
argument. Yeah, you really screwed up by not starting a
Unix-to-the-home-user company back in the early '80s. That would
have gone over like a turd in the punchbowl.
| Indeed, the Internet already existed for
| something like two decades at this point. Mosaic, Netscape, and a
| number of other browsers were already out there, running on a LOT
of
| different kinds of computers, getting a lot of people in touch
with
| one another.
Still very difficult to *get* *connected*. Browsers have never been
all that difficult to use. Actually *connecting* your operating
system to the Internet at that point in time was _well beyond the
capabilities of the layperson_!
|
| Giving Microsoft credit for the proliferation of IP (as if every
| installation of Windows '95 gave IP to someone who didn't already
have
| it) is, well, a misrepresentation. IP was already everywhere by
the
| time Windows '95 came on to the scene.
This is interesting. You must be talking about prefessional
computers that were not in the home environment. Apple had an
extremely shoddy IP implementation at the time (still is IMO) but
that was it! If IP was everywhere, how do you explain the sudden
explosion of Internet subscribers since Win95's debut? It must be
because all of those home-user Unix workstations were just sitting
in the person's closet.
Is it such a blow to your ego to admit that the ease of use of
Windows 95 played a tremendous role in the addoption of the internet
by hundereds of millions of people?
-B
Brent Davies
| >>>>> On Fri, 03 Sep 1999 23:06:14 GMT,
| "Brent Davies" <brentdavi...@home.com> said:
|
| Brent> Cite some examples of MS's so-called "crippled" IP stack.
|
| Three examples are: winnuke, bonk, and teardrop. Another
excellent
| example is the inability of NT to be used as a reliable sniffer.
| Observe the inability of NT to serve as a host platform for the
| Network Flight Recorder (http://www.nfr.net/).
All problems that have been fixed. And don't try and tell me that
the main TCP/IP code hasn't needed some fixing here and there.
|
| Brent> So you're telling me that BSD, Trumpet, and Chameleon enjoy
an
| Brent> install-base equal to at least a majority of current
Internet
| Brent> users? Interesting...
|
| Your argument is fundamentally flawed because it assumes that the
| current Internet users are using the same product that brought IP
to
| them in the first place. That's both ridiculous and irrelevant.
Go read my reply to Nick. I've been beat up over making a comment
that is subjective opinion. There is no proof one way or the other.
For that reason, agree with you. This is totally ridiculous.
-B
Brent Davies
|
|
| Uh, the old FINWAIT problem where a Microsoft computer wouldn't
| properly close TCP connections.
|
| The various fragmentation attacks, many of which affected
Microsoft's
| TCP/IP stacks but not others.
|
| Try the various "security bug tracking" sites on the net. They
have
| lists of dozens of Denial of Service attacks against MS TCP/IP
that
| I'm probably not even aware of, because I'm not silly enough to
| leave Windows machines on the open internet and thus don't have to
| care.
I am currently under the impression that a great many of the
problems you are describing here were problems within the TCP/IP
spec itself. I have been lead to believe that most of the DoS
problems and fragmentation problems affected most, if not all, of
the machines running the protocol.
I currently have little time to run that kind of a research project,
so I'll have to ask that this issue be tabled, at the least, for the
time being. I don't have any information to refute what you've
said, and I won't likely have time to gather such information in the
near future.
|
| Then there is the inability of a single-processor WNT machine to
| saturate a 100mbit network link. Rather sad. You have to buy a
| N-way Xeon with special "server network cards" that offload most
| of the IP processing from NT just to get half a gigabit, whilest
| a single processor BSD machine can pull 700 mbit over gigabit
| ethernet with the initial test release of it's driver. (and since
| you'll whine about citing, I suggest you go look at all the
"Server
| ethernet cards," out there, which are interestingly enough only
| useful on NT. Be sure to ask their vendors what they do. If you
| want the FreeBSD numbers, they have a mailing list search web
page.)
