what *is* the approval process? [was Re: License Committee Report for July 2007]

[Luis Villa-3]

On 7/30/07, Timothy McIntyre <tmcintyre@...> wrote:
>
>  The OSI, to its credit, requires compliance with the following procedures,
> in order to have a license approved:
>
> http://www.opensource.org/docs/certification_mark.html#approval

But Russ linked to:

http://opensource.org/approval

These two links are radically different documents (particularly the
second link's requirement of a 'legal analysis' of the OSD). Which of
them is actually the correct process?

Luis

[Russ Nelson]

Luis Villa writes:
 > On 7/30/07, Timothy McIntyre <tmcintyre@...> wrote:
 > >
 > >  The OSI, to its credit, requires compliance with the following procedures,
 > > in order to have a license approved:
 > >
 > > http://www.opensource.org/docs/certification_mark.html#approval
 >
 > But Russ linked to:
 >
 > http://opensource.org/approval
 >
 > These two links are radically different documents (particularly the
 > second link's requirement of a 'legal analysis' of the OSD). Which of
 > them is actually the correct process?

Hrm.  That's got to be my fault, since it was my action item to
"Preserve existing URLs".  The real "old website" approval process is
in certification_mark.php, which is no longer accessible and even if
it was, uses old PHP code.  I've looked, and it has the same content
as .../approval.  Thus I feel safe in saying that
http://opensource.org/approval has been the proper procedure to
follow, before and after the website upgrade.

My apologies for the confusion.  That page has been changed to point
to .../approval.

--
--my blog is at    http://blog.russnelson.com   | People have strong opinions
Crynwr sells support for free software  | PGPok | about economics even though
521 Pleasant Valley Rd. | [phone]       | they've never studied it.
Potsdam, NY 13676-3213  |     Sheepdog          | Curious how that is!

[Luis Villa-3]

On 7/31/07, Russ Nelson <nelson@...> wrote:

> Luis Villa writes:
>  > On 7/30/07, Timothy McIntyre <tmcintyre@...> wrote:
>  > >
>  > >  The OSI, to its credit, requires compliance with the following procedures,
>  > > in order to have a license approved:
>  > >
>  > > http://www.opensource.org/docs/certification_mark.html#approval
>  >
>  > But Russ linked to:
>  >
>  > http://opensource.org/approval
>  >
>  > These two links are radically different documents (particularly the
>  > second link's requirement of a 'legal analysis' of the OSD). Which of
>  > them is actually the correct process?
>
> Hrm.  That's got to be my fault, since it was my action item to
> "Preserve existing URLs".  The real "old website" approval process is
> in certification_mark.php, which is no longer accessible and even if
> it was, uses old PHP code.  I've looked, and it has the same content
> as .../approval.  Thus I feel safe in saying that
> http://opensource.org/approval has been the proper procedure to
> follow, before and after the website upgrade.
>
> My apologies for the confusion.  That page has been changed to point
> to .../approval.

My apologies to everyone else for resurrecting a much more onerous process.

So... what lawyer is going to volunteer to write the necessary
analysis for v3?

(crickets)

[Board: this could be your cue to step in, note that v3 has already
undergone the most extensive legal vetting of any license in history,
agree with Larry and Rick in the other thread, waive the analysis, and
let us all move on with life.]

Luis

[Chris DiBona]

Indeed. I hate to sound all political, but is there really a change
that OSI won't approve this license?

That said, I think it will be and so far has been on our hosting
system, quite popular. Wihtin a week of its being allowed, we had over
100 projects registered.

Chris

On 7/31/07, Luis Villa <luis@...> wrote:

