OT: Two ways Microsoft sabotages Linux desktop adoption
mike.mccarty at sbcglobal.net
Thu Feb 16 19:24:05 UTC 2006
Joel Rees wrote:
> 2006-02-15 (水) の 23:36 -0600 に Mike McCarty さんは書きました:
>>It states that
>>section 6 states the terms for "distribution of such executables". So
>>one might argue that if one links such an executable on his own
>>by linking it at load time, but does not distribute it, then 6 has no
>>force, since it is not distributed.
> That's the point.
Well, I just went back and checked, and this point is
actually addressed by LGPL, and it specifically states
that this is not the proper interpretation. The work
is still considered as being a derivative work, regardless
of how the link takes place.
Here is the rlevant section...
6. As an exception to the Sections above, you may also combine or link a
"work that uses the Library" with the Library to produce a work
containing portions of the Library, and distribute that work under terms
of your choice, provided that the terms permit modification of the work
for the customer's own use and reverse engineering for debugging such
b) Use a suitable shared library mechanism for linking with the Library.
A suitable mechanism is one that (1) uses at run time a copy of the
library already present on the user's computer system, rather than
copying library functions into the executable, and (2) will operate
properly with a modified version of the library, if the user installs
one, as long as the modified version is interface-compatible with the
version that the work was made with.
[END OF QUOTATION]
> Publication is the primary thing which is regulated in both GPL and
> LGPL, not execution. That is one of the major differences between the
> attitude of traditional EULAs and the approach taken by the GPL. (And it
> is one of the big reasons you do a lot of compiling FreeBSD and its
> packages from source, although I think they ride the line a little thin
> in a number of cases.)
> I might assume that your lawyer friends did not take the time to
I'm sure you may presume anything you like.
> understand the license well enough to recognize such fine distinctions
> as the above. I would bet they assumed restrictions on running the
> programs because those have been the whole point of about half the terms
> of EULAs until the GPL, and also because you described their responses
> in such terms. (If it was free advice, one could understand why they
> would not take the time, but I don't recall how much you said you paid
> them for the analysis.)
No, the issue was, and always remained, protection of trade
secrets. Anything which gets linked, in whatever manner,
with LGPL, must provide to the customer a copy of the
entire work in a manner suitable for him to modify for his
own use. And that gives away the trade secret, which is
then lost in perpetuity.
> I don't begrudge them an error for two reasons, one being that time is
This is not an error. I got onto that for a moment.
> the most valuable thing to a professional, the other that this is an
We argued for *days*.
> easy mistake to make in an economic climate where licenses are usually
> designed to control what people do with a thing as much as whether they
> re-distribute the thing. But if you're going to come here and take the
The issue is disclosing trade secret.
> time to disparage the license which made a lot of programmers feel free
> to help build Linux, you should first take the time to understand what
> it is you disparage.
Where have I disparaged ANYTHING AT ALL?
Man, this last bit is annoying in several ways. I spent
*days* discussing this with lawyers. I haven't disparaged the
license. Every decision has consequences, both good and bad.
The use of GPL and LGPL means that Linux will never have
proprietary device drivers, which is an *intended* consequence
of both of those licenses. I don't see how pointing that
out is disparaging.
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
More information about the users