Henning Makholm [EMAIL PROTECTED] writes:
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
The more worrisome case, alas, is the person who *does* keep the .xcf,
but won't distribute it.
No, that is the *easy* case - here we can all agree that the .xcf is
source and must be
On Tuesday, Jul 1, 2003, at 03:12 US/Eastern, Thomas Bushnell, BSG
wrote:
Anthony DeRobertis [EMAIL PROTECTED] writes:
It's true that the GPL wording implies that there is a single
preferred form,
Yep. The GPL was designed for compiled programs, and it shows in
several places.
The
On Tuesday, Jul 1, 2003, at 16:35 US/Eastern, Thomas Bushnell, BSG
wrote:
Brian T. Sniffen [EMAIL PROTECTED] writes:
Nonsense. I edit multiple images into a single image all the time,
but
rarely save an XCF file: multiple layers live in the image-editor's
memory, but never hit the disk.
Brian T. Sniffen [EMAIL PROTECTED] writes:
Nonsense. I edit multiple images into a single image all the time, but
rarely save an XCF file: multiple layers live in the image-editor's
memory, but never hit the disk. There is no persistent form which
represents source any more than there is
Thomas Bushnell, BSG said:
Brian T. Sniffen [EMAIL PROTECTED] writes:
Nonsense. I edit multiple images into a single image all the time,
but rarely save an XCF file: multiple layers live in the
image-editor's memory, but never hit the disk. There is no persistent
form which represents
Brian T. Sniffen [EMAIL PROTECTED] writes:
It certainly does: if there is no persistent form, it isn't the source.
But an xcf is not some weird thing which is not normally persistent.
It is easily and trivially persistent.
Otherwise, the elisp code which is generated (and used, but usually
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
The more worrisome case, alas, is the person who *does* keep the .xcf,
but won't distribute it.
No, that is the *easy* case - here we can all agree that the .xcf is
source and must be distributed.
--
Henning Makholm The
Anthony DeRobertis [EMAIL PROTECTED] writes:
Well, for executable works, we have things like For an executable
work, complete source code means all the source code for all modules
it contains, plus any associated interface definition files, plus the
scripts used to control compilation and
Anthony DeRobertis [EMAIL PROTECTED] writes:
It's true that the GPL wording implies that there is a single
preferred form,
Yep. The GPL was designed for compiled programs, and it shows in
several places.
The relation between a xcf and a gif is precisely one of compilation.
Thomas Bushnell, BSG said:
Anthony DeRobertis [EMAIL PROTECTED] writes:
It's true that the GPL wording implies that there is a single
preferred form,
Yep. The GPL was designed for compiled programs, and it shows in
several places.
The relation between a xcf and a gif is precisely one
On Wednesday, Jun 25, 2003, at 04:21 US/Eastern, Edmund GRIMLEY EVANS
wrote:
I think I would have to distribute both the original source and the
LaTeX source and an explanation of what happened.
You forgot one of the most important parts: The perl script. I'd
suggest distributing the
On Wednesday, Jun 25, 2003, at 11:22 US/Eastern, Joe Moore wrote:
Nick Phillips said:
In the situation where I take a simple GPL'd C program, compile it to
assembler, then hand-optimise the assembler before altering the code,
initially in small ways, eventually completely re-writing the whole
Thomas Bushnell, BSG said:
Joe Moore [EMAIL PROTECTED] writes:
Modifying in an interesting and useful way like running strip on the
binary?
I don't think the GPL can be subverted so easily:
Say, to change a string constant, usually pretty easy. I wish I had
said that the first time around.
On Wednesday, Jun 25, 2003, at 00:16 US/Eastern, Thomas Bushnell, BSG
wrote:
Anthony DeRobertis [EMAIL PROTECTED] writes:
Well, The source code for a work means the preferred form of the work
for making modifications to it. For an executable work, complete
source code means all the source
Anthony DeRobertis [EMAIL PROTECTED] writes:
Well, The source code for a work means the preferred form of the work
for making modifications to it. For an executable work, complete
source code means all the source code for all modules it contains,
plus any associated interface definition
People don't edit program binaries usually, so let's take a realistic
example: I produce a bidirectional bilingual dictionary using some
Perl scripts that automatically generate LaTeX source for both sides
of the dictionary from a marked up version of one direction. However,
just before going to
Thomas Bushnell, BSG said:
It seems to me that if you are right, then there is no way to enforce
the GPL: because then someone could simply modify the object file in
some interesting and useful way (say, to change a string constant,
usually pretty easy), and then claim that the C code isn't
Joe Moore [EMAIL PROTECTED] writes:
Thomas Bushnell, BSG said:
It seems to me that if you are right, then there is no way to enforce
the GPL: because then someone could simply modify the object file in
some interesting and useful way (say, to change a string constant,
usually pretty
Anthony DeRobertis [EMAIL PROTECTED] writes:
On Monday, Jun 23, 2003, at 02:44 US/Eastern, Thomas Bushnell, BSG
wrote:
The reason is quite clear: because otherwise one could very trivially
escape the GPL's requirements entirely, by making some little
modification directly to the
Henning Makholm [EMAIL PROTECTED] writes:
What? Who on earth said it was illegal to edit things other than the
preferred form?
