restarting License (anti-)Proliferation
Scott Shattuck
idearat at mindspring.com
Tue Aug 5 16:19:53 UTC 2008
Personally I think these ideas lead to a very slippery slope. It's one
thing to announce that upon further review a particular license
doesn't actually comply with the OSI definition. It's entirely another
to say one is preferred and another is deprecated. The former is a
clear responsibility of the OSI as the arbiter of what constitutes
compliance. The latter begins to step into a realm where the OSI can
be affecting business outcomes by passing quality judgements on
licenses in the absence of sufficient detail. Stating that a license
is "preferred" or "deprecated" is a fitness-of-purpose judgment in
some sense -- one which can't be made without taking into
consideration the goals of the licensor, licensee, and the various
specifics of the licensed software, the nature of the consuming
application and subsequent distribution requirements. In my mind
attempting to place a broadly sweeping value judgement on a license in
the absence of such information is a fools errand.
What I believe might be useful would be to publish "commentaries"
which describe what the community considers to the be pros and cons of
a particular license with respect to a set of common licensing
scenarios. This information would then allow both licensors and
licensees to get perhaps a clearer view of the tradeoffs each license
choice embodies. This is the kind of information the software
community needs -- not labels which can't really assist in making
informed licensing decisions.
ss
On Aug 5, 2008, at 9:06 AM, Smith, McCoy wrote:
> At the risk of proliferating categories, shouldn't there be a category
> (or perhaps a subcategory within "compliant") for "retired" or
> "deprecated" licenses? Those to me seem like they should be of an
> even
> lesser status than the "compliant" ones.
> McCoy
>
> -----Original Message-----
> From: Russ Nelson [mailto:nelson at crynwr.com]
> Sent: Tuesday, August 05, 2008 7:00 AM
> To: license-proliferation at opensource.org;
> license-proliferation-discuss at opensource.org;
> license-proliferation-2 at opensource.org
> Cc: license-discuss at opensource.org; osi at opensource.org
> Subject: restarting License (anti-)Proliferation
>
> [ Note the Reply-To. Please join the new list to do the work of the
> committee, or if you want to discuss the committee, do it on
> license-discuss. -russ ]
>
> [ I also want to call out explicitly Larry Rosen, Bruce Perens, Chris
> Dibona, and Van Lindberg as people who have expressed a strong opinion
> on the subject of license proliferation. Apologies to anyone whom
> I've left out. -russ ]
>
> I got a motion through the OSI board to restart the License
> (anti-)Proliferation committee. Here's the text of the motion:
>
> Mr. Nelson moves that we form a license proliferation committee to
> evaluate all existing licenses into two tiers - an upper tier and
> a lower tier of licenses (e.g. "recommended" and "compliant"). The
> role of this committee would be to establish criteria for
> assigning the tier for each license, use a new
> license-proliferation mailing list for discussion and come up with
> a final list of two tiers of licenses. Mr. Nelson will be chairing
> this new committee. The Board will select the two terms that are
> used. The deadline for presenting the draft recommendations from
> the committee back to the board will be October 2008. Ms. Cooper
> calls the vote, Mr. Tiemann seconds and the motion is passed
> unanimously.
>
> You may ask "aren't we doing a rewind?" No. Here's why:
>
> o We asked the previous committee to do the wrong thing, at which
> they proceeded to do a good job, but which was still wrong.
> Ask a wrong question and you get a wrong answer every time.
> o This committee is standing; the previous was ad-hoc.
> o This is going to be a public process, unlike the previous effort.
> o This committee will create a process to categorize the licenses;
> The previous committee categorized the licenses.
>
> Here's the problem statement:
>
> The problem of license proliferation has two countervailing
> aspects. Too many approved licenses increases the cost of using
> Open Source because of the quantity of licenses that must be
> understood. Each one fragments the community and reduces code
> sharing between projects. On the other hand, too few approved
> licenses means that others will claim "but our software's license
> complies with the OSD; read it for yourself" which weakens the
> brand name.
>
> The trouble is that we have only one flavor of cookie to hand out
> (a single "OSI Approved" trademark). With two flavours, we can
> give one to all licenses which comply with the OSD, and the other
> one to all licenses which we recommend to reduce licensing costs.
> But how to make this distinction? How do we do it without
> alienating somebody because their favorite license didn't make the
> list? How do we do it so that new licenses, which start off as
> merely Compliant and not Recommended, can get promoted? How do we
> de-Recommend some licenses, such as the Artistic 1.0 (currently on
> the losing end of a legal battle)?
>
> Answering these questions is the work of the committee.
>
> I suggest that this committee should come up with a published
> criteria which anyone can apply against the licenses to decide
> which ones we recommend. It should be a process for which anyone
> can understand the rationale. Yet, it will likely need tweaking,
> thus a standing committee. We have laws because human judgement
> isn't fair enough, but we have judges because laws are never fair
> enough.
>
> Once the committee is satisfied with its work, it will present its
> results to the Open Source Initiative for approval as policy. The
> board has requested that this be accomplished by the October board
> meeting (2nd Wednesday). I'm the chair of the committee.
> Membership of the committee is open to all, although disruptive
> members will be invited to comment on license-discuss instead.
>
> Join the committee by sending any piece of email to
> license-proliferation-2-subscribe at opensource.org. You will
> receive a subscription confirmation. Reply to it.
>
> Please start the discussion by reviewing the work of the ad-hoc
> committee: http://www.opensource.org/proliferation
>
> --
> --my blog is at http://blog.russnelson.com | Software that needs
> Crynwr sells support for free software | PGPok | documentation is
> software
> 521 Pleasant Valley Rd. | +1 315-323-1241 | that needs repair.
> Potsdam, NY 13676-3213 | Sheepdog |
More information about the License-discuss
mailing list