> On 7/31/07, Russ Nelson <nelson@...> wrote:
> > Luis Villa writes:
> >  > On 7/30/07, Timothy McIntyre <tmcintyre@...> wrote:
> >  > >
> >  > >  The OSI, to its credit, requires compliance with the following procedures,
> >  > > in order to have a license approved:
> >  > >
> >  > > http://www.opensource.org/docs/certification_mark.html#approval
> >  >
> >  > But Russ linked to:
> >  >
> >  > http://opensource.org/approval
> >  >
> >  > These two links are radically different documents (particularly the
> >  > second link's requirement of a 'legal analysis' of the OSD). Which of
> >  > them is actually the correct process?
> >
> > Hrm.  That's got to be my fault, since it was my action item to
> > "Preserve existing URLs".  The real "old website" approval process is
> > in certification_mark.php, which is no longer accessible and even if
> > it was, uses old PHP code.  I've looked, and it has the same content
> > as .../approval.  Thus I feel safe in saying that
> > http://opensource.org/approval has been the proper procedure to
> > follow, before and after the website upgrade.
> >
> > My apologies for the confusion.  That page has been changed to point
> > to .../approval.
>
> My apologies to everyone else for resurrecting a much more onerous process.
>
> So... what lawyer is going to volunteer to write the necessary
> analysis for v3?
>
> (crickets)
>
> [Board: this could be your cue to step in, note that v3 has already
> undergone the most extensive legal vetting of any license in history,
> agree with Larry and Rick in the other thread, waive the analysis, and
> let us all move on with life.]
>
> Luis
>

--
Open Source Programs Manager, Google Inc.
Google's Open Source program can be found at http://code.google.com
Personal Weblog: http://dibona.com

[Luis Villa-3]

On 7/31/07, Chris DiBona <cdibona@...> wrote:
> Indeed. I hate to sound all political, but is there really a change
> that OSI won't approve this license?
>
> That said, I think it will be and so far has been on our hosting
> system, quite popular. Wihtin a week of its being allowed, we had over
> 100 projects registered.

Indeed. The FSF went to great lengths to ensure that v3 was credible,
and has apparently succeeded. If OSI doesn't approve it, people will
question OSI, not FSF/v3. (When I've mentioned today's discussion in
IRC, the universal reaction has been 'you mean OSI hasn't already
approved it?')

Luis

[Russ Nelson]

Luis Villa writes:
 > Indeed. The FSF went to great lengths to ensure that v3 was credible,
 > and has apparently succeeded. If OSI doesn't approve it, people will
 > question OSI, not FSF/v3. (When I've mentioned today's discussion in
 > IRC, the universal reaction has been 'you mean OSI hasn't already
 > approved it?')

We haven't approved it because it hasn't gone through license-discuss
yet.  Pace Andrew Oliver, we don't approve licenses that haven't gone
through the process.

--
--my blog is at    http://blog.russnelson.com   | People have strong opinions
Crynwr sells support for free software  | PGPok | about economics even though
521 Pleasant Valley Rd. | [phone]       | they've never studied it.
Potsdam, NY 13676-3213  |     Sheepdog          | Curious how that is!

[Dag-Erling Smørgrav-2]

Russ Nelson <nelson@...> writes:

> Hrm.  That's got to be my fault, since it was my action item to
> "Preserve existing URLs".  The real "old website" approval process is
> in certification_mark.php, which is no longer accessible and even if
> it was, uses old PHP code.  I've looked, and it has the same content
> as .../approval.  Thus I feel safe in saying that
> http://opensource.org/approval has been the proper procedure to
> follow, before and after the website upgrade.
>
> My apologies for the confusion.  That page has been changed to point
> to .../approval.

OK, Russ, now I'm getting angry.

Firstly, in <87r6ov93cp.fsf@...> on June 1st, I wrote the
following to you and the board:

> OK, first, allow me to point out that you have two different pages (the
> other is <URL:http://opensource.org/approval>) listing two slightly
> different procedures for license approval and two different email
> addresses.

You answered in <18016.24333.862925.706713@...> later the
same day:

> Hrm!  This is very not good.  I've been pointing people to
> certification_mark.html for years now, and yet it appears as if
> certification_mark.php is newer ... and that's what's on the new
> website.  But that's our proble, not yours.

So please don't pretend you weren't aware of this.

Secondly, I specifically asked you what was missing from my application
other than that I had sent it to the wrong address.  Your answer was:

> Yes, you should have sent it to license-discuss instead of us.  The
> procedure tells you to get an analysis from a licensed practitioner of
> the law, and yet .... we haven't held everybody to that standard.

I interpreted this, and the result of the (very brief) discussion on
license-discuss, to mean that a legal analysis was not required in this
case.

Thridly, I posted the Simplified BSD License to license-discuss on June
4th.  The board was already aware of it since I had mistakenly sent it
to them first.  *You* were already aware of it, having discussed the
application with me.  At no point after I posted it to license-discuss
did you or anyone else mention anything about it not being submitted
properly, until now - and you *still* won't tell me what's missing.

Fourthly, license-approval bounces, so there is nowhere else to submit
licenses to than license-discuss.

