Linux-Advocacy Digest #102, Volume #34 Tue, 1 May 01 17:13:03 EDT
Contents:
Re: Richard Stallman what a tosser, and lies about free software ("Ayende Rahien")
Re: Richard Stallman what a tosser, and lies about free software ("Ayende Rahien")
Re: Richard Stallman what a tosser, and lies about free software ("Ayende Rahien")
Re: Bill Hudson admits that he, Dave Casey, V-man and Redc1c4 (WesTralia)
Re: Another Windows pc gets Linux ("Richard J. Donovan")
Re: IE ("Ayende Rahien")
Re: Winvocates confuse me - d'oh! ("Ayende Rahien")
Re: Is StarOffice 5.2 "compatible" w/MS Office 97/2000? (Fred K Ollinger)
Re: Is StarOffice 5.2 "compatible" w/MS Office 97/2000? (Fred K Ollinger)
----------------------------------------------------------------------------
From: "Ayende Rahien" <[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
Date: Tue, 1 May 2001 23:50:16 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in comp.os.linux.advocacy on Tue, 1 May 2001 05:51:57
> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> Said Ayende Rahien in comp.os.linux.advocacy on Mon, 30 Apr 2001
> >> [...]
> >> >No, you claim that the implentation of API is important to developers
who
> >> >code against this API, right?
> >>
> >> No. What gave you that idea?
> >
> >Qoutes:
> >
> >"designing the API and writing the code to support it are not two
> >independent things."
>
> An abstraction error on your part. Just because the designing and the
> writing are two "independent" things doesn't mean the API and the code
> are the same "thing". They are and they are not, depending on how you
> are using the terms. Sorry; I didn't invent the rules of language, I
> just try to understand and use them.
API is about designing, you design your API according to your application's
needs, when you've the API, *then* you start coding.
You are currect about the language, though, you are misusing the
programmers' dialect, you use incorrect terms constantly, which appears to
be largely
> "The API" is important to program developers, as is therefore "the code"
> which implements it. How the code "works" doesn't matter, but then
> again, neither does how the API "works". Just whether it does, THAT it
> does.
No, the API is the only thing that matters.
The API in about *interface*, a way to communicate with other part of the
the program, it has nothing to do with the code.
Again, you seem to be confusing basic stuff here.
API is a declaration, a way of stating something.
In this case, your application's interface.
You say, "You'll give me money, I'll give you a book"
Code is the action.
You accept the money, count it, go to where you can get the book, and give
it to whomever gave you the money.
The one who gave you the money doesn't know if you have a stock of copies of
the book at home or if you went to a bookstore and bought it from there.
One important point in API is to limit to a minimum the amount of data you
disclouse.
> [...continued catalog of Ayende's mistaken understanding of my words
> snipped...]
>
> [...]
> >> I know enough about programming to know that you can't write a program
> >> that uses a library that doesn't yet exist and expect it to work just
> >> because it matched a draft of the API documentation.
> >
> >Not by a long shot.
> >You can't make an exectuable of a program that relies on a
not-yet-existing
> >library, but you can certainly program against it.
>
> In theory, maybe. In practice? The API/library/code has to "exist"
> somehow for a programmer to "use" it. If you want to claim someone is
> using something that doesn't exist, I can understand the idea
> metaphorically. But analytically it is an empty concept.
No, in practice, API calls that doesn't exist are used all the time.
They need to exist only when you are ready to compile the code.
> >In Ada, you can *compile* against a spec, and then you write the spec's
> >body, and you make an executable out of the whole thing.
>
> Sounds neat, though obviously I don't really understand what you mean.
> I think you're just confusing coding for "spec", as mandated by the
> terminology used in ADA, but I could be entirely wrong.
In Ada, you write like this:
demo.ads (ada spec)
package demo is
-- all sort of stuff
end demo;
demo.adb (ada body)
package body demo is
-- al sort of stuff
end demo;
The ads file is the package API, the adb file is the implementation.
> >T. Max, macros doesn't even come close to any degree of disipline
> >programming.
>
> Sorry; they're procedures which must be properly coded to function.
> That's programming, though of a notoriously simple sort. I don't know
> or care what "typing" means, and I'm sure that's enough for you to
> convince yourself I therefore don't have a valid understanding of
> whether a program is 'derivative' of a library. But, hey, I'm not the
> one that made that case; I'm just pointing out why it hasn't yet been
> successfully refuted.
