Linux-Advocacy Digest #211, Volume #34 Sat, 5 May 01 12:13:02 EDT
Contents:
Re: Justice Department LOVES Microsoft! (T. Max Devlin)
Re: Justice Department LOVES Microsoft! (T. Max Devlin)
Re: Justice Department LOVES Microsoft! (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
Re: Richard Stallman what a tosser, and lies about free software (T. Max Devlin)
----------------------------------------------------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:40 GMT
Said Daniel Johnson in comp.os.linux.advocacy on Fri, 04 May 2001
>"Rick" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Daniel Johnson wrote:
>> > "Rick" <[EMAIL PROTECTED]> wrote in message
>> > news:[EMAIL PROTECTED]...
>> > > Daniel Johnson wrote:
>> > > Then you might want to eplain why Windows ran on top of DOS.
>> >
>> > Because users had lots of DOS apps and they
>> > wouldn't switch if it means giving them up.
>>
>> SO, Windows DID run on DOS.
>
>Oh yes. Most certainly. Even Windows 3 was
>an orthodox DOS program, when in real
>mode.
It is still a DOS program, if unorthodox (known as a "DOS extender")
when in protected or enhanced mode.
>[snip]
>> > Being a "better DOS than MS-DOS" is damning it with
>> > faint praise. MS-DOS was *terrible*; DR-DOS was
>> > only slightly less terrible.
>>
>> Note: no response. I will ask again:
>> Since Windows ran on top of DOS. And DR-DOS was a better DOS than
>> MS-DOS, how can you support your point? (See above point)
>
>I have already supported it; I do so by poitned
>to important features that Windows has and DR-DOS
>has not.
These features are the 'DOS extender' I was talking about. Very similar
to EMM386, if quite a bit more complex to provide a 'middleware
platform', or whatever you call the 'enhanced GUI' thing that is
Windows, sans all of the OS traits attributed to Windows, which are
actually DOS.
>Your commentary seems irrelevant to the point
>I was making.
Your commentary seems like pathetic squirming by a sock puppet or a
troll.
>[snip]
>> > Yes, those guys couldn't say "no" when Bill wanted
>> > Windows apps. In the days of Windows 1 and 2, that
>> > was important.
>>
>> They also had inside info. Thats waht landed M$ in trouble.
>
>They may have done, for all the good it did them. But
>it isn't what landed MS in trouble, as far as I can
>see.
It was the PPL; the FTC tried to figure out just what was illegal about
it. It was obvious to any anti-trust lawyer you looked that it was
illegal, since it resulted in monopoly, but there wasn't any
codification of law or precedent that precisely explain why it had no
competitive merits. You know how gullible judges are, and the case was
very hard to put together. Almost five years of investigation didn't
lead to a trial. Meanwhile, the DoJ took an interest in Microsoft's
force bundling of Windows with DOS, and took over the case. A couple
years later, MS signed a consent decree to avoid prosecution, promising
not to do either ever again. They broke both promises within the year,
of course.
Now do you see why you don't get the benefit of the doubt, smiling while
you defend these thieves?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:41 GMT
Said Daniel Johnson in comp.os.linux.advocacy on Fri, 04 May 2001
>"Ayende Rahien" <[EMAIL PROTECTED]> wrote in message
[...]
>Microsoft at this point decided they had had
>enough of IBM and informed them that they
>no longer wished to collaborate. They wanted
>to pursue Windows instead.
This is where your telling goes awry, I think. MS kept the industry
supporting OS/2, sometimes quite forcefully (or at least forcefully
excluding them from developing on Windows, insisting that OS/2, not
Windows, was "the replacement for DOS".)
>They knew a good thing when they saw one, and
>they weren't about to go down with OS/2, no
>matter what they'd said about it in the past.
OS/2 was a clearly superior product, but lacked the necessary feature to
gain "market acceptance". Microsoft had a monopoly in DOS, and OS/2
wasn't DOS. Windows could be DOS (but yet pretend not to be DOS, even
to the point of not being DOS, but being a VMS-like thing specially
created to not be DOS, without being anything else other than whatever
OS Windows needs.)