Interesting that you think substantiating statistics or comments
like those you made above is such an unnecessary chore. Are you
someone that I should believe above all other things? Perhaps you
ought to introduce yourself before we talk.
|
|
|
| >| Who brought TCP/IP to the masses? -- BSD did. Who wrote the
TCP/IP
| >| stack everyone associates with early Windows TCP/IP? -- Trumpet
| > and
| >| Chameleon.
| >
| > So you're telling me that BSD, Trumpet, and Chameleon enjoy an
| > install-base equal to at least a majority of current Internet
users?
| > Interesting...
|
| Funny how you completely misunderstand the issue.
|
| It isn't about who made the OS the stuff is installed on, it's
about
| who made it possible for the stuff to get there.
No. YOU are flaming ME for what I stated. This issue goes back to
what I started. I stated that MS became the world's most successful
DISTRIBUTOR of TCP/IP when they added it to Windows 95. Go back and
read again. I know better than to claim that MS contributed
anything technical to the protocol.
[snip]
Have you realized how ridiculous this debate is? This point is not
a provable issue. I admit that I can't prove what I'm saying. You
do what you want with that. There's too much going on in this forum
to waste any more time with such a stupid discussion. I regret
starting it.
-B
void
wrote:
>
>void <fl...@interport.net> wrote in message
>news:slrn7t05ur...@interport.net...
>|
>| Not at all. Sendmail is one program, and dog-slow compared to the
>fully
>| modular postfix or qmail. Ease of implementation? Exchange makes
>even
>| MCSEs complain.
>
>I'm an MCSE and in my 2+ years of Exchange implementations, I
>haven't had cause to complain about it. Who might you be talking
>about? Are these recent graduates of the multitudes of "Boot Camps"
>around town?
No, the diploma-mill McSystems Engineers don't don't how bad they have it.
The people who make complain about Exchange usually have something to
compare it against. Do you? What other MTAs have you configured, and
on what platforms? Since I'm asking, I may as well save you the trouble
and admit that I've never set up Exchange. Just sendmail and postfix,
and I have played around with qmail but found it to be unintuitive and
hard to configure, so I decided not to use it.
>| Ease of recovery? Again, Exchange has generated some
>| of the worst war stories I've ever heard.
>
>I've had my fair share of recoveries (mostly due to bad hardware, or
>one of your "MCSEs" mucking up the system), but by following the
>many disaster recovery whitepapers available, it isn't difficult at
>all.
I've never tried it, so you win this one by default.
The only sendmail disaster recovery I've had to do was after I installed
Solaris 2.6 sendmail binaries on a production Solaris 2.5.1 host. Oh
no, there was another instance that wasn't my fault, where the queues on
an outgoing mail server filled due to a spammer. I just moved the
queues to partitions that still had free space and symlinked them. Unix
supports moving open files, between partitions or disks or whatever,
transparently to the process(es) which have them open.
Then, after checking with the local mail czar, I deleted all the spam,
being careful not to delete anything else by accident.
>| It's much easier to recover
>| flat files than opaque huge databases.
>
>It may be faster, but I haven't found it to be any easier.
Faster is better, if you value your time. But the relative merits of
databases and flat files have been debated in these forums ad nauseam
already, I have little desire to rehash those discussions.
void
>Matt Curtin <cmcu...@interhack.net> wrote:
>
>: Lest you believe the sort of cruft that Microsoft spews about their
>: role in bringing IP to the masses, I feel it necessary to engage in
>: the office of reminding you that whilst the Trade Press was quoting
>: Bill Gates saying "the Internet is the frothiest thing I've ever
>: services to Windoze users.
>
>Right... products which ran on Microsoft operating systems. Your point?
They could have run on any OS. Win16 was hardly a network-ready environment.
void
>| On Fri, 03 Sep 1999 06:00:59 GMT, Brent Davies
><brent...@spam-home.com>
>| wrote:
>| >
>| >Well, even though I agree, I don't see how this makes MS look any
>worse
>| >than Apple, Novell, or Digital, just to name a few. All had
>proprietary
>| >protocols, and all those protocols failed. What's the point?
>|
>| What's up with the attention span around here? I was refuting
>Chad's
>| baseless assertion that Microsoft brought TCP/IP to the masses
>like
>| Prometheus making off with BBN's fire.
>
>same behavior around here. Are you telling me that you can't
>identify that behavior when someone else does it?
Huh?
I made a statement in support of a certain conclusion, and you said that
you agreed, but didn't see how my statement led to another conclusion.