Do you have inter-relaying procedures? Did you've tried working in
multi-programmer enviroment? Did you *design* this?
> >Consider this statement:
> >"Likewise, a library must be written before its API can be correctly
> >described."
>
> Note the importance of the term "correctly". I didn't put it there by
> accident.
An API which is tied to its implementation is flawed, period.
If you need the library to describe the API then there is something
seriously wrong here.
Again, this in an API:
[function return type] [function name] [function parameters] [function
action]
Different languages has different ways to represent this, but I think that
most coders can read this.
An API that must wait its implementation is not one that was correctly
designed, or designed at all.
You write the API *first*, then the implementation.
> >T. Max, a spec ( a class interface, a set of function, whatever you want
to
> >call it, it's an API), in any place where programming is a serious
bussiness
> >is not something that you throw together in half an hour over a cup of
> >coffee.
>
> Indeed; because it requires an intuitive "design" process which occurs
> 'unconsciously', we might say, which is effectively "coding the
> library", in your head, though very loosely. You will have some idea of
> how you COULD code the library, even if you haven't chosen at all how
> you WILL code the library, or the spec would, quite simply, suck.
Again, you are making basic mistakes in terminlogy, not to mention that you
suddenly seem to think that merely thinking about a problem means that you
solved it.
Naturally, when you design an API, you've an idea about how you would
implement it, no, the two are not connected to each other.
You might run into an obstacle in the implementation, and have to change it,
(performance, memory size, etc), the API doesn't change to reflect that.
The library isn't written yet, period.
The API is already defined correctly.
If I build a phone book application, I can hold all the data in an array,
linked list, binary tree, or any of the other dozens of data structures that
exist out there.
When I'm writing the API to think phone book application, I might have an
idea about what I'll use, but that can change, and drastically.
That doesn't impact the "correctness" of the API.
> I would not go so far as to say I could provide a library specification;
> I don't know enough about programming. Thus, designing your 'spec' and
> writing your 'code' both are, and are not, the same thing.
Not at all, see above for explanation and an example, not to mention that
that isn't what you claimed in the first place.
> >Writing a spec can, and often does, take longer than writing the code
> >itself.
>
> Let me guess; the more time you (intelligently) spend writing the spec,
> the less time it takes to code, right?
Yes, because you've a picture of the program's *layout*, instead of having
to pigure it out on your own.
And no, that has nothing to do with the implementation.
> >Writing a spec is designing the API, which I feel is the hardest part in
> >programming.
>
> Well, if its "part of programming", doesn't that make it the same as
> "writing code", since that's what makes it "programming"?
No, it's not, anymore than drawing a picture of a building is a part of
engineering it.
Sure, you may refer to the picture to do the general stuff, but the
engineering and the arcitecting part of it are completely seperated.
Likewise in programming.
I code when I "engineer", I design when I "arictect".
I don't consider the time when I write down the design of a program as
coding.
Programming isn't all about coding.
> >So, no, nothing has to be written before an API can be correctly
described.
>
> "Written", maybe not, so I'll say "coded", but not written down. Very
> loosely coded, as you've illustrated, because you might use either an
> array or a tree or whatever for certain parts. But the library "exists"
> already, and so a program which presupposes this existence is
> "derivative", if it uses the library's code in its execution (regardless
> of which 'copy' of the library code it might require.)
Written is equal to coded.
And what you call library coded very loosely is the library's API.
Do you *now* get it?
What I have already is the interface of the library, not it's code, that is
called, strangely enough, API!
> >I'll try give an example.
> [...gedanken experiment ignored...]
> >I can assure you that *no one* is working in the way you described.
>
> I have not described any way for anyone to work; I've pointed out that
> this is the way programming always "works", according to the use of the
> terms we've got, like 'library', 'code', 'program', 'develop', 'copy',
> 'software', 'derivative', etc.... Every one of them is a problem,
> unless you're talking to another programmer who will take on faith your
> usage. Once you start using these terms outside the technical world,
> though, and start talking to laymen and lawyers and judges, you assume
> that the "meanings" of each are just as fixed, and they are not. They
> are abstractions, not labels for metaphysical objects or
> characteristics.