>At this time IBM and MS had their intellectual
>properties rather tangled up, but they sorted them
>out, assigning bits to each company.
>
>One gets the feeling IBMs lawyers were asleep
>at the switch on this one. You're not going
>to believe how they split this stuff up.
Lawyers aren't businessmen, goofball. Businessmen, not lawyers,
negotiate deals; lawyers just right them down. One gets the feeling IBM
was still under the impression they could out-compete the
anti-competitive. They should have known better, of course, still
climbing out from under several consent decrees themselves. But of
course, in the end, that explains the "asleep at the switch" angle: they
were perfectly aware of what was going on, and in retrospect have done
an admirable job of living through it.
>IBM got OS/2 1.x and the still unfinished 2.0 but
>would pay royalties to Microsoft for the various
>code MS had written for these products as long
>as they continued to use this code.
>
>This included the file system and the graphics
>subsystem, and it totalled up to a few hundred
>dollars a box.
>
>MS agreed that they would license Windows
>source code to IBM for a pre-arranged royalty,
>which would increase every year. These royalties
>were pretty low though- around twenty
>dollars a box, or so. IBM was allowed to
>tinker with the code, too.
>
>MS got the OS/2 3.0 code they had been working
>on lock, stock and barrel. It would become
>Windows NT in a few years time.
Perhaps you're confabulating their use of SMB from LanManager for NT
networking. OS/2 3.0 code is "Warp", and MS never had anything to do
with that.
>IBM and Microsoft then went at each other
>great guns. But that's another story.
No, IBM competed and Microsoft continued to illegally monopolize.
That's pretty much the end of the story.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.ms-windows.nt.advocacy,comp.sys.mac.advocacy,comp.os.ms-windows.advocacy
Subject: Re: Justice Department LOVES Microsoft!
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:42 GMT
Said JS PL in comp.os.linux.advocacy on Fri, 4 May 2001 12:34:15 -0400;
[...]
>I don't care what the problem is. I prefer an OS that works well without all
>the hours of configuration.
I prefer an OS that works consistently without all the hours of
reconfiguration.
>On the plus side, setting up connection sharing was a no-brainer.
>
>I'll keep it around despite some of the cosmetic flaws. It is worthy of my
>limited hardware resources.
Bwah-ha-ha-ha-ha. I think you misspelled "Linux kicks the shit out of
monopoly crapware!", JS PL.
Thanks for your time. Hope it helps.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy,misc.int-property
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:43 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 22:44:23
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Ayende Rahien in comp.os.linux.advocacy on Wed, 2 May 2001 21:08:31
>
>
>> [...]
>> >No, I'm not talking about copyright here. I'm talking software design.
>> [...]
>>
>> Well, that's the problem then.
>
>When you are talking about API, you talk about software design, when you
>make erronomous statement about API, I'll correct you.
Are you TOTALLY clueless?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:44 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 23:02:22
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Austin Ziegler in comp.os.linux.advocacy on Thu, 3 May 2001
>> >On Thu, 3 May 2001, T. Max Devlin wrote:
>> >> Said Roberto Alsina in comp.os.linux.advocacy on 2 May 2001 21:08:02
>> >>> It has been done, IN PRACTICE. That you, who has never exercised the
>> >>> craft claim that what has already been done is impossible, is quite
>> >>> irritating.
>> >> I'm sure. No, it has not been done in practice; it is impossible in
>> >> practice to write a program which requires a library that doesn't yet
>> >> exist in any way.
>> >
>> >For the nth time, this is false.
>>
>> For the nth+1 time, your contention is flawed.
>
>No, it isn't.
>Prove it. By example.
You cannot prove something is impossible by example, Ayende. I'd think
if you're smart enough to program you should be smart enough to know
that already.
>We have provided plenty of examples, you're stuck with "It's not true, mommy
>told me so!"
No, I've stuck quite clearly with, "You seem to be misunderstanding what
I am saying; none of those are examples, nor could there be any
examples, because what you are suggesting is 'true' is physically
impossible, like throwing paints into the air and having them land in
the shape of the Mona Lisa."
>> >The program may not be *functional*,
>> >but you can definitely write such a program. [...]