The GPL says that. It says that you mustn't distribute anything
without giving the preferred form for it, and if we go by your premise
that gifs are not the
Anthony DeRobertis [EMAIL PROTECTED] writes:
What I can't do is edit the image with layers in xcf, then flatten it
and call that source. Or call the unedited assembly output of gcc
source.
NOR can you call the edited assembly the complete source. If you
could, the GPL would be pointless.
Henning Makholm [EMAIL PROTECTED] writes:
The GPL'ed source contains ugly xpm's that upstream created pixel for
pixel in Emacs because he knew no better and thought he was only
making a proof-of-concept implementation anyway. I import the xpm into
the Gimp, painstakingly separate the raw
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Henning Makholm [EMAIL PROTECTED] writes:
The history you were considering was a .xcf (or the like), which
someone then modified a few parts of in its gif output.
A few parts was never in the history I am talking about. Someone
distributed a
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Henning Makholm [EMAIL PROTECTED] writes:
The GPL'ed source contains ugly xpm's that upstream created pixel for
pixel in Emacs because he knew no better and thought he was only
making a proof-of-concept implementation anyway. I import the
* Henning Makholm [EMAIL PROTECTED] [030624 13:19]:
The GPL'ed source contains ugly xpm's that upstream created pixel for
pixel in Emacs because he knew no better and thought he was only
making a proof-of-concept implementation anyway. I import the xpm into
the Gimp, painstakingly
Henning Makholm a dit:
Do you mean by that that if I use an editor that does not have a
save format that losslessly reproduces all of its internal state,
then I can only distribute the output under the GPL if I also ship a
revivable core dump of the editor?
Well, xcf does not reproduces all
On Tuesday, Jun 24, 2003, at 02:16 US/Eastern, Thomas Bushnell, BSG
wrote:
NOR can you call the edited assembly the complete source.
I disagree there. Let me make it clear that I'm taking about making
major changes: so much so that the executable is substantially
different from the
Anthony DeRobertis [EMAIL PROTECTED] writes:
I don't think an interpretation of the GPL that says I wrote this
code in C. Forever is C must it stay! is correct.
Right. All I'm saying is you must distribute the C code; I don't care
whether you continue to make changes in that language.
How
Henning Makholm [EMAIL PROTECTED] writes:
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Henning Makholm [EMAIL PROTECTED] writes:
The history you were considering was a .xcf (or the like), which
someone then modified a few parts of in its gif output.
A few parts was never in the
Henning Makholm [EMAIL PROTECTED] writes:
Do you mean by that that if I use an editor that does not have a
save format that losslessly reproduces all of its internal state,
then I can only distribute the output under the GPL if I also ship a
revivable core dump of the editor?
Is there such
On Tuesday, Jun 24, 2003, at 13:29 US/Eastern, Thomas Bushnell, BSG
wrote:
Anthony DeRobertis [EMAIL PROTECTED] writes:
I don't think an interpretation of the GPL that says I wrote this
code in C. Forever is C must it stay! is correct.
Right. All I'm saying is you must distribute the C
On Tuesday, Jun 24, 2003, at 13:30 US/Eastern, Thomas Bushnell, BSG
wrote:
Henning Makholm [EMAIL PROTECTED] writes:
Do you mean by that that if I use an editor that does not have a
save format that losslessly reproduces all of its internal state,
then I can only distribute the output
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes:
Precisely when the xcf is the exact source of the actual gif in
question. If the gif has been modified on its own, then the source is
now the combination of both the xcf and the gif.
Would you agree that there could come a point where the gif
Anthony DeRobertis [EMAIL PROTECTED] writes:
On Tuesday, Jun 24, 2003, at 13:30 US/Eastern, Thomas Bushnell, BSG
wrote:
Henning Makholm [EMAIL PROTECTED] writes:
Do you mean by that that if I use an editor that does not have a
save format that losslessly reproduces all of its internal
Anthony DeRobertis [EMAIL PROTECTED] writes:
On Tuesday, Jun 24, 2003, at 13:29 US/Eastern, Thomas Bushnell, BSG
wrote:
Anthony DeRobertis [EMAIL PROTECTED] writes:
I don't think an interpretation of the GPL that says I wrote this
code in C. Forever is C must it stay! is correct.