Correct, and you are a laymen in computer programming, so you use the terms
incorrectly.
According to your own words, you just spent a whole lot of effort arguing
about something just because you failed to understand a technical term.
> >T. Max, there are a number of people around here who program for a
*living*,
> >we *know* what we are saying. You don't, period.
>
> No, I'm sorry, you are entirely mistaken. I do understand what you are
> saying. I just don't agree with it. That might be enough to pop the
> circuit breakers in a programmer's head, as you guys are pretty tightly
> wired. You have to be, to do what you do, which I certainly am not
> capable of doing. I can assure you that your impression that I do not
> "understand" it, simply because I am not capable of doing it, is flawed.
Part of understanding is doing it. You can read a whole book about
programming, and you wouldn't know how to program until you sat down and
wrote a lot of your own code.
> As for not knowing what *I* am talking about, that is a problem you
> have; I understand what I'm saying perfectly, and I will again assure
> you that it is perfectly reasonable, though you might still disagree
> with it, if you bothered to try to understand it. Once you understand
> it, you may be surprised to find you agree with it. You might not, but
> at least you'd be able to use the abstractions competently to explain
> what is wrong with my theory.
You use CS terminology incorrectly, which caused a lot of confustion because
you seemed to be unabled to grasp what API *is*.
------------------------------
From: "Ayende Rahien" <[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
Date: Tue, 1 May 2001 23:51:00 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in comp.os.linux.advocacy on Mon, 30 Apr 2001
> >"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> Said Stefaan A Eeckels in comp.os.linux.advocacy on Sun, 29 Apr 2001
> >
> >> >Well, one of my colleagues is writing an application to a Java
> >> >.jar that's not yet implemented (I finished the spec, he started
> >> >on his application after about the third draft, when we felt it
> >> >was stable enough). I'll have the classes implemented when he'll
> >> >start testing. Hint: writing a program != coding. There's a lot
> >> >to do before the first line of code is written, or before the
> >> >first test is run.
> >>
> >> That's like saying "writing a book != authoring", and illustrates
> >> clearly why everyone gets so confused by software copyright.
> >
> >Here is a perfectly legal API:
> [...]
>
> My consideration regards real APIs, not "legal" ones, or any other form
> of thought experiment.
That is a real API.
It's just expressed in a language neutral language.
------------------------------
From: "Ayende Rahien" <[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
Date: Tue, 1 May 2001 23:51:57 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in comp.os.linux.advocacy on Tue, 1 May 2001 11:58:39
> +0200;
> >
> >"Isaac" <[EMAIL PROTECTED]> wrote in message
> >news:[EMAIL PROTECTED]...
> >> On Tue, 1 May 2001 05:51:57 +0200, Ayende Rahien <[EMAIL PROTECTED]> wrote:
> >> >
> >> >Qoutes:
> >> >
> >>
> >> < quotes snipped >
> >>
> >> Man is that stuff funny when it's all collected together like that.
> >> I note that you addressed the quotes point by point, but I'm not
> >> sure that was necessary. You don't need a comedian when the
> >> straight man is this good.
> >
> >I'd a similar discussion with him about the usefulness of following MS'
> >recommendations about storing user info in the registry, as well as
several
> >other things related to this, several months ago.
>
> You're going for troll points here, Ayende, you realize that, don't you?
>
> >I'd wrote a program that he claimed was impossible, as well as explain
him
> >(several times) that putting user related stuff in HKLM is not good, he
> >somehow managed to make every sentence into MS-makes-crapware arguement.
> >At least this time he is only ignorant, and not biased.
>
> You utterly failed to write a usable program for managing file types,
> mime types, and associations. You hacked together a couple of function
> calls, and figured the matter done, which was a pathetic sham, as I
> never said this was not exceptionally easy. Just pathetically useless.
That wasn't the requirement, remember?
<Qoute url="http://groups.google.com/groups?oi=djq&ic=1&selm=an_699310481";>
How about you deliver a "simple
application" that does *both* associate multiple extensions with an
existing file type or a new file type, regardless of whether those
extension are registered with another existing file type. We'll see how
non-simple it is, eh?
