Tools That Big TIme Developes User
Steve
the tools they use:
Dave Pearson
> Dumb article, but it has quotes from big time linux developers about the
> tools they use:
>
> http://localhost:8080/zeera/webapps/isp/index_home.jsp
But I don't have anything listening on that port on my machine.
--
Dave Pearson: | lbdb.el - LBDB interface.
http://www.davep.org/ | sawfish.el - Sawfish mode.
Emacs: | uptimes.el - Record emacs uptimes.
http://www.davep.org/emacs/ | quickurl.el - Recall lists of URLs.
Steve
> Dumb article, but it has quotes from big time linux developers about
> the tools they use:
>
> http://localhost:8080/zeera/webapps/isp/index_home.jsp
The pitfalls of tabbed browsing. Now, for the url of the article:
http://www.linuxworld.com/linuxworld/lw-1999-06/lw-06-vcontrol_1.html
Per Abrahamsen
> The pitfalls of tabbed browsing. Now, for the url of the article:
> http://www.linuxworld.com/linuxworld/lw-1999-06/lw-06-vcontrol_1.html
Apparently, Alan Cox doesn't use an editor at all.
cat > file.c
I can respect that. If it is human to error, I guess Alan have no
need of an editor.
Derrell...@unwireduniverse.com
> steves...@yahoo.com (Steve) writes:
>
> Apparently, Alan Cox doesn't use an editor at all.
>
> cat > file.c
Yes, but we real Gurus don't bother with that; it's faster to avoid that step
completely and just do
cat > file.o
If, in the unlikely case that an error is made, we can always use hexl-mode to
correct it, if adb isn't available. :-)
Derrell
David Kastrup
I find this extremely inconvenient, in particular entering the
debugging symbol in .o files. I prefer they way you type them into
a.out. With the advent of dynamic libraries, this has become much
more feasible.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
Email: David....@t-online.de
Simon Josefsson
I don't think you could release the result as GPL though, as I
remember it required that the "source" that is released is in a form
that is convenient for programmers to work on.
Hm. I wonder if I can GPL my Intercal hacks.
David Kastrup
Sigh. I don't know why everyone considers it necessary to spread
rumors, lies and misinformation about the GPL when it is so easy to
just read it. It is not that it is kept secret anywhere.
Here are relevant extracts:
1. You may copy and distribute verbatim copies of the Program's
source code as you receive it, [...]
Key phrase: "as you receive it".
The definition of "source" code only applies to section
3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the
following: [...]
which is a special exception for the case the work is _not_
redistributed as received. In that case, we have
The source code for a work means the preferred form of the work
for making modifications to it.
Of course, the preferred form in this case is the file itself since
nothing else exists.
Anyhow, this is all missing the point. You stated: "I don't think you
could release the result as GPL though, as I remember it required that
the "source" that is released is in a form that is convenient for
programmers to work on." and this is a brazen lie. Of _course_ I can
release _anything_ that I wrote myself under any license I desire. I
can _perfectly_ well release a binary-only program under the GPL, if
I am in the mood. Of course, the recipients will not be allowed to
further redistribute their binaries, since they can't comply with the
GPL's terms. So it would not be more than a marketing gag, mostly,
but entirely possible.
> Hm. I wonder if I can GPL my Intercal hacks.
Why not?
Simon Josefsson
> The definition of "source" code only applies to section
>
> 3. You may copy and distribute the Program (or a work based on it,
> under Section 2) in object code or executable form under the terms of
> Sections 1 and 2 above provided that you also do one of the
> following: [...]
>
> which is a special exception for the case the work is _not_
> redistributed as received. In that case, we have
>
> The source code for a work means the preferred form of the work
> for making modifications to it.
>
> Of course, the preferred form in this case is the file itself since
> nothing else exists.
How do you know if there is source code or not? Consider someone
distributing binaries only, without source code, under GPL, and claims
there is no source code. Is this OK?
> Anyhow, this is all missing the point. You stated: "I don't think you
> could release the result as GPL though, as I remember it required that
> the "source" that is released is in a form that is convenient for
> programmers to work on." and this is a brazen lie. Of _course_ I can
> release _anything_ that I wrote myself under any license I desire.