Jeremy Hankins [EMAIL PROTECTED] writes:
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes:
Precisely when the xcf is the exact source of the actual gif in
question. If the gif has been modified on its own, then the source is
now the combination of both the xcf and the gif.
Would you
On Tuesday, Jun 24, 2003, at 16:37 US/Eastern, Thomas Bushnell, BSG
wrote:
Yes, there might be such a case, but I would say that a few edits
isn't such a case. And that the usual scenario isn't this at all;
it's people who simply throw away the xcf or outright refuse to
distribute it.
On Tuesday, Jun 24, 2003, at 16:36 US/Eastern, Thomas Bushnell, BSG
wrote:
Why would C stay the preferred form for modifying a work for eternity,
even when the current work bares hardly a resemblence to its C
original?
It is *PART* of the source. Not the whole source, but part of it.
Henning Makholm [EMAIL PROTECTED] writes:
In such a case, the layered format is the preferred form,
Perhaps for you. Not for everybody.
No, for everybody: for the simple reason that if you have distributed
that, then the raw pixels of the gif are still accessible and can be
edited. I wish
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Henning Makholm [EMAIL PROTECTED] writes:
for the simple reason that if you have distributed that, then the
raw pixels of the gif are still accessible and can be edited,
But if they are not the preferred form, it is illegal to edit them
On Sunday, Jun 22, 2003, at 08:11 US/Eastern, Henning Makholm wrote:
But if they are not the preferred form, it is illegal to edit them (at
least, it it illegal to distribute the edited gifs). So what's the
point of being *able* to do so?
If you merge the layers of an image, then edit
On Monday, Jun 23, 2003, at 02:44 US/Eastern, Thomas Bushnell, BSG
wrote:
The reason is quite clear: because otherwise one could very trivially
escape the GPL's requirements entirely, by making some little
modification directly to the binary for some program, and then
claiming that the binary
Scripsit Anthony DeRobertis [EMAIL PROTECTED]
On Sunday, Jun 22, 2003, at 08:11 US/Eastern, Henning Makholm wrote:
But if they are not the preferred form, it is illegal to edit them (at
least, it it illegal to distribute the edited gifs).
If you merge the layers of an image, then edit
On Fri, Jun 20, 2003 at 02:03:30PM -0500, Steve Langasek wrote:
SL I really don't think that the form that contains the *most*
SL information is necessarily the best, because this prevents someone
SL from improving the source by removing *extraneous* information. If
SL two forms of source
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Henning Makholm [EMAIL PROTECTED] writes:
Some people would prefer to edit a layered format,
because it gives them better control and allow them to change the
semantic contents of the image without losing any overlaid effects.
Other
Florian Weimer [EMAIL PROTECTED] writes:
Sam Hartman [EMAIL PROTECTED] writes:
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in a particular instance what
the preferred form for modifications was?
I know of one Debian
Jeremy Hankins [EMAIL PROTECTED] writes:
The difference is that a gif is a lot richer than binary software, in
the sense of humans being able to do stuff with it.
This is a real difference.
So we have two ways of distinguishing between binary and source:
Way one says that binaries cannot be
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Way two says that source is the preferred form for modification, and
binaries are not.
Yes, but it is not unambiguous what this would result in in the case
of graphics. Some people would prefer to edit a layered format,
because it gives them
Henning Makholm [EMAIL PROTECTED] writes:
Scripsit [EMAIL PROTECTED] (Thomas Bushnell, BSG)
Way two says that source is the preferred form for modification, and
binaries are not.
Yes, but it is not unambiguous what this would result in in the case
of graphics. Some people would prefer
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes:
Again, again, again, I'm not interested here in the definition of
free or proprietary; just with the copyleft. In the context of
the copyleft, if you destroy the source, the object code does not
somehow mutate into source, and as a result the
On Tue, Jun 17, 2003 at 02:47:58PM +0200, Thomas Hood wrote:
Objection #1: The license must not force the licensee to keep around
old crufty versions of the source.
Answer: Using my definition, it doesn't. The licensee is only required
to provide the most informative form at his disposal.
Sam Hartman [EMAIL PROTECTED] writes:
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in a particular instance what
the preferred form for modifications was?
I know of one Debian package whose source code is the output of a
CASE
On Wed, Jun 18, 2003 at 02:19:12PM -0700, Thomas Bushnell, BSG wrote:
snip
I'd say this is a pretty strong argument for why we shouldn't talk
about preferred forms at all in our definition of freedom.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ | Dept.
On Thu, Jun 19, 2003 at 09:41:55AM +0100, Andrew Suffield wrote:
On Wed, Jun 18, 2003 at 02:19:12PM -0700, Thomas Bushnell, BSG wrote:
snip
I'd say this is a pretty strong argument for why we shouldn't talk
about preferred forms at all in our definition of freedom.
