[License-discuss] comprehensiveness (or not) of the OSI-approved list
rick at linuxmafia.com
Thu May 23 02:15:42 UTC 2019
Quoting Henrik Ingo (henrik.ingo at avoinelama.fi):
> And me, and others... This was the quickly growing consensus.
Including me, emphatically. For whatever that's worth.
I note with appreciation Richard Fontana reminding us of the then-recent
context of the MXM/MPEG-LA matter. That certainly makes his comments at
the time very clear (again). At this remove, I can't recall clearly, but
doubt that I engaged with l-r concerning MXM for reasons lost to time's
passage. But, irrespective of that, Richard's point that there's
probably been inconsistency in roughly similar cases based on bias for
or against the proposer is doubtless well taken. I can only say that
CC0 seemed an open-and-shut case, so I said so. If I'd been advised
that very similar no-patent-grant wording in MXM had drawn hostility
(did it?), then I'd have said 'Well, that's unfortunate, but I'm looking
at only CC0 today, and the decision looks clear to me, so I'm not
bothered by some _other_ commenters having previously disliked a
different licence now claimed to have been similar.'
The point is well taken, IMO, that the notion of l-r consensus being
anything like dispositive should be approached with skepticism.
On a separate matter, someone elsewhere in this thread had a useful
taxonomy of positions in the OSI community on patents, which I don't
have in front of me. FWIW, my own view is:
An open source licence is a necessary but not sufficient condition
of covered software being open source. The licence conveys copyright
and sometimes adjacent bundles of legal rights, but inherently can never
solely suffice to determine the status of covered code.
A limiting example can illustrate why: Today, I declare that a codebase
is 2-clause BSD licensed. I post tarballs with compiled binaries.
Somehow I just don't get around to posting source code, and it appears
that, as far as observers can determine, I might be the only person to
possess appropriate source trees, or perhaps in the worst case I might
have been careless with a recursive rm operation and accidentally
deleted all usable source trees.
Is that software covered by an open source licence? Absolutely. Is the
software open source? Not really until someone finds a source tree.
Why? Because access to source code is yet another necessary but not
sufficient condition for a codebase to be open source, in addition to
licensing rights to copyright title and sometimes other things.
Hypothetical #2: Now that someone has found the missing source code, it
transpires that its algorithms are encumbered by strongly asserted
patent rights (let's say, 'RAND' terms) in the Kingdom of Ruritania, and
nowhere else. The newly discovered Ruthenian patent will expire
December 31, 2019.
With this revelation, is my open source code _still_ open source? I
would say, yes, in all but one of the world's countries, as far as we
know (absent further problems such as submarine patents in the Kingdom
of Latveria). No in one country through the end of the year. Yes in
that country starting New Year's Day.
Implicit in my position is that whether a codebase is open source
depends on whether OSD rights are prevented from being lawfully
available on grounds of either impossibility (absence of source code) or
any legal impediment that functions to nullify the licence's substantive
 I'm curious about regulars' assessment of this personal view, and of
alternatives (including as to blind spots and errors). Thanks. (On the
other hand, if I'm only belabouring the obvious, apologies for that.)
Cheers, "I am a member of a civilization (IAAMOAC). Step back
Rick Moen from anger. Study how awful our ancestors had it, yet
rick at linuxmafia.com they struggled to get you here. Repay them by appreciating
McQ! (4x80) the civilization you inherited." -- David Brin
More information about the License-discuss