Unless you are violating the license when doing so. E.g., you can't
release code you wrote under a non-GPL-compatible license if your code
uses GPLd code.
> I can _perfectly_ well release a binary-only program under the GPL,
> if I am in the mood. Of course, the recipients will not be allowed
> to further redistribute their binaries, since they can't comply with
> the GPL's terms. So it would not be more than a marketing gag,
> mostly, but entirely possible.
>
>> Hm. I wonder if I can GPL my Intercal hacks.
>
> Why not?
Because the source code for my work is not in a preferred form for
making modifications to it, thus violating the license. I guess the
problem is the word "preferred", because 1) there is no objectively
"preferred" form of something and 2) it is not clear whom (publisher
or user) decides what the preferred form or not.
It seems the GPL assumes there is source code for something, which
isn't always true.
(Isn't it wonderful how you can make almost any discussion into a
license flamewar? I wonder why this is so.)
David Kastrup
> David Kastrup <David....@t-online.de> writes:
>
> > The definition of "source" code only applies to section
> >
> > 3. You may copy and distribute the Program (or a work based on it,
> > under Section 2) in object code or executable form under the terms of
> > Sections 1 and 2 above provided that you also do one of the
> > following: [...]
> >
> > which is a special exception for the case the work is _not_
> > redistributed as received. In that case, we have
> >
> > The source code for a work means the preferred form of the work
> > for making modifications to it.
> >
> > Of course, the preferred form in this case is the file itself since
> > nothing else exists.
>
> How do you know if there is source code or not? Consider someone
> distributing binaries only, without source code, under GPL, and claims
> there is no source code. Is this OK?
Sure, as long as everything was written by himself and he is not
violating the license of somebody else. See below.
> > Anyhow, this is all missing the point. You stated: "I don't think you
> > could release the result as GPL though, as I remember it required that
> > the "source" that is released is in a form that is convenient for
> > programmers to work on." and this is a brazen lie. Of _course_ I can
> > release _anything_ that I wrote myself under any license I desire.
>
> Unless you are violating the license when doing so. E.g., you can't
> release code you wrote under a non-GPL-compatible license if your code
> uses GPLd code.
In which case it is not your code. If you use the code from someone
else, you need to comply with the license under which you received the
code. It is up to the person which licensed this code to you how he
interprets this license, and it is up to the respective legal system
to find whether his interpretation applies in case you and the
licensor disagree over the meaning of the license.
> > I can _perfectly_ well release a binary-only program under the GPL,
> > if I am in the mood. Of course, the recipients will not be allowed
> > to further redistribute their binaries, since they can't comply with
> > the GPL's terms. So it would not be more than a marketing gag,
> > mostly, but entirely possible.
> >
> >> Hm. I wonder if I can GPL my Intercal hacks.
> >
> > Why not?
>
> Because the source code for my work is not in a preferred form for
> making modifications to it, thus violating the license. I guess the
> problem is the word "preferred", because 1) there is no objectively
> "preferred" form of something and 2) it is not clear whom (publisher
> or user) decides what the preferred form or not.
Rubbish. If there is no other form in existence, your hacks _are_ the
preferred (and sole) form of modification. Of course the only
relevant opinion about that is that of the licensor, and in case he
wants to press it, the court's interpretation of scope and meaning of
the license.
> It seems the GPL assumes there is source code for something, which
> isn't always true.
Problems with your reading comprehension?
> (Isn't it wonderful how you can make almost any discussion into a
> license flamewar? I wonder why this is so.)
Because some people would rather start an uninformed flame war rather
than to read the license in question and engage their brain?
Kai Großjohann
> The source code for a work means the preferred form of the work
> for making modifications to it.
>
> Of course, the preferred form in this case is the file itself since
> nothing else exists.
Well, a.out being the only format doesn't make it the preferred
format.