I agree. It's just too
Mark Rafn [EMAIL PROTECTED] writes:
There are a number of icons and images in
products whose original creator preferred to edit in photoshop, with crazy
psd files that contain layering, gamma, and other useful information. I
made further modifications to the resulting GIF file. My
(I apologize for the fact that this won't thread.
Future messages should thread properly.)
I agree that for the purposes of writing guidelines, it is
not necessary to be either precise or objective. Therefore my
comments below relate to the definition of 'preferred form ...'
in the context of
On Tue, Jun 17, 2003 at 02:47:58PM +0200, Thomas Hood wrote:
Objection #2: This definition would make it harder to produce free
software using non-free tools.
Answer: If your codebase is in a proprietary format and you work
on this using proprietary tools, and you release binary and source
Thomas Hood [EMAIL PROTECTED] writes:
License texts must be objective, fair and reasonably precisely
worded if they are to be enforceable.
This is mostly true, but it is very important to understand that we
are talking about *legally* objective, fair, precise. Such things as
the preferred
J.D. Hood [EMAIL PROTECTED] writes:
I suggest that the definition of 'preferred form for
making modifications' be information-theoretical.
No. It is *human*, and focused on actual, real, genuine, human
preferences. This is far better than trying to find a specific
technical definition of
Richard Braakman [EMAIL PROTECTED] writes:
On Sun, Jun 15, 2003 at 05:15:14PM -0400, Sam Hartman wrote:
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in a particular instance what
the preferred form for modifications was?
I
J.D. Hood said:
I suggest that the definition of 'preferred form for
making modifications' be information-theoretical.
When source code is compiled into binary code there is a
loss of information, as indicated by the fact that you
cannot get the original source back, given only the binary
On Mon, Jun 16, 2003 at 04:10:16PM +0200, Thomas Hood wrote:
Thomas Bushnell wrote:
No. It is *human*, and focused on actual, real, genuine, human
preferences. This is far better than trying to find a specific
technical definition of those preferences: much better instead to use
the
Thomas Hood said:
Joe Moore [EMAIL PROTECTED] wrote:
Finally, there is a very lossy conversion which must be Free,
and that is linguistic translation.
Nope. If you are distributing the binary with English UI then
I don't want the source with the UI translated lossily into
Romanian.
You've
Thomas Hood [EMAIL PROTECTED] writes:
The focus on human preferences tends to end up either in subjective
assessments or in speculation about what other people prefer.
Should these questions be settled by conducting surveys?
There's nothing wrong with subjectivity -- note that the
On Mon, Jun 16, 2003 at 04:10:16PM +0200, Thomas Hood wrote:
J.D. Hood wrote:
On the other hand, if there is a set of different forms
each of which is convertible into the others by means of
freely available tools then any member of the set is as
good as any other.
Andrew Suffield [EMAIL
Thomas Hood [EMAIL PROTECTED] writes:
Richard Braakman [EMAIL PROTECTED] writes:
I know of one thorny problem in this area: many graphics are distributed
as .png or .jpg files, even though their creator probably used a richer
format like .xcf.
Thomas Bushnell wrote:
Is it not obvious
On Monday, Jun 16, 2003, at 10:10 US/Eastern, Thomas Hood wrote:
In general if you possess both a non-indent(1)ed version of the
program you are distributing and version of the program that you
have run through indent(1), then I want the non-indent(1) ed version.
Generally, one doesn't run
On Sun, Jun 15, 2003 at 10:41:24PM -0700, Thomas Bushnell, BSG wrote:
Richard Braakman [EMAIL PROTECTED] writes:
On Sun, Jun 15, 2003 at 05:15:14PM -0400, Sam Hartman wrote:
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in
Richard Braakman [EMAIL PROTECTED] writes:
Is it not obvious that the preferred form is .xcf?
It is preferred, but does that make the other formats non-free?
I'm not sure. The talk about preferred form first comes up in the
requirement of the GPL to provide source. I don't know whether or
J == J D Hood [EMAIL PROTECTED] writes:
J I suggest that the definition of 'preferred form for making
J modifications' be information-theoretical.
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in a particular instance what
On Sun, Jun 15, 2003 at 05:15:14PM -0400, Sam Hartman wrote:
Why? What real-world problem does this solve? Have we actually run
into situations where it was not obvious in a particular instance what
the preferred form for modifications was?
I know of one thorny problem in this area: many
On Sun, Jun 15, 2003 at 07:47:32PM +0100, J.D. Hood wrote:
I suggest that the definition of 'preferred form for
making modifications' be information-theoretical.
When source code is compiled into binary code there is a
loss of information, as indicated by the fact that you
cannot get the
72 matches
Mail list logo