Since I never said that that statement led to that conclusion, it seems
to me that you're attacking a straw man.
void
wrote:
>
>You misrepresenting SOB!!! That was not my complete statement. I
>KNOW DAMN WELL that someone else would have done it. My question
>was, could that company have done it AS QUICKLY AS Microsoft did?
How can you praise their speed when they were the last to market with
TCP/IP on their own OS?! It makes no sense to me.
void
wrote:
>>news:slrn7t05ls...@interport.net...
>|
>| Microsoft's role in bringing TCP/IP to the masses is comparable to
>| that of Tucows or shareware.com.
>
>Who would have done it otherwise, and as quickly? Apple? They
>_still_ don't have their IP stack right, IMO.
What's your problem with it?
void
>
>Giving Microsoft credit for the proliferation of IP (as if every
>installation of Windows '95 gave IP to someone who didn't already have
>it) is, well, a misrepresentation. IP was already everywhere by the
>time Windows '95 came on to the scene.
Even when I was doing freelance desktop consulting[0] in 1997, I
encountered Trumpet Winsock not infrequently.
It's also well known among desktop support people that MS's built-in
dialer program is often replaced by software from AOL or other online
services. This causes no end of troubles.
[0] This makes it sound more glamorous than it actually was. I got
more out of the people I met than out of the work itself. My
favorite client was probably the crotchety jazzhead with the evil
Windows 3.1 laptop. His wife had a "Monorail" running Windows 95.
Remember those? Supercheap PCs before the idea's time had come.
void
wrote:
>
>Still very difficult to *get* *connected*. Browsers have never been
>all that difficult to use. Actually *connecting* your operating
>system to the Internet at that point in time was _well beyond the
>capabilities of the layperson_!
Balls! I knew plenty of laypeople who successfully installed and
configured Trumpet Winsock. Their ISPs would give them install disks,
they'd pop them in and run "setup", and everything would be done for
them. Maybe they'd have to pick a phone number from a list and type it
in.
>| Giving Microsoft credit for the proliferation of IP (as if every
>| installation of Windows '95 gave IP to someone who didn't already
>| have it) is, well, a misrepresentation. IP was already everywhere by
>| the time Windows '95 came on to the scene.
>
>This is interesting. You must be talking about prefessional
>computers that were not in the home environment.
I'm talking about home computers.
>Apple had an extremely shoddy IP implementation at the time (still is IMO)
>but that was it!
I still want to know what your problem with it is.
>If IP was everywhere, how do you explain the sudden explosion of Internet
>subscribers since Win95's debut? It must be because all of those home-user
>Unix workstations were just sitting in the person's closet.
You have it backwards. Windows got TCP/IP because the internet was
becoming a big thing. Having TCP/IP built into Windows made things
smoother for Windows users, of course, but Windows users were already
quite capable of connecting to the internet without advanced knowledge.
void
<cmul...@hipcrime.vocab.org> wrote:
>
>Wrong answer. Lan Manager, while first released by IBM, followed by AT&T,
>3COM, Banyan and a host of others was developed by Microsoft under contract
>by IBM. This was as I stated an OEM only release as such noone knew MS was
>involved. Source (other than my rememberance of the time): "Mastering
>Windows NT Server, Sixth Edition," Mark Minasi, 1998.
Ok, so Microsoft, AFAICT, has come up with one original TCP/IP application.
Your point.
>>Microsoft's role in bringing TCP/IP to the masses is comparable to that
>>of Tucows or shareware.com.
>
>With the exception that MS provided the platform for the TCP/IP to run on,
>and delivered that to the masses.
I know that Windows is the most popular operating system for the most
popular computer architecture. I fail to see how that makes Microsoft
responsible for one iota of the internet's success. I and others have
shown again and again that all the necessary software elements were
available from other vendors. You can repeat that Microsoft distributed
the elements until you're blue in the face, but that doesn't answer my
argument that others would have if Microsoft hadn't.
void
wrote:
>
>I am currently under the impression that a great many of the
>problems you are describing here were problems within the TCP/IP
>spec itself.
No. At least, I remember that the fixes did not change any of the
protocol semantics.
>I have been lead to believe that most of the DoS
>problems and fragmentation problems affected most, if not all, of
>the machines running the protocol.
Many were cross-platform, but some were MS-only.
