<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
On 18/12/10 8:11 PM, Paolo Giarrusso wrote:
<blockquote
cite="mid:AANLkTikLNLLLtaJz3SocraCeocuXo2Do+_PDQomPBVgA@mail.gmail.com"
type="cite">
<pre wrap="">On Fri, Dec 17, 2010 at 16:06, <a class="moz-txt-link-rfc2396E" href="mailto:exarkun@twistedmatrix.com"><exarkun@twistedmatrix.com></a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">On 02:48 pm, <a class="moz-txt-link-abbreviated" href="mailto:tobami@googlemail.com">tobami@googlemail.com</a> wrote:
</pre>
<blockquote type="cite">
<pre wrap="">@Jean-Paul: I thought it would only be a problem if the license was
GPLv3?
</pre>
</blockquote>
<pre wrap="">Yes, that's right. But I got sad anyway.
</pre>
<blockquote type="cite">
<pre wrap="">But in any case, what is the conclusion, if any? Can the Mercurial API
be used from LGPL code or not?
</pre>
</blockquote>
<pre wrap="">This seems unanswerable. If you believe some people, yes. If you
believe other people, no. The only way to resolve the conflict is to do
it, and either get sued by the copyright holder or not get sued by the
copyright holder (you could argue that the outcome of the lawsuit must
also be considered, but by the time you've gotten sued you probably
don't care anymore).
</pre>
</blockquote>
<pre wrap="">Well, one point is that the matter is complex enough to make you
suable. Moreover, the point that you're technically not linking seems
extremely technical and specific to Python, and even to CPython - a
JIT compiler links together the code, even more than it happens for C
libraries. Therefore, explaining it to a judge would take an extremely
expensive lawyer.
</pre>
</blockquote>
<br>
The GPL doesn't mention linking at all. It applies to "<span
class="Apple-style-span" style="border-collapse: separate; color:
rgb(53, 56, 42); font-family: 'Times New Roman'; font-size: 16px;
font-style: normal; font-variant: normal; font-weight: normal;
letter-spacing: normal; line-height: normal; orphans: 2;
text-indent: 0px; text-transform: none; white-space: normal;
widows: 2; word-spacing: 0px;"><span class="Apple-style-span"
style="font-family: arial,helvetica,clean,sans-serif;
line-height: 20px; text-align: left;">the Program or any
derivative work under copyright law: that is to say, a work
containing the Program or a portion of it, either verbatim or
with modifications and/or translated into another language</span></span>".
Linking is only relevant because traditional static linking causes
parts of the library to be included in compiled binaries, which
automatically makes them derivative works.<br>
<br>
Downstream recipients of your code are also perfectly within their
rights to combine GPL code with code released under an incompatible
license; provided they never distribute such code, the GPL imposes
no restrictions. So the fact that the JIT compiler links code
together on the end user's system is not relevant. The GPL terms
only come into play if the CodeSpeed source can be considered a
"derivative work under copyright law" of Mercurial.<br>
<br>
If CodeSpeed is designed to be support several version control
systems via independently installable modules, then I don't see how
it could be considered a derivative work of Mercurial. The
integration module might be, but to accept that the existance of
such a module makes CodeSpeed a derivative work means that CodeSpeed
is also a derivative work of every version control system, or at
least will be as soon as a module is released for each system (which
could be done by other people). I find that rather ludicrous.<br>
<br>
That said, I'm not a lawyer, and I'm applying common sense reasoning
here, not knowledge of how the legal system actually treats things.<br>
<br>
-- Ben<br>
</body>
</html>