>>
>> Find a non-programmer who calls a random bunch of characters that do not
>> perform function "a program". Why would you even bother writing a
>> program that is not functional? Just trying to 'cheat' copyright law,
>> without actually getting into any trouble?
>
>Find a non-programmer who is interested in application development.
Find a copyright attorney who's interested in application development.
Then get in touch with the FSF if they believe, as Les and Roberto and
Austin (and now you, although I'm still pretty sure you don't know what
position it is you are defending so forcefully) that the GPL is not
enforceable. Quite whining to me about it; I was just trying to explain
to you nimrods why your criticisms of their position are both flawed and
foolish.
>Such things can be of many uses for a programmer. From proof of concept to
>just studying the way the application work to debugging.
So long as it remains solely useful to a programmer, to be honest, I
don't give a fuck. Any snippet of code is supposedly protectable under
copyright, but that doesn't make it a "program". From 'proof of
concept' to 'example' to 'debugging aid', it might be code, but it ain't
a "program". And, honestly, if you're a programmer and you disagree on
this point, YOU ARE MISTAKEN.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:46 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 22:56:31
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
[...]
>> I'm sure. No, it has not been done in practice; it is impossible in
>> practice to write a program which requires a library that doesn't yet
>> exist in any way.
>
>It's done in pratice since just about forever.
>Here is a code that rely on non existing library.
It is WITHOUT ANY CONCEIVABLE DOUBT, the case that you misunderstood
what "it" is. Perhaps you wish to fudge the concept of a 'non-existent
library'. Perhaps you are under the impression this code is
'functional' or 'useful'. I doubt even you would be ignorant enough to
believe anyone would pay you for it though. So theoretically, it might
be copyrightable, but practically speaking, I don't think you need to
waste your time registering it or anything.
By "in theory" I meant to exclude such moronic ideas as those coming
from you right now. This "here is code" is THEORY. PRACTICE is when
you make a real program that real people use.
>Here is the library's API:
[...]
>Now, I can gurantee you that anyone who code in C++ can:
>A> implement the API
>B> predict what the output of the program would be according to the API,
>*without* implementing the API.
Make money off it? That's the only part that counts in copyright, you
know: you gets paid.
>Give it to your friend with the extensive experiance in programming, T. Max,
>and convice him to say why he think that A & B are incorrect.
My friends are no more concerned with THOUGHT EXPERIMENTS like this than
I am.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:47 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 23:33:45
[...]
>> You are mistaken about what "it" is.
>
>Programming to an API without implementation.
No, creating a program that is derivative of something that doesn't
exist.
>Possible, in practice, in fact, used just about everywhere in application
>development.
No real-world software has EVER been written to an *API* that doesn't
"exist". If you write the program and say "some day their will be a
library that..." then you have created the API. Whether you documented,
coded, or 'implemented' it first is entirely irrelevant, you've defined
it enough to use it, so it exists. You cannot use an API that doesn't
exist. Nor is coding in a library the only thing which can be called
"implementation".
This isn't a situation of my ignorance of programming; I've explained
that already several times, and don't understand why it is so hard to
understand; you'd think programmers would be smart enough to figure it
out after a dozen exchanges! The question has nothing to do with
programming, it isn't a matter of what is or is not possible in
programming, and it's getting to the point where programmers appear
particularly ill-equipped to understand the *COPYRIGHT ISSUES* involving
derivative works, and the FSF's claim, which does stand up however much
you waste my time arguing against a point you don't grasp.
The MOST you can say is that the position has not been FULLY tested in
court. It's a lame excuse for a position, but at least it is better
than just saying "the judge was wrong; he doesn't understand copyright",
like people do when they read the Lasercomb America v. Reynolds case.
Stamping your feet and saying "it just is not so and I know because I'm
a programmer and you don't so there" isn't going to get you anywhere.
Now, would you like to try to understand the point, Ayende, now that
you've stumbled into the middle of it, or would you rather keep
trolling?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:48 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 23:33:50
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
[...]
>> The adverb "simple" was supposed to give you a clue: I wouldn't have
>> included it if "braindead" were acceptable.