What am I to conclude?  That your little club is invitation-only, and I
wasn't invited?

DES
--
Dag-Erling Smørgrav
Senior Software Developer
Linpro AS - www.linpro.no

[Luis Villa-3]

On 8/1/07, Ernest Prabhakar <ernest.prabhakar@...> wrote:
> Hi Luis,
>
> On Jul 31, 2007, at 7:46 PM, Luis Villa wrote:
> > So... what lawyer is going to volunteer to write the necessary
> > analysis for v3?
> >
> > (crickets)

For what it is worth, I didn't meant this to imply that writing the
analysis was difficult or likely to fail, or that the lawyers on the
list are lazy, just that I'm guessing most of them have better ways of
spending their very limited time.

> That's probably not the gating factor here.   The real issue is
> ensuring the "many eyes" of this committee have evaluated GPLv3 vs.
> OSD. In practice, that usually means one person takes a stand and
> claims that it satisfies (or not) the various tenets of the OSD, and
> various people debate it back and forth.

That process was started a month ago (by Dibona and Flaschen), but
apparently that was insufficient to get it on the committee's radar. I
have assumed that the long-form analysis by a lawyer was the sticking
point, since every other meaningful/substantial part of the analysis
was already in motion a month ago. But if I'm wrong, and we're all OK
with ignoring that part of the rules, I've got no problem with that.
:)

> So, how about you, Luis? Are you willing to go on record saying that
> GPL3 is point-by-point compatible with the OSD?

(1) I'm not a lawyer, which the rules seem to require. (Just a law student.)
(2) At least one lawyer has already gone on the record doing so, so if
the requirement is 'be on the record', and not 'point by point
analysis', then we're already good to go.
(3) I've already got such an analysis written (informal, IANAL,
IANFSF, IANSFLC, yadda, yadda), but I decided not to post it in hopes
we could all agree with Larry and Rick's one-line assessments and move
on quickly with our lives.

Luis

[Rodent of Unusual Size-2]

Dag-Erling Smørgrav wrote:
>
> So please don't pretend you weren't aware of this.

He didn't say that he wasn't aware of it.  He admitted
that it was his action item to clear it up, and he dropped
the ball.  Having forgotten about something isn't the
same as pretending it doesn't exist.

> Thridly, I posted the Simplified BSD License to license-discuss on June
> 4th.  The board was already aware of it since I had mistakenly sent it
> to them first.  *You* were already aware of it, having discussed the
> application with me.

Keeping on top of things that go through the process correctly
is quite enough without having to try to track things that made
a false start.  So if someone makes such a false start and then
submits a licence correctly, it should treat it as though it
never did anything else -- because once the submitter is informed
of the false start, the whole thing drops off our radar except
for what happens subsequently on license-discuss.

> At no point after I posted it to license-discuss
> did you or anyone else mention anything about it not being submitted
> properly, until now - and you *still* won't tell me what's missing.

I have no comment on that, not having an opinion.  If someone
at OSI dropped the ball, I readily apologise on its behalf.
However, two months on license-discuss, particularly when there's
a lot of other traffic, doesn't seem an excessive interval
to me.

> Fourthly, license-approval bounces, so there is nowhere else to submit
> licenses to than license-discuss.

Another bug; thanks for mentioning it.  When did you last try sending
to that eddress?

> What am I to conclude?  That your little club is invitation-only, and I
> wasn't invited?

No, that we're human volunteers and fallible.  So sorry we're
not computerised automatons that never make mistakes.
--
#ken P-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist      http://Apache-Server.Com/

"Millennium hand and shrimp!"

[Luis Villa-3]

On 8/1/07, Luis Villa <luis@...> wrote:

> On 8/1/07, Ernest Prabhakar <ernest.prabhakar@...> wrote:
> > So, how about you, Luis? Are you willing to go on record saying that
> > GPL3 is point-by-point compatible with the OSD?
>
> (1) I'm not a lawyer, which the rules seem to require. (Just a law student.)
> (2) At least one lawyer has already gone on the record doing so, so if
> the requirement is 'be on the record', and not 'point by point
> analysis', then we're already good to go.
> (3) I've already got such an analysis written (informal, IANAL,
> IANFSF, IANSFLC, yadda, yadda), but I decided not to post it in hopes
> we could all agree with Larry and Rick's one-line assessments and move
> on quickly with our lives.