</Qoute>
------------------------------
From: WesTralia <[EMAIL PROTECTED]>
Crossposted-To:
alt.destroy.microsoft,us.military.army,soc.singles,soc.men,misc.survivalism,alt.fan.rush-limbaugh
Subject: Re: Bill Hudson admits that he, Dave Casey, V-man and Redc1c4
Date: Tue, 01 May 2001 15:51:48 -0500
"Aaron R. Kulkis" wrote:
>
> "T. Max Devlin" wrote:
> >
> > Said Aaron R. Kulkis in alt.destroy.microsoft on Tue, 01 May 2001
> > >chrisv wrote:
> > >> "Aaron R. Kulkis" <[EMAIL PROTECTED]> wrote:
> > >> >billh wrote:
> > >> >> "Aaron R. Kulkis"
> > >> >>
> > >> >> > That which supports strategic goals as opposed to tactical goals.
> > >> >>
> > >> >> LOL!!! You've out done yourself, wannabe.
> > >> >
> > >> >Then, Bill, I'm sure you can provide us with a better concise
> > >> >definition of "strategic," right, asshole.
> > >>
> > >> Once again Kookis resorts to his "I'll make the other guy explain it,
> > >> then say 'yep that's right' when he's done" tactic, when his BLUFF has
> > >> been CALLED!
> > >
> > >He's claimin my definition is wrong, so the burden of providing a
> > >better defininition is upon him.
> >
> > Bwah-ha-ha-ha-ha. That's the same tactic you used last time, when your
> > bluff was called. And the time before that, and the time before that,
> > and the time before that.
> >
> > You know, Aaron, if you killed the attitude, killed the sig, and tried
> > to avoid anti-trolling, you might well be an intelligent and reasonable
> > person, I think, even if some of your ideas are kind of goofy (more so
> > even than their conservative roots). As it stands, you're just a troll
> > who happens to hate Microsoft, no different in any other respect from
> > the sock puppets or windroids.
> >
>
> Actually in my spare time I am a prima ballerina, NASCAR driver, astronaut,
> barbie doll collector, professional basketball player, map maker, spelling
> bee champ, President of the United States, overweight talk show host, aluminum
> can collector, war-hero, and food-stamp recipient.
>
This I may believe for once.
--
------------------------------
From: "Richard J. Donovan" <[EMAIL PROTECTED]>
Subject: Re: Another Windows pc gets Linux
Date: Tue, 01 May 2001 20:58:46 GMT
>
> It shines on the really large documents, too. Did a 1500 page one a while
> back. (Word2000 choked miserably on it. Tables screwed up and headers
> mangled after every repagnation. Even WordPerfect8 for Linux did a better
> job than Word.)
WP 8 for Linux, specifically, WP 8.1 as in the Corel 1.1 distro IIRC, is
a great program. But these posts encourage me to try such further Linux
applications.
------------------------------
From: "Ayende Rahien" <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.destroy.microsoft
Subject: Re: IE
Date: Tue, 1 May 2001 23:54:26 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Chad Everett in alt.destroy.microsoft on 1 May 2001 11:18:42 -0500;
> >On Tue, 1 May 2001 16:38:01 +0100, Michael Pye <[EMAIL PROTECTED]>
> wrote:
> >>
> >>"T. Max Devlin" <[EMAIL PROTECTED]> wrote
> >>> No. Seriously. Honestly. Chalk it up, if you want, to the fact that
> >>> I've never been spoiled by something better. I can certainly imagine
a
> >>> browser which is less clumsy, "better presented" (whatever that was
> >>> supposed to mean), lower on memory use or quicker to load. Its just
> >>> that I have never found any. Even after I looked. Seriously.
> >>> Honestly.
> >>
> >>Well, I find IE is most of those for Windows. I may not like the
authors,
> >>but it is undoubtably a better browser...
> >
> >If you run it on Windows. Ever try the Solaris version of IE? What a
joke.
>
> That's what happens when you have to port the registry to Unix. Ugh.
Wine did it quite successfully, I believe.
------------------------------
From: "Ayende Rahien" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.nt.advocacy
Subject: Re: Winvocates confuse me - d'oh!
