selling GPL sources
rick at linuxmafia.com
Tue Sep 20 23:32:04 UTC 2005
Quoting Ben Tilly (btilly at gmail.com):
> Question. Suppose that you are supposed to publically provide source
> and I wish to receive it. If I am unable to accept it in the form
> that you wish to distribute it, to what extent do you need to bend to
> accomodate my needs?
> For instance suppose that I live in Kenya, and do not have reliable
> phone service, let alone internet, and I wish to receive the source
> code to Linux. Do you have to be ready to mail a CD to me? I believe
> that the FSF's intent is that you do. I don't know what a judge would
<shrug> We may never know, since I doubt that even the Internet can
dredge up someone deranged enough to litigate the matter.
But, to reiterate, in the general case, "FSF's intent" would _not_ be
relevant to any litigation that hypothetically could arise: The judge
would look to indications of the _licensor's_ intent (and, if we're
talking contract law, the nature of terms agreed to between the
On the Internet, people can spin up hypothetical questions all day long,
and speculate wildly about how they'd be decided. Judges don't
particularly like to play that game: Thus requirements prerepquisites
like a cause of action, standing, etc.
But, speaking even as one of those Internet people, and as a non-lawyer,
I can point out some known tendencies in civil litigation: When there's
a formal written document such as a licence to indicate intent, judges
will tend to look to its plain wording first.
In this case (GPLv2), the licence says "offer [...] a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange". If I had to guess, I'd
speculate that a judge wouldn't really care that you live in Kenya or on
the far side of the moon. The question is: _What kind of offer_
substantively meets the redistributor's obligation under clause 3b?
I suspect the judge would ponder whether ftp/Web sites are "a medium
customarily used for software interchange", mutter to himself "Hell yes",
and throw plaintiff out on his ass.
Which is part of why _even_ the Internet probably won't produce someone
deranged enough to attempt that particular legal theory.
> For a more arbitrary instance, suppose that you make code publically
> available by putting them up on a proprietary network. Can you then
> claim, "Anyone can get access to the code, they just have to register
> for the proprietary network!" Is that allowed? I'd hope not, and
> would hope that a judge would agree.
I think the judge would throw you out on your ass hard enough that you
would bounce _twice_.
> But then what is the difference between said proprietary network and
> the Internet?
The difference is that judges are not Turing machines.
You do not tell them to behave like idiot savants (i.e., like
computers), and follow stupidly deterministic rules: They apply their
judgement according to their understanding of proper standards of
evidence, of due diligence and good faith dealings, of precedent, and of
the surrounding circumstances of particular cases.
You know, Ben: I get really tired of people thinking that judges cannot
arrive at a particular judgement because the commentator can't reduce it
to an algorithm. Judges arrive at decisions anyway. If you want to
understand how they work, tear yourself away from your terminal and
study law. (A prior professional existence required me to formally
study business law for several years, though I never became an attorney,
and I'm rusty on many particulars.)
> FWIW I suspect that that item was written in a time when access to the
> Internet in the USA was far from universal. So the situation in some
> ways resembled more closely the Kenyan example than the current USA.
That's nice, but I was speaking of the situation prevailing in this
decade, in our usual-suspect common-law countries; not in any previous
decade, and not in the Kingdom of Bhutan.
> No, but my understanding is that judges are likely to be swayed by
> their understanding of how people historically understood the license.
That would be considered relevant only to the extent of indicating how
the licensor is likely to have meant, and only if the plain language is
somehow really obscure. In this case, however, the plain language is
not in the least obscure.
> I've heard that this is part of why the FSF wrote their FAQ, to help
> ensure that the GPL would be interpreted as they want it interpreted.
Possibly. And I would hope that judges would not be very persuaded by
quite transparent advocacy.
> I'll bet that Richard *intended* the requirement to be physical media,
> so if you remind him, he'd be likely to make that fix.
Feel absolutely free-as-in-freedom to do it yourself. I can even supply
his USPS address, if you want to lavish postage on it.
Rick Moen Frater Magnus vos spectat.
rick at linuxmafia.com
More information about the License-discuss