On re-reading, this was unnecessarily combative in tone; I apologize
for that. The offer stands to publish the analysis with appropriate
disclaimers, but I do think we'd all be better served if we just
skipped the discussion altogether.

Luis

[Rishab Aiyer Ghosh]

i think it is fair to implement the rules in context. the need for a legal analysis to be included with a submitted licence assumes that such analysis is new and not previously available to license discuss.

there are reams of legal analyses of gplv3 that are publicly available. while not necessarily point by point with reference to the OSD they tend to cover all points, imho. so perhaps a simpler way of dealing with this would be for anyone to state if they have concerns that the gplv3 does NOT fulfil some OSD criterion.

-rishab


On Wed, Aug 01, 2007 at 10:47:56AM -0400, Luis Villa wrote:

> On 8/1/07, Luis Villa <luis@...> wrote:
> > On 8/1/07, Ernest Prabhakar <ernest.prabhakar@...> wrote:
> > > So, how about you, Luis? Are you willing to go on record saying that
> > > GPL3 is point-by-point compatible with the OSD?
> >
> > (1) I'm not a lawyer, which the rules seem to require. (Just a law student.)
> > (2) At least one lawyer has already gone on the record doing so, so if
> > the requirement is 'be on the record', and not 'point by point
> > analysis', then we're already good to go.
> > (3) I've already got such an analysis written (informal, IANAL,
> > IANFSF, IANSFLC, yadda, yadda), but I decided not to post it in hopes
> > we could all agree with Larry and Rick's one-line assessments and move
> > on quickly with our lives.
>
> On re-reading, this was unnecessarily combative in tone; I apologize
> for that. The offer stands to publish the analysis with appropriate
> disclaimers, but I do think we'd all be better served if we just
> skipped the discussion altogether.
>
> Luis

[Luis Villa-3]

On 8/1/07, Ernest Prabhakar <ernest.prabhakar@...> wrote:

> Hi Luis,
>
> On Aug 1, 2007, at 7:47 AM, Luis Villa wrote:
> > On re-reading, this was unnecessarily combative in tone; I apologize
> > for that.
>
> No problem. :-)
>
> > The offer stands to publish the analysis with appropriate
> > disclaimers, but I do think we'd all be better served if we just
> > skipped the discussion altogether.
>
> Um, skip *which* discussion? The meta-discussion of whether the
> process is being followed,

That was my initial suggestion, but...

> did you want us to simply "stipulate"
> that the GPLv3 *is* OSD-compliant without further discussion on-list?

... that is what Larry and Rick suggested, and it seems reasonable to me.

Luis

[Dag-Erling Smørgrav-2]

I am still waiting for someone to tell me what the *actual*
application process is so I can resubmit the Simplified BSD License.

DES
--
Dag-Erling Smørgrav
Senior Software Developer
Linpro AS - www.linpro.no

[Chuck Swiger]

Hi, DES (and others)--

Dag-Erling Smørgrav wrote:
> I am still waiting for someone to tell me what the *actual*
> application process is so I can resubmit the Simplified BSD License.

In the past, one needed to create both a plain-text and an HTML version of the
license and submit them to both this mailing list and to the license-submit
address which you apparently found was not working.

For a submission which is an updated version of an already approved license,
such as the case for the Simplified BSDL and for the GPLv3/LGPLv3, the
requirement for a detailed legal analysis and comparison/rationale for the
license to be approved compared with other existing licenses has generally not
been needed.

To my mind, the Simplified BSDL is obviously OSD-compliant and should be
approved; the only issue I could see is whether it qualifies as being
sufficiently distinctive from the existing "new" or "modified" BSDL for
another, separate approval to be meaningful.

--
-Chuck

PS: Thanks for trying to focus attention on the purpose of this list, which is
reviewing licenses which have been submitted, and not ad-hoc legal debates on
whether the GPL is enforcable.  :-)

[Russ Nelson]

Dag-Erling Smørgrav writes:
 > You answered in <18016.24333.862925.706713@...> later the
 > same day:
 >
 > > Hrm!  This is very not good.  I've been pointing people to
 > > certification_mark.html for years now, and yet it appears as if
 > > certification_mark.php is newer ... and that's what's on the new
 > > website.  But that's our proble, not yours.
 >
 > So please don't pretend you weren't aware of this.