Nick Manka
"Brent Davies" <brentdavi...@home.com> writes:
>| Try the various "security bug tracking" sites on the net. They
> have
>| lists of dozens of Denial of Service attacks against MS TCP/IP
> that
>
> I am currently under the impression that a great many of the
> problems you are describing here were problems within the TCP/IP
> spec itself. I have been lead to believe that most of the DoS
> problems and fragmentation problems affected most, if not all, of
> the machines running the protocol.
Most of them, possibly all of them, were unforeseen implementation
consequences. Many machines were affected, few as fatally as Windows,
many barely at all. Most of the non-Windows ones were also invulnerable
to most of the truly silly ones.
> I currently have little time to run that kind of a research project,
> so I'll have to ask that this issue be tabled, at the least, for the
> time being. I don't have any information to refute what you've
> said, and I won't likely have time to gather such information in the
> near future.
Granted.
> Interesting that you think substantiating statistics or comments
> like those you made above is such an unnecessary chore. Are you
I believe that anyone who is attempting to argue what you are
attempting to argue needs to be a lot more aware of the subject
matter. Most of this is what I consider public knowledge, that is
that anyone who wants to understand the issues can put a little
effort into it and get a reasonably thorough understanding back.
> No. YOU are flaming ME for what I stated. This issue goes back to
I'm responding to your response to something ultimately tracing
back to a stupid comment from Chad.
The place I pick up on it, <M6Vz3.19081$Rn.1...@news.rdc2.occa.home.com>,
states:
"""""
"| Microsoft's role in bringing TCP/IP to the masses is comparable to
"that
"| of Tucows or shareware.com.
"
"Who would have done it otherwise, and as quickly?
"""""
> what I started. I stated that MS became the world's most successful
> DISTRIBUTOR of TCP/IP when they added it to Windows 95. Go back and
> read again. I know better than to claim that MS contributed
> anything technical to the protocol.
That's not what it looks like you said.
It sure -looks- like you said that Microsoft expedited the deployment
of TCP/IP to the masses, while they really had absolutely nothing
to do with it. By the time Microsoft got around to putting TCP/IP
in Windows, anyone who knew how to spell it had already gotten it
from someone else.
However, as you've requested a reprieve, please don't feel obligated
to reply. I understand you may not agree with my interpretation of
events.
> Have you realized how ridiculous this debate is?
Rather, but I probably find it ridiculous in different ways.
void
>
>Most people do not like running seperate programs for everything. They'd
>rather have one program running all day with one interface rather than 5 or
>10 programs that all give different interfaces and require different
>training.
That's a specious argument, because if you put a nice GUI on an average
user's screen, they won't know or care if it is backended by one
monolithic app or a suite of small modular programs.
void
<cmul...@hipcrime.vocab.org> wrote:
>
>void wrote in message ...
>
>>What's up with the attention span around here? I was refuting Chad's
>>baseless assertion that Microsoft brought TCP/IP to the masses like
>>Prometheus making off with BBN's fire.
>
>articles?
It was Chad Myers. See <http://x36.deja.com/getdoc.xp?AN-520266384&CONTEXT=
36426396.1383464963&hitnum=10>.
My news reader does thread articles.
void
wrote:
>
>Do you realize how many steps you are adding to the process by
>seperating these functions into their own "daemons"? Then imagine a
>10,000 node network where 100,000 messages calendar/meeting
>requests, etc. pass through the system every day. These added steps
>will severely impact performance.
Oh, come on, Brett. Modularity does not imply poor performance and it
hasn't for years. What a canard.
Brent Davies
| On Fri, 03 Sep 1999 23:03:28 GMT, Brent Davies
<brentdavi...@home.com>
| wrote:
| >
| >You misrepresenting SOB!!! That was not my complete statement.
I
| >KNOW DAMN WELL that someone else would have done it. My question
| >was, could that company have done it AS QUICKLY AS Microsoft did?
|
| How can you praise their speed when they were the last to market
with
| TCP/IP on their own OS?! It makes no sense to me.
Well, it's really by no attempt on their part. It was kind of a
defacto thing. Windows 95 was on every PC purchased and TCP/IP had
been added to it, so the distribution was incredibly fast.