>
>Simple is simple, if you wanted complex, you should've asked for more
>functionality.
Quite squirming. I did not post the sentence as a request to begin
with, moron. Your continued attempts to intentionally misunderstand the
situation is getting rather pitiful. I explained why no popular,
convenient, functional applications were around for fully managing all
of the intricacies and details of extensions, file types, and
application associations. You PROVED my point, by cobbling something
together to prove how "simple" it was, which only succeeded in proving
how useless such simple implementations are.
I've TRIED to explain to you in several ways why you are wasting both
our time, and failing utterly to cast any shadow on either my honesty or
integrity, no matter how hard you try or how successful you might
self-evaluate your efforts to be.
I wanted a practical application for doing something as an end user; you
gave me an amateurish blob of Win32 calls that act on the registry,
similar to three other programs I had ALREADY described to you, and
pointed out that they were ALREADY unacceptable. Your pathetic
knock-off isn't bad enough to be 'unacceptable', having failed to rise
to the level of "valid", since you never actually wrote the program,
just designed it in your head and assumed it would work. It allowed me
to change reg keys, but then again, I already have regedit, and to be
honest managing extensions/types/associations is far easier using
regedit than your cobbled-together piece of shit; it would be more
trouble than its worth, so, no, it doesn't merit the title 'simple
application'. Just a lousy example which perfectly made MY point.
>> >It does what you asked for, it does so efficently, with as little code as
>> >possible. You don't ask more than a programmer.
>>
>> I think you mean "from a programmer", and yes, indeed, I do.
>
>No, you don't.
>Answer this: The application does what you asked for, yes or no?
No. But that doesn't address the issue. The right question is "Is this
acceptable, simply because it does what I thought you were asking for?"
The answer, of course, is still "no", but it gives you an idea of what
"more" one asks from a programmer. If you were a good programmer
(professional, I mean, not simply competent) then you would not present
such a lame excuse for functionality, EVEN IF IT WERE what I asked for,
and it is most thoroughly NOT, regardless of how confused you are on
that point. I asked for "a simple application to effectively manage
extensions, file types, and associations", what I got was a student
project that didn't even merit a passing grade.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:50 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 23:33:38
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Ayende Rahien in comp.os.linux.advocacy on Thu, 3 May 2001 01:14:57
>
>> >For computer programmers, there is a terminology of their own, which
>tries
>> >not to leave ambiguaties
>> >It would help if you told us what you *think* an API is, then we could
>> >disillution you.
>>
>> The issue is not what I think an API is; I have an entirely correct
>> understanding of the term. The discussion concerned copyright, and
>> whether a program is derivative of a library which it requires.
>
>State it.
>It being what you think API is.
I'm tempted to ignore the post, or avoid the question, just to frustrate
you, since in fact the issue is entirely off-topic. Still, just so you
know: Application Programming Interface, and that is what it is, not
just what I think it is. There's nothing about any of those terms which
requires additional explanation (though you're probably a tad confused
on 'interface', but that's not important now) but if you would like to
learn more, feel free to ask.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy,misc.int-property
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:51 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 21:56:59
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Ayende Rahien in comp.os.linux.advocacy on Thu, 3 May 2001 01:22:52
>> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>> [...]
>> >No, if I want to know what an API is, I don't go to the documentation.
>> >I go the the API's documentation.
>>
>> Is it only in English that these statements are incomprehensible?
>
>It's English.
I don't think so. I think it is the fact that you used the term
'documentation' twice, in a conflicted kind of way. "I don't go to the
documentation, I go to the documentation".
>> >What is an API documentation?
>> >API can be divided into two parts:
>> >Functions declaration.
>> >Text that describe what those functions does.
>> >
>> >You would need to be more spesific about the documentation part,
>> >documentation of what?
>>
>> Yes, that was my point.
>
>That documentation is part of the API.
(!)
>An API is not complete without the documentation of what its function does.
You mean the library won't work if a programmer makes a function call
unless the function is documented?
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy,misc.int-property
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:53 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 23:04:46
>"WesTralia" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Ayende Rahien wrote:
>> >
>>
>> [...]