I confess to not being perfect, and making mistakes.  I also apologize
for them in general, and this one specifically.

 > Secondly, I specifically asked you what was missing from my application
 > other than that I had sent it to the wrong address.  Your answer was:

It's crucial to my reporting mechanism that you use the "For
Approval:" prefix on your email's subject.  There's far too much
chatter on license-discuss for me to be able to pick out discussion of
a license unless it occurs under a structured subject.

 > Fourthly, license-approval bounces, so there is nowhere else to submit
 > licenses to than license-discuss.

It bounces when you send it email which isn't an approval request.  I
installed that filter because of the incredibly HUGE amount of
unwanted email that an address published on the Internet for nearly
ten years receives.

--
--my blog is at    http://blog.russnelson.com   | People have strong opinions
Crynwr sells support for free software  | PGPok | about economics even though
521 Pleasant Valley Rd. | [phone]       | they've never studied it.
Potsdam, NY 13676-3213  |     Sheepdog          | Curious how that is!

[Russ Nelson]

Luis Villa writes:
 > That process was started a month ago (by Dibona and Flaschen), but
 > apparently that was insufficient to get it on the committee's radar.

No, the problem there is that we require everyone to follow the
process, even the best of friends.  If we didn't, we would (very
reasonably) be accused of cronyism.

--
--my blog is at    http://blog.russnelson.com   | People have strong opinions
Crynwr sells support for free software  | PGPok | about economics even though
521 Pleasant Valley Rd. | [phone]       | they've never studied it.
Potsdam, NY 13676-3213  |     Sheepdog          | Curious how that is!

[Luis Villa-3]

On 8/3/07, Russ Nelson <nelson@...> wrote:

>  > Secondly, I specifically asked you what was missing from my application
>  > other than that I had sent it to the wrong address.  Your answer was:
>
> It's crucial to my reporting mechanism that you use the "For
> Approval:" prefix on your email's subject.  There's far too much
> chatter on license-discuss for me to be able to pick out discussion of
> a license unless it occurs under a structured subject.

I assume this has been proposed and shot down before, but just in
case, I'd strongly suggest that this particular problem be solved by
forcing this list to stay on topic and moving the off-topic chatter
elsewhere, rather than by admitting failure on the signal-noise
problem and relying on specific subject headers.

Luis

[Russ Nelson]

Luis Villa writes:
 > I assume this has been proposed and shot down before, but just in
 > case, I'd strongly suggest that this particular problem be solved by
 > forcing this list to stay on topic and moving the off-topic chatter
 > elsewhere, rather than by admitting failure on the signal-noise
 > problem and relying on specific subject headers.

Been tried, didn't work; current system works.

--
--my blog is at    http://blog.russnelson.com   | People have strong opinions
Crynwr sells support for free software  | PGPok | about economics even though
521 Pleasant Valley Rd. | [phone]       | they've never studied it.
Potsdam, NY 13676-3213  |     Sheepdog          | Curious how that is!

[Dalibor Topic]

Russ Nelson <nelson <at> crynwr.com> writes:

>
> Luis Villa writes:
>  > I assume this has been proposed and shot down before, but just in
>  > case, I'd strongly suggest that this particular problem be solved by
>  > forcing this list to stay on topic and moving the off-topic chatter
>  > elsewhere, rather than by admitting failure on the signal-noise
>  > problem and relying on specific subject headers.
>
> Been tried, didn't work; current system works.
>

In light or Mr. Terekhov turning this list into gnu.misc.discuss [1] over the
past couple of days, and flooding it the way he did with debian-legal, when he
did his performance act there a little while ago, I'd be interested to know:

 what it would take to try one more time, with a vengeance?

cheers,
dalibor topic

[1] I believe his major achievement there is a 600+ messages thread on GPLv3,
the evilness of FSF and who knows what.

[Rick Moen]

Quoting Dalibor Topic (robilad@...):

> In light or Mr. Terekhov turning this list into gnu.misc.discuss [1] over the
> past couple of days, and flooding it the way he did with debian-legal, when he
> did his performance act there a little while ago, I'd be interested to know:
>
>  what it would take to try one more time, with a vengeance?

Just as a brief reminder:  Killfiles work, too.

--
Cheers,                 English is essentially an imprecise dialect of Java,
Rick Moen               without the object orientation.
rick@...     --Julian Morrison, http://ccil.org/~cowan/essential.html