If they had never added IP to their OS, it still would have been
added by someone else. If my saying that makes everyone happy, can
we all go home? =}
-B
Brent Davies
| On Fri, 03 Sep 1999 19:12:27 GMT, Brent Davies
I've used MS-Mail, Post.Office, and Sendmail. I found Sendmail to
be as you described qmail. I am still working on Sendmail, with all
of the rulesets and toys burried under the surface. I have it
working in a standard config, which wasn't hard, but there as some
non-standard things I'd like to do with it, but am having trouble.
MS-Mail was a horrible excuse for an MTA, and Post.Office simply got
the job done. The only groupware server I've worked on is Exchange.
|
| >| Ease of recovery? Again, Exchange has generated some
| >| of the worst war stories I've ever heard.
| >
| >I've had my fair share of recoveries (mostly due to bad hardware,
or
| >one of your "MCSEs" mucking up the system), but by following the
| >many disaster recovery whitepapers available, it isn't difficult
at
| >all.
|
| I've never tried it, so you win this one by default.
|
| The only sendmail disaster recovery I've had to do was after I
installed
| Solaris 2.6 sendmail binaries on a production Solaris 2.5.1 host.
Oh
| no, there was another instance that wasn't my fault, where the
queues on
| an outgoing mail server filled due to a spammer. I just moved the
| queues to partitions that still had free space and symlinked them.
Unix
| supports moving open files, between partitions or disks or
whatever,
| transparently to the process(es) which have them open.
That is a feature I'd pay for ! ! (I'm being quite serious)
|
| Then, after checking with the local mail czar, I deleted all the
spam,
| being careful not to delete anything else by accident.
|
| >| It's much easier to recover
| >| flat files than opaque huge databases.
| >
| >It may be faster, but I haven't found it to be any easier.
|
| Faster is better, if you value your time. But the relative merits
of
| databases and flat files have been debated in these forums ad
nauseam
| already, I have little desire to rehash those discussions.
Per your request, I'll try to avoid the database merits issue. But
given the fact that I have my reasons for liking the all-in-one
groupware database design, I think the time required for recovery is
acceptable. Besides, I've only had to do it twice in more than 10
times as many configurations (different clients).
-B
Brent Davies
| On Sat, 04 Sep 1999 03:00:26 GMT, Brent Davies
<brentdavi...@home.com>
| wrote:
| >
| >I am currently under the impression that a great many of the
| >problems you are describing here were problems within the TCP/IP
| >spec itself.
|
| protocol semantics.
|
| >problems and fragmentation problems affected most, if not all, of
| >the machines running the protocol.
|
Ok, I think I crossed some "jargon" line that I didn't see. If
"spec" was the wrong word, I apologize. But there were many
cross-platform problems. Some were MS only, yes. But at least I
can get an MS IP stack to stay up with a current implementation.
Yeah, yeah. Too little, too late, right?!?
-B <-----too tire to debate, but too damn stubborn to leave before
reading everything!
Brent Davies
| On Fri, 03 Sep 1999 19:13:34 GMT, Brent Davies
<brentdavi...@home.com>
| wrote:
| >
[snip]
| >
| >I was replying to only one of your claims. You are famous for
that
| >same behavior around here. Are you telling me that you can't
| >identify that behavior when someone else does it?
|
| Huh?
|
| I made a statement in support of a certain conclusion, and you
said that
| you agreed, but didn't see how my statement led to another
conclusion.
| Since I never said that that statement led to that conclusion, it
seems
| to me that you're attacking a straw man.
Because I am too tired, you have totally confused me. Suffice it to
say that I obviously didn't care that much, since I've forgotten
what the original issue was. Besides that, I mistook you for Jedi
in my comment about "being famous for that behavior around here".
I'll concede because of my mistake.
-B
Brent Davies
| On Fri, 03 Sep 1999 19:14:52 GMT, Brent Davies| >news:slrn7t05ls...@interport.net...
| >|
| >| Microsoft's role in bringing TCP/IP to the masses is comparable
to
| >| that of Tucows or shareware.com.
| >
| >_still_ don't have their IP stack right, IMO.
|
| What's your problem with it?
That's the second time you've asked, but I'm too tired right now to
get into it. Sometime in the next couple of days I'll explain to
you my experiences with it. Ok?