Date: Tue, 1 May 2001 23:57:15 +0200
"T. Max Devlin" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Said Ayende Rahien in comp.os.linux.advocacy on Tue, 1 May 2001 01:24:11
> >As a side note:
> >C:\Documents and Settings\Ayende>debug
> >-f 0:0 ffff 5
> >
> >C:\Documents and Settings\Ayende>
> >
> >It didn't even crash cmd.exe :-)
>
> No doubt a hacked patch, rather than a fix to the fundamental flaws
> embedded in Windows, due to its historical development as a DOS
> extender.
Please notice the path & the application name, and don't display you
ignorance.
This is not a DOS extender, this is an NT system, which has no DOS roots at
all.
------------------------------
From: [EMAIL PROTECTED] (Fred K Ollinger)
Crossposted-To: comp.unix.advocacy,alt.solaris.x86,comp.unix.solaris
Subject: Re: Is StarOffice 5.2 "compatible" w/MS Office 97/2000?
Date: 1 May 2001 20:59:10 GMT
jim ([EMAIL PROTECTED]) wrote:
: "Ivo" <[EMAIL PROTECTED]> wrote in message
: news:9c44vf$79v$[EMAIL PROTECTED]...
: > What Can I say after all those discussions:
: > Use MS Windows & Word AND OF course
: > enjoy the ALL 65000 viruses (Probably more).
I enjoy virii. :)
: OK and all the Secretaries and people who do the REAL typing on desktops
: that NEVER get on the internet (eg AIRGAP) will absolutely FLIP for
: STAROFFICE! {snigger} not in a while......
I hated star office, too much like MS office which is why I dumped the whole
pile of crap.
: > Personally for me - I love Unix (Solaris/FreeBSD) and linux
: > And Staroffice is enought for me to edit docs.
: > For e-mail - use HTML, plain text -
: I think VI and all the useless other editors are an excuse for sad little
: 40-50 yr old UNIX fat tards like yourself to keep whatever "knowledge" you
Useless is in the eye of the beholder. I use vi for _everything_. I'm
writing a novel in it. I learned it when I tried to use MS word and it
_failed_. So for me MS word is useless. The only diff is that you either
paid for Word or your employer did or you _stole_ it. So you either got
ripped off or are a thief if you use word. My $200 (or whatever word is now)
is going into better hardware.
: have and NOT share it with anyone - problem is, with a click of a mouse your
: Gramma can send an email to her friend - and guess what - UNIX didn't do it!
Who cares? And hey, what about that alleged BSD code for networking. Or
are you coveniently forgetting that again?
: A smelly Finn didn't do it! It took a man that is richer than your dreams -
Who are you to call people smelly? Rude and has nothing to do with the topic.
: it took a man that would tell you who was BOSS! Oh, and btw I use plain
: text, html and the next best thing!
Gates is not my boss. Maybe he is yours? Defending someone who overcharges,
I do hope you work for him, either way, if you defend someone like that then
they own you.
Fred
------------------------------
From: [EMAIL PROTECTED] (Fred K Ollinger)
Crossposted-To: comp.unix.advocacy,alt.solaris.x86,comp.unix.solaris
Subject: Re: Is StarOffice 5.2 "compatible" w/MS Office 97/2000?
Date: 1 May 2001 21:03:20 GMT
Vid Strpic ([EMAIL PROTECTED]) wrote:
: Joerg Schilling said unto us in comp.unix.advocacy :
: > In article <[EMAIL PROTECTED]>,
: > Rich Teer <[EMAIL PROTECTED]> wrote:
: >
: >>UNIX does have GUI email clients too, ya know. Probably got there before
: >>Windoze, too.
: >
: > I know Sun's mailtool (GUI) since 1985, so it _was_ there in 1984!
: >
: > What did M$ this thime ?
: M$ Windows 1.0 dates somewhere in 1985., so ... ;)))
And they had a gui email client for windows 1.0? Reference please. How many
were installed on grandmother's computers by this time?
Not that it's a good idea. I think gui
sucks for this purpose, there's no need to have to use a mouse to do email.
It's slower, need more system resources and is more cumbersome. Also more
painful on the eyes, I like the black screen.
Fred
: Dogs come when you call. Cats have answering machines.
: --
: ))
: C|~~| Vid Strpic, IRC:*@Martin, [EMAIL PROTECTED], /bin/zsh.
: `--' (I don't speak for my employer, just for myself.)
: C|N>K UNIX fundamentalist - and an average chauvinistic male.
------------------------------
** 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
******************************