I think that you can't release a Cobol program under the GPL since
nobody in their right mind would prefer Cobol as a format to work
on. Instead, it should be rewritten in Perl or something.
kai
--
A large number of young women don't trust men with beards. (BFBS Radio)
David Kastrup
> David Kastrup <David....@t-online.de> writes:
>
> > The source code for a work means the preferred form of the work
> > for making modifications to it.
> >
> > Of course, the preferred form in this case is the file itself since
> > nothing else exists.
>
> Well, a.out being the only format doesn't make it the preferred
> format.
>
> I think that you can't release a Cobol program under the GPL since
> nobody in their right mind would prefer Cobol as a format to work
> on. Instead, it should be rewritten in Perl or something.
Could you please stop spouting nonsense? I can release my own stuff
under any license I bloody want. I am licensing the stuff, period.
And it is entirely up to me how I interpret or enforce this license.
And only if there is any disagreement between my interpretation and
that of a licensee, an additional opinion gets important: that of the
courts.
Dave Pearson
> I think that you can't release a Cobol program under the GPL since nobody
> in their right mind would prefer Cobol as a format to work on. Instead, it
> should be rewritten in Perl or something.
A case of "Worse is Better"?
Geoff Raye
>Kai.Gro...@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
>> I think that you can't release a Cobol program under the GPL since
>> nobody in their right mind would prefer Cobol as a format to work
>> on. Instead, it should be rewritten in Perl or something.
>
>Could you please stop spouting nonsense? I can release my own stuff
>under any license I bloody want. I am licensing the stuff, period.
>And it is entirely up to me how I interpret or enforce this license.
>And only if there is any disagreement between my interpretation and
>that of a licensee, an additional opinion gets important: that of the
>courts.
Oh, come on. I thought it was funny.
And on to the more important subjects:
I have not yet moved properly onto the True Path of Emacs with respect
to newsgroup reading. If I'm to do so, I've heard that this Path
has multiple ways to travel. I've heard that there is an option of
GNUS. Furthermore, some seem to suggest that GNUS and Gnus are
different options. Are those the ideal, or should I consider some other
EMACS-based reading?
Thank you,
Geoff
--
Geoff Raye \ All irregularities will be handled by the forces
ge...@raye.com \ controlling each dimension. Transuranic heavy
\ elements may not be used where there is life.
Kai Großjohann
> I have not yet moved properly onto the True Path of Emacs with respect
> to newsgroup reading. If I'm to do so, I've heard that this Path
> has multiple ways to travel. I've heard that there is an option of
> GNUS. Furthermore, some seem to suggest that GNUS and Gnus are
> different options.
There was GNUS 4.x. I don't recommend to use it, it's very old.
Then there was Gnus 5.x. The current version is Gnus 5.9 and comes
with Emacs 21. XEmacs probably has Gnus 5.8.8 which is approximately
the same thing.
Dave Pearson
> Kai.Gro...@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
>
> > I think that you can't release a Cobol program under the GPL since
> > nobody in their right mind would prefer Cobol as a format to work on.
> > Instead, it should be rewritten in Perl or something.
>
> Could you please stop spouting nonsense? [SNIP]
,----[ C-h f with-sense-of-humour-failure RET ]
| with-sense-of-humour-failure is a Lisp macro in `humour'.
| (with-sense-of-humour-failure &rest BODY)
|
| Execute the forms in BODY with your sense of humour in
| failure mode. Sometimes useful for testing purposes, often
| used in error.
`----
David Kastrup
> * David Kastrup <David....@t-online.de>:
>
> > Kai.Gro...@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> >
> > > I think that you can't release a Cobol program under the GPL since
> > > nobody in their right mind would prefer Cobol as a format to work on.
> > > Instead, it should be rewritten in Perl or something.
> >
> > Could you please stop spouting nonsense? [SNIP]
>
> ,----[ C-h f with-sense-of-humour-failure RET ]
> | with-sense-of-humour-failure is a Lisp macro in `humour'.
> | (with-sense-of-humour-failure &rest BODY)
> |
> | Execute the forms in BODY with your sense of humour in
> | failure mode. Sometimes useful for testing purposes, often
> | used in error.