-B
Brent Davies
| In article <eX%z3.19313$Rn.1...@news.rdc2.occa.home.com>,
| "Brent Davies" <brentdavi...@home.com> writes:
|
|
| > what I started. I stated that MS became the world's most
successful
| > DISTRIBUTOR of TCP/IP when they added it to Windows 95. Go back
and
| > read again. I know better than to claim that MS contributed
| > anything technical to the protocol.
|
| That's not what it looks like you said.
Alright. Well, the sentence above matches my intent much more
closely than anything I may have written earlier in this thread. If
you have no problem with that statement, and the meaning behind it,
let's call this debate to a close. Hmm?
-B
Jerome Chan
(void) wrote:
> On 3 Sep 1999 20:44:44 GMT, Stephen Edwards <ja20...@primenet.com> wrote:
> >Matt Curtin <cmcu...@interhack.net> wrote:
> >
> >: Lest you believe the sort of cruft that Microsoft spews about their
> >: role in bringing IP to the masses, I feel it necessary to engage in
> >: the office of reminding you that whilst the Trade Press was quoting
> >: Bill Gates saying "the Internet is the frothiest thing I've ever
> >: seen", products like Trumpet Winsock and Chameleon were providing IP
> >: services to Windoze users.
> >
> >Right... products which ran on Microsoft operating systems. Your point?
>
> They could have run on any OS. Win16 was hardly a network-ready environment.
That's like saying computer desks help introduce the internet because
Microsoft windows run on top of desktops. :P
Chad Mulligan
void wrote in message ...
<brentdavi...@home.com>
>wrote:
>>
>>Do you realize how many steps you are adding to the process by
>>seperating these functions into their own "daemons"? Then imagine a
>>10,000 node network where 100,000 messages calendar/meeting
>>requests, etc. pass through the system every day. These added steps
>>will severely impact performance.
>
>Oh, come on, Brett. Modularity does not imply poor performance and it
>hasn't for years. What a canard.
Ben I've got to agree with you on this one. OS/2 could and did handle this
type of mission with separate applets for the user, with an invisible
infrastructure SOM
Chad Mulligan
void wrote in message ...
><cmul...@hipcrime.vocab.org> wrote:
>>
>>Wrong answer. Lan Manager, while first released by IBM, followed by AT&T,
>>3COM, Banyan and a host of others was developed by Microsoft under
contract
>>by IBM. This was as I stated an OEM only release as such noone knew MS was
>>involved. Source (other than my rememberance of the time): "Mastering
>>Windows NT Server, Sixth Edition," Mark Minasi, 1998.
>
>Ok, so Microsoft, AFAICT, has come up with one original TCP/IP application.
>Your point.
>
You said:
"
>Assert failed. Someone (Brett I think) mentioned LanMan, but I think IBM
>came up with that first. So Microsoft has still brought nothing to
>TCP/IP that others haven't done first -- in a user-friendly fashion no
"
You were also wrong about Lan Manager. So stop making baseless assertions.
>
>I know that Windows is the most popular operating system for the most
>popular computer architecture. I fail to see how that makes Microsoft
>responsible for one iota of the internet's success. I and others have
>shown again and again that all the necessary software elements were
>available from other vendors. You can repeat that Microsoft distributed
>the elements until you're blue in the face, but that doesn't answer my
>argument that others would have if Microsoft hadn't.
>
But the platform that was already in the home when the Internet craze took
off was which one?
Chad Mulligan
Brent Davies wrote in message ...
>
>Chad Mulligan <cmul...@hipcrime.vocab.org> wrote in message
>news:ysWz3.640$i03....@typhoon-sf.snfc21.pbi.net...
>|
>[snip]
>| >
>|
>| BTW, MS and Exchange, and Active Directory, use LDAP an
>established standard
>| protocol. That's why Outlook can work like an exchange client
>without
>| needing exchange. Let's see Notes do that.
>
>You want established standards? The Exchange Directory service
>completely conforms with X.500; the internal Exchange Server message
>delivery protocol is X.400 (try to delete your mailbox's X.400
>address. All hell breaks loose!); Internet Mail Connector
>completely conforms to SMTP/POP3 standards; etc.
>
>How much more "standardized" do you want it to be? ("you" used in
>general. Chad, I know that you're in agreement).