>>
>> > An API is not complete without the documentation of what its function
>does.
>>
>>
>> An API, is an API, is an API! It's always nice to have a documented
>> API, but whether an API is documented or not is absolutely exclusive
>> of whether the API is complete or not.
>
>A bunch of function declaration is not an API.
>You need to know *what* those function does for it to be an API.
Ayende, this is purely what we call an "epistemological" argument.
"Epistemology" is the philosophy of words, the study of *Meaning*.
There is little hope for this discussion, particularly given the
difficulty translating (I know English is not your first language, and
despite your obvious capabilities, I think use of words becomes too
arbitrary when discussing epistemology for their to be much productive
interaction.)
Whether a "bunch of functions" "is" or "is not" "an API" is a matter of
perspective and context only; there is nothing in the term itself which
can absolutely determine the distinction.
>Sure, you can use a function declaration in your program, but I doubt that
>incorporating std::vector<char> XDdasSR433(int,double,std::string); into
>your code is going to be very helpful.
Epistemological discussions always become interchangeable, eventually,
with ontological discussions. Ontology is the philosophy of existence,
the study of *Being*. If you know what XDasSR433 does, lack of
documentation will not prevent it from functioning. You may say that
the only way to know what it does is to read the documentation, but that
is certainly not true: you might simply read the source code for the
function, eh? What is "knowledge", what are we saying when we claim a
mental construct (program, API, library) "exist", when is it correct to
say "an API *is*" any other thing? These are philosophical questions,
and on Usenet our choice is to avoid them, or just wait for the
conversation to work its way around to "Hitler".
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
From: T. Max Devlin <[EMAIL PROTECTED]>
Crossposted-To: gnu.misc.discuss,comp.os.ms-windows.advocacy
Subject: Re: Richard Stallman what a tosser, and lies about free software
Reply-To: [EMAIL PROTECTED]
Date: Sat, 05 May 2001 15:13:54 GMT
Said Ayende Rahien in comp.os.linux.advocacy on Fri, 4 May 2001 21:54:02
>"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Said Roberto Alsina in comp.os.linux.advocacy on 2 May 2001 18:59:37
>
>> >>In the real world, an application program ROUTINELY needs to know more
>> >>about a function than the API documentation itself can provide.
>> >
>> >You know this because of your extensive programming eperience, right?
>>
>> No, I know it because people who have extensive programming experience,
>> who's opinions I trust, and who understand my point correctly, say it is
>> so.
>
>Extensive mean that they read Visual Basic for Dummies and created their own
>Hello World program *without* looking in the book?
None of them use Visual Basic at all, last I talked to them.
>*No one* with any reasonable programming experiance would say this.
No, no one with any reasonable programming experience would fail to hope
that this isn't so. But the real world says different, as updating
libraries routinely break applications (written to the same API, but a
different implementation) and they are not as idealistic as you are, I
guess.
>An API is a contract between a library and the application.
>The application knows that when it does X, then library would do Y, because
>it's spesified in the contract.
I can go along with the analogy of a contract between a library and an
app, but I'm afraid you've stretched the metaphor too far for me, when
you say that the application "knows" things because of it. There's
nothing about being 'in the contract' (which I presume from the other
things you've said you would identify closely with the documentation of
the API) which prevents the library from functioning the way the library
code functions, regardless of any promises made which contradict this
physical fact.
>That contract hide as much as possible from the user's of the library,
>because the point is to give the library as much freedom as possible in
>implementing itself.
Now contracts "hide" things, as well as allowing programs the ability to
"know" things. Its all getting too metaphysical for me.
>If you code against the API, and your code doesn't work correctly, either
>you, or the library, broke the contract.
>And that is quite rare.
Rare, but always possible, and so expecting it to happen might not make
sense, but assuming it cannot happen is entirely bogus. Therefore (if
you don't mind skipping several steps to see my conclusion) a program
can only be written to "an API", yet it is still derivative of *the
library*, if there is only one library that has ever implemented that
API.
--
T. Max Devlin
*** The best way to convince another is
to state your case moderately and
accurately. - Benjamin Franklin ***
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.advocacy.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
