You are viewing a plain text version of this content. The canonical link for it is here.
Posted to derby-dev@db.apache.org by "Jean T. Anderson" <jt...@bristowhill.com> on 2006/06/23 02:17:37 UTC

catch-22: Derby, Mustang, and JCP issue

David posted a good summary of the legal catch-22 at [1]. But the
shortest story is:

 + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
 + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.

Let's keep this thread confined to the JCP issue Andrew raised that to
roll a release candidate qualifies as "creation".[2] And those release
candidates will be generally available.

How do we handle this cleanly?

-jean


[1]
http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
[2] http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html

Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Jean T. Anderson wrote:
> David posted a good summary of the legal catch-22 at [1]. But the
> shortest story is:
> 
>  + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>  + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
> 
> Let's keep this thread confined to the JCP issue Andrew raised that to
> roll a release candidate qualifies as "creation".[2] And those release
> candidates will be generally available.
> 
> How do we handle this cleanly?
> 
> -jean
> 
> 
> [1]
> http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
> [2] http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html

I goofed up these URLs and both point to the start of the thread. I
meant to point to these:

[1]
http://www.nabble.com/Re%3A-Proposal-for-10.2-release-schedule-p5003242.html
[2]
http://www.nabble.com/Re%3A-Proposal-for-10.2-release-schedule-p5003602.html


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Daniel John Debrunner wrote:
> Geir Magnusson Jr wrote:
>> Daniel John Debrunner wrote:
>>
>>> Jean T. Anderson wrote:
>>>
>>>
>>>> David posted a good summary of the legal catch-22 at [1]. But the
>>>> shortest story is:
>>>>
>>>> + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>>>> + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
>>>>
>>>> Let's keep this thread confined to the JCP issue Andrew raised that to
>>>> roll a release candidate qualifies as "creation".[2] And those release
>>>> candidates will be generally available.
>>> I don't think the JCP rules apply since we are not creating an
>>> implementation of JSR221.
>>
>> How could that be?  Where did the information for the APIs come from?
> 
> Isn't an implementation of JSR221 