Yes, while at work today I did some other research on MS TCP/IP and learned
some interesting things:
1. NetBIOS over TCP/IP is a standard supported by RFC
2. It is not a protocol, but an API to the TDI driver.
3. When a NetBIOS machine connects with another, all traffice is
multiplexed through that single connection.
4. Digging around in microsoft.com I discovered an entire specification
draft and API for QOS routing dated within the last 12 months. Interesting
reading that.
5. I actually couldn't find anything in the MS documentation about TCP/IP
that wasn't supported by an established standard.
The paper is available under technical references, in the network and
communications section of the NT server web page at microsoft. I'm not
going to post a URL there's enough info there for any one with a clue to
find the paper.
Thus ends more FUD.
>
>[snip]
>
>-B
>
Matt Curtin
Brent> Mac also had a GUI OS before MS came around.
...as did the Lisa, vrarious Lisp machines, and the STAR,
Brent> So? Who got it on to millions of desktops?
The Macintosh reached millions. Heck, the Apple IIgs, with its color
GUI reached 1 million units.
Brent> I am simply stating that MS succeeded in bringing these
Brent> capabilities to the masses, where the other companies did not
Brent> succeed in as wide a distribution.
I'm not going to disagree that Windows is (now) more widely deployed
than any of these others. But the assertion that Microsoft is to be
given credit for "enabling" millions of people to use computers, the
Internet, etc., and for doing so faster than anyone else, is silly.
Microsoft didn't do it "faster" than anyone else. As usual, Microsoft
didn't get involved until the group known in marketing circles as the
"early majority" were ready for it. Without Microsoft, we wouldn't
have fewer computer users today. Without Microsoft, we wouldn't have
less sophisticated GUIs than we have today. It's really that simple.
They haven't been "growing" any markets: they've been using marketing
and a ruthless business-is-war mentality to take markets from others.
Matt> Of course, if they wanted an operating system with IP built in,
Matt> such were *already available*.
Brent> Such that they could actually by used by laypeople?
Yes indeed; systems in which they can be quite productive. Firstly,
let me observe that Unix wasn't the only platform out there that was
IP capable at the time. Secondly, it's worthwhile to point out that a
little training goes quite a long way. Secretaries and clerical staff
used Unix and tools like troff very happily for many years.
Brent> Still very difficult to *get* *connected*. Browsers have never
Brent> been all that difficult to use. Actually *connecting* your
Brent> operating system to the Internet at that point in time was
Brent> _well beyond the capabilities of the layperson_!
I don't buy this argument at all. I have known too many nontechnical
people who installed the package, clicked on the icons, and went
surfing.
Brent> This is interesting. You must be talking about prefessional
Brent> computers that were not in the home environment.
Not exclusively.
Brent> Apple had an extremely shoddy IP implementation at the time
Brent> (still is IMO) but that was it!
Except for the previously-mentioned packages that gave Windoze users
connectivity. Not to mention that other, "more exotic" platforms also
had this capability.
Brent> If IP was everywhere, how do you explain the sudden explosion
Brent> of Internet subscribers since Win95's debut? It must be
Brent> because all of those home-user Unix workstations were just
Brent> sitting in the person's closet.
Do yourself a favor and look at the numbers. It isn't like the
Internet was some sort of obscure technology that no one was using
until Windoze '95 appeared, at which point, it exploded into the
ubiquitous matrix we have now. The Internet was already in a state of
exponential growth. Microsoft and Windows '95 were doing nothing more
than surfing the exponential growth curve. Microsoft isn't any kind
of "benefactor" bringing the power to people who couldn't have it
otherwise.
--
Matt Curtin cmcu...@interhack.net http://www.interhack.net/people/cmcurtin/
Matt Curtin
"Brent Davies" <brentdavi...@home.com> said:
Brent> Cite some examples of MS's so-called "crippled" IP stack.
Matt> Three examples are: winnuke, bonk, and teardrop. Another
Matt> excellent example is the inability of NT to be used as a
Matt> reliable sniffer. Observe the inability of NT to serve as a
Matt> host platform for the Network Flight Recorder
Matt> (http://www.nfr.net/).
Brent> All problems that have been fixed.
Wrong. NT *still* can't support NFR. The IP stack sucks by
comparison to *BSD.
