[beyond-licensing] quick review of scope, resources, and goals

Luis Villa luis at lu.is
Mon Apr 11 02:08:45 UTC 2016


I'm afraid I'm jumping in to the middle of this discussion, not having been
part of the initial discussion. I'm going based on what I've seen on the
wiki, and the initial discussions we had about this topic 1+ year ago at
the board level.

(Overall, to be clear, I strongly support the "beyond licensing" exercise -
I think it is critical to the future of OSI and FOSS more broadly. And I
support software freedom; I think it is ethically important and am a
dues-paying member of FSF, though I would have opposed the addition of
"software freedom" to the OSI mission statement had I still been on the
board, mostly for the same reasons I'm going to give below.)

(tl;dr: open source's developer-focused pragmatism is related to, but quite
different than, any existing definition of software freedom; blurring those
distinctions risks losing lots of important context and learnings.)

On Sun, Apr 10, 2016 at 8:46 AM Allison Randal <allison at opensource.org>
wrote:

> In fact, software freedom is the direct cause of the practical
> benefits of open source. And at some point, those abstract benefits of
> software freedom are important in telling the difference between open
> source and a license that only shares the source code but doesn't grant
> the right to use, modify, and redistribute it. At first glance it might
> seem "like open source", but the fauxpen source license only gives you a
> fraction of the practical benefits of open source.
>
> Open source and software freedom aren't two different things, they're
> two ways of talking about the same thing.
>

I don't think that's correct; perhaps more importantly for this group, it
isn't consistent with a "beyond licensing" approach to open source.

Definitionally, the reason why software freedom isn't the same as open
source "beyond licensing" should be obvious: no one has articulated a
coherent, useful definition of software freedom that goes beyond licensing,
so you can't go "beyond licensing" without also going beyond software
freedom.[1] More arguably, FSF's freedoms are (correctly!) focused on the
freedoms of the *user*,[2] open source has focused on *developers.*[3]

Practically, I think there are many activities/behaviors that would widely
be labeled as "open source" but:

   - have nothing to do with licensing/source availability
   - are not part of "software freedom" as anyone has ever defined it that
   I'm aware of (i.e., if you failed to do them, FSF would still sign off, but
   many open source developers would be grumpy)
   - are being rapidly embraced/extended by proprietary software (so are
   very relevant to the future of both free and open software, and certainly
   to the future of OSI)

Some examples of these activities, just offhand (don't have time to
elaborate a ton tonight):

   - *revision control:* historically something done poorly by proprietary
   software development; brought to the forefront as a best practice, and best
   tools developed aggressively, by FOSS communities; now assumed by all
   software (free, open, or otherwise) as a minimum necessary for beginning
   any development. Nothing to do with freedom per se; widely associated with
   open source development.
   - *publicly-readable bug tracking:* key part of non-licensing open
   source community-focused development; now adopted widely even by Apple
   <https://blackpixel.com/writing/2012/02/radar-or-gtfo.html>. Nothing to
   do with freedom or licensing.
   - *collaborative decision-making:  *So little to do with "freedom" that
   RMS is used as the posterchild for how to do this wrong. Considered a key
   part of any healthy open source community.
   - *communication and distributed development:* IRC and mailing lists to
   tie together distributed developers were rare in proprietary software
   pre-OSS; they're now de rigeur to the point of supporting multiple
   "unicorns" (Slack and Github).
   - *distributed governance: *free software licenses are one mechanism for
   ensuring distributed governance, but they aren't the only one; this is why
   you see standards bodies and industry consortia adopting many of our
   governance patterns, not just (and often not ever) our licensing.

The common theme among all of these, of course, is how developers work
together. Doing that right gives you most of the benefits of open source,
regardless of licensing. That's why proprietary software is embracing these
things pretty wholesale. It's telling, I think, that no one calls these
"free" development practices; they're *open* development practices (recent
example <https://github.com/WhiteHouse/source-code-policy/issues/129>).

So, yeah, I think OSI does itself, open development practices, and the
broader movement a disfavor by tying itself too deeply to software freedom.
When those are (incorrectly) touted as synonyms, important nuance is lost.

Sorry for the brevity - this is an important topic that deserves less force
and more nuance, but I'd rather put the issues on the table badly than not
at all.

Luis
[1] As some of you even saw in person, I've started talking about a
non-licensing theory of software freedom myself, but it's very definitely
in an overall vacuum: http://lu.is/?p=2917
[2] http://www.gnu.org/philosophy/free-sw.en.html uses "user" 16 times,
developer 6x, "community" 2x
[3] ESR engaged in a lot of revisionist history in old versions of
http://www.opensource.org/history, but one claim that I think was accurate
(and left in when I rewrote it) was the statement that OSI grew out of a
determination to "advocate for the superiority of an open *development*
process". (emphasis added by me in this email)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/beyond-licensing_lists.opensource.org/attachments/20160411/a829f475/attachment.html>


More information about the beyond-licensing mailing list