> `----
Problem is that BODY is not ignored, so you should be sure not to
raise any exceptions not handled by the particular clause. In the
above case, this means
s/release/redistribute/
and in particular when keeping in mind what passage Kai actually
quoted before unnecessarily contradicting it.
Kai Großjohann
> | with-sense-of-humour-failure is a Lisp macro in `humour'.
Ha! And there I was, carefully removing gnu.misc.discuss from the
Newsgroups list, but no...
To David: Un' ich sach noch, Wennä! Abba nein, er hört nich.
Lute Kamstra
> Kai.Gro...@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
>
> > I think that you can't release a Cobol program under the GPL since
> > nobody in their right mind would prefer Cobol as a format to work
> > on. Instead, it should be rewritten in Perl or something.
>
> Could you please stop spouting nonsense? I can release my own stuff
> under any license I bloody want. I am licensing the stuff, period.
> And it is entirely up to me how I interpret or enforce this license.
> And only if there is any disagreement between my interpretation and
> that of a licensee, an additional opinion gets important: that of the
> courts.
Not all licenses are enforceable. Legislation supersedes licensing
conditions. For example, I remember a case in Germany where a
Microsoft license prohibited the transfer of (ownership of / the right
to use) software from one person to another. The German courts
overruled this condition, in effect changing the license conditions.
Lute.
--
(error "`~/.signature' too rude")
Nix
> ra...@meow.raye.com (Geoff Raye) writes:
>
>> I have not yet moved properly onto the True Path of Emacs with respect
>> to newsgroup reading. If I'm to do so, I've heard that this Path
>> has multiple ways to travel. I've heard that there is an option of
>> GNUS. Furthermore, some seem to suggest that GNUS and Gnus are
>> different options.
>
> There was GNUS 4.x. I don't recommend to use it, it's very old.
> Then there was Gnus 5.x. The current version is Gnus 5.9 and comes
> with Emacs 21. XEmacs probably has Gnus 5.8.8 which is approximately
> the same thing.
Quite so.
--
`There's something satisfying about killing JWZ over and over again.'
-- 1i, personal communication
Kai Großjohann
> Not all licenses are enforceable. Legislation supersedes licensing
> conditions. For example, I remember a case in Germany where a
> Microsoft license prohibited the transfer of (ownership of / the right
> to use) software from one person to another. The German courts
> overruled this condition, in effect changing the license conditions.
Look at it this way: otherwise the licenses would change law, and we
can't have that.
Shrink-wrapped licenses don't have any effect in Germany because when
you buy a piece of software you make a contract, and it is not
possible for anything that you don't know to be part of the contract.
But you don't see the shrink-wrapped license until after you have
bought the software. Therefore the license does not take effect.
Tim Vanhoof
> David Kastrup <David....@t-online.de> wrote:
> >Kai.Gro...@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> >> I think that you can't release a Cobol program under the GPL since
> >> nobody in their right mind would prefer Cobol as a format to work
> >> on. Instead, it should be rewritten in Perl or something.
> >
> >Could you please stop spouting nonsense? I can release my own stuff
> >under any license I bloody want. I am licensing the stuff, period.
> >And it is entirely up to me how I interpret or enforce this license.
> >And only if there is any disagreement between my interpretation and
> >that of a licensee, an additional opinion gets important: that of the
> >courts.
>
> Oh, come on. I thought it was funny.
>
> And on to the more important subjects:
> I have not yet moved properly onto the True Path of Emacs with respect
> to newsgroup reading. If I'm to do so, I've heard that this Path
> has multiple ways to travel. I've heard that there is an option of
> GNUS. Furthermore, some seem to suggest that GNUS and Gnus are
> different options. Are those the ideal, or should I consider some other
> EMACS-based reading?
Unfortunately, people who have quotes from dodgy 1970s sci-fi shows in
their sigs are not allowed to use Emacs. (see
/usr/share/emacs/etc/FORBIDDEN for details)
Per Abrahamsen
> I find this extremely inconvenient, in particular entering the
> debugging symbol in .o files.
Why bother? Debugging symbols is for code with bugs.