Brent> And don't try and tell me that the main TCP/IP code hasn't
Brent> needed some fixing here and there.
Other vendors' TCP/IP implementations didn't seem to require three and
four patches to fix what's essentially one problem that made itself
manifest in different ways.
TNC
>
> TNC <gr...@mpinet.net> wrote:
>
> : Chad Myers wrote:
> Hmmm... a search of "TCP/IP" at http://www.microsoft.com/ yields
>
> http://www.microsoft.com/NTServer/nts/techdetails/compares/TCPIntrowp.asp
> http://www.microsoft.com/SYSPRO/deploy/tcpintro/tcpintro14.htm
>
> Is this the documentation that doesn't exist?
>
> I guess I wasted a lot of money on those WindowsNT books, because I'm sure
> there's no documentation whatsoever in them about TCP/IP.
MS constantly uses undocumented APIs to gain an internal advantage and
to break competitor's products. This is _well documented_ about them.
Your NT books don't tell you what you need to know to _really_ compete
with MS. This is not the case with open systems (note there is a
difference between open systems and open source, one can build a closed
source but open system in the sense that the APIs and protocols are
fully documented and available. MS has never and will never do this).
> As I understand it, MS developer kits are meant for serious developers,
> and not "hackers". OpenSource users and developers often have a difficult
> time grasping the point of commercial software... to make something
> commercially viable... therefore, to make money by providing a service or
> product professionally.
I guess Redhat doesn't make any money huh? Open source people have a
different business model, that's all.
>
>
> So, you're suggesting that the _only_ reason they made those modifications
> was to screw up Sun's Java agenda?
Um... pretty much.
> I guess it's too much to consider that
> Microsoft may have actually been attempting to improve upon it for use
> with Windows operating systems.
So if I buy the rights to distribute your song/novel etc... and then
change it all around, change the meaning, etc... and THEN continue to
distribute it as though it is still your product, you WON'T sue me? As
for MS "improving" it, this is interesting. It's interesting because
everyone else in the industry seemed content to create a Java that was
Sun compliant. Now this forces us to conclude one of the following:
1) MS was trying to corrupt and destroy Java by making
developers/consumers choose between their version and everyone else's.
2) MS developers are just so damn smart, so GODLIKE in their
intellectual superiority to the rest of us dolts that they, only they
were able to see how this product might be better. And that those
modifications just *happened* to only work on Windows.
Hmm... which is more plausible.
MS's policy of "embrace and extend" is not something I or a bunch or
paranoid Linux advocates made up. It is something that _comes from MS_
in the form of the "Halloween documents". I don't have the URL but put
that phrase in your search engine and you'll see what your favorite
bastards are really up to.
>
> Even if Microsoft is responsible for hindering Java, I would sooner thank
> them, than detest them.
Really, why? because it threatens platform based monopolies?
> It wasn't weak... it wasn't even an analogy. It was typical Mark
> S. Bilk nonsense.
Technically it was an analogy - a better one might have been that MS
works to convince you to buy air from them at $1.50/litre instead of
just breathing what's around you.
--
"Gun control proposals are nothing more than a modern liberal
suggestion that government, which is unable to protect its citizens,
makes sure those citizens cannot defend themselves."
- Robert. H. Bork
Hobbyist ©
> > Joke's on you... I have a wooden leg.[0]
>
> A good match for your head.....
I'm sure that you've improved Jerry.
A couple weeks back you would have surely typed:
"A good match for your head, you fuck-wit..."
or
"A good match for your head, shit for brains ..."
LOL.
--
-=Ali=-
Erik Funkenbusch
> On Fri, 3 Sep 1999 15:43:11 -0500, Erik Funkenbusch <er...@visi.com>
wrote:
> >
> >Most people do not like running seperate programs for everything. They'd
> >rather have one program running all day with one interface rather than 5
or
> >10 programs that all give different interfaces and require different
> >training.
>
> That's a specious argument, because if you put a nice GUI on an average
> user's screen, they won't know or care if it is backended by one
> monolithic app or a suite of small modular programs.
The "nice gui" is the app in this case, with a single user interface.
I was talking about prgrams such as elm, tin, etc..
Your talking about using backend MTA's and such with a single viewer app.