[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Fsfe-uk] Report from BCS OSSG meeting on a new FS license

From: Alex Hudson
Subject: [Fsfe-uk] Report from BCS OSSG meeting on a new FS license
Date: Wed, 25 Jul 2007 09:25:15 +0100

Hey everyone. (lengthy; sorry).

So, not too many people turned out yesterday - I'm guessing that for
many, travel is still an issue in these fair Isles - but there weren't
too many people who were in favour of the new license at this meeting.

Well, in fact, no-one really spoke in favour of it: even amongst the
committee there who supposedly were interested in it, it was difficult
to tell who the protagonists were. The person who proposed it (I think;
at the time he was owning up to it he wasn't making it very clear)
didn't really say much, he just took quasi-minutes of the meeting.
Another person took us through an introduction to the idea, but he also
didn't profess to be in favour of it, and basically just read out the
text on the slides, in a kind of "it is alleged.. " fashion. Really
strange: it's very difficult to argue against something when no-one will
adequately make a case _for_ it :)

Some notes on the meeting.

I believe those quasi-minutes are going to be published on the internet,
so I'm not going to try and give a summary of the overall meeting, just
some thoughts I had afterwards. I believe there will be time for
feedback on the ideas presented there for some period of time, I'll try
to notify this list if I see it before anyone else.

The general push from the meeting was one of constructive criticism I
think. While it was difficult to pin anyone down on exactly why this
license was needed, the "worry" was that the GPL concerns businesses who
want to get around copyleft and are unsure if they can (this came across
as the driving reason to me, anyway). I tried to get some positive
statements about what this new license would do; all I received were
more proposed failures in the GPL  drafting. It was very difficult to
get any kind of elevator pitch about what a new license could do, only a
"GPL makes my clients uncertain" type of response.

The only other reasons put forward were a kind of "The EU have their own
license, so we should have our (local license for local people) also,
and that would help public-sector take-up" (?!). I entirely missed an
opportunity to propose my Yorkshire Public License, but it did seem to
be grudgingly conceded that such parochialism was somewhat out of step
with the modern, copyright-harmonised, world.

Therefore, attendees asked the committee to a. look at publishing some
advice under the BCS examining some issues which potentially unsettle
businesses, encouraging them to take up free software, and b. do some
research on those companies who are being discouraged by the licenses,
before moving any further forward with a new license.

I have a feeling at least some on the committee are not terribly
familiar with licenses at all. Hearing that the GPL can be terminated,
he said, "So, it's unsafe then?". That theme seemed to stick with him,
until Mark Taylor (OSC) made a pretty impassioned point about how the
"Is it safe?" question is very 1990s and that free software had moved
somewhat ahead in that time.

There didn't seem to be much awareness of where free software was being
used in the public sector, how much traction it had in the private
sector, or the extent to which small businesses were using it. In many
ways, it did feel like going back ten years, worrying about how people
might make money from it, is it safe, is it sustainable, etc. etc. -
arguments I thought I'd heard the last of many years ago. Why the
licensing was being seen as a limiting factor on take-up wasn't entirely
clear to me, except that perhaps it was a larger factor in the
experiences of those there.

Some notes on the GPL.

To this attendee, at least, virtually the entire event was spent talking
about the failures of the GPLv2. No other license was even raised,
except to say "BSD, MIT, etc.," (they didn't mention the Apache license,
bizarrely) "are also somewhat popular". 

Interestingly, the lawyer present (who also professed to not being
persuaded a new license was needed, and said it would be very difficult
to write, but who seemed to be seen as a protagonist by the other
committee members...) seemed to intimate strongly that a lot of advice
he gives is basically about copyleft dodging, and believes that the
GPLv2 copyleft is severely limited, particularly with respect of use of
GPLv2 libraries, at least under E&W law. And has presumably been
advising clients of that for some time, who are now almost certainly
violating the spirit of the GPLv2, if at least not the letter.

I didn't get an exact description of this fault in the GPL from him, but
by piecing bits together, it seems to stem from a reading of the second
sentence in paragraph 0: that the words after the colon are not
tautologous to those before it. He believes the words before it take
precedence (that "work based on the program" means "a derived work as
defined in copyright law"), and that many uses of GPL'd software fall
not under derived works (I don't think E&W law even uses that term tbh)
but are collective works.

I _suspect_ that basically he is advocating an NVidia approach to
getting around the copyleft, in that so long as you don't distribute
things together, you can license the stuff you link to the GPL any way
you want. I don't really see how that gets around contributory
infringement, nor do I see why being a collective work would make a
difference (perhaps that's relevant to the contributory factor?), so
it's likely I don't have all the pieces in that puzzle.

It also seems to me that the GPLv3 has mostly addressed the concerns
raised; the definitions are a lot more broad, at some other points
(e.g., what happens to licenses when copyright ownership is transferred,
how you can retrieve the situation if you're in non-compliance, etc.)
have been pretty thoroughly addressed. I didn't get the impression that
any kind of analysis of the GPLv3 had been done, so most of the points
raised may or may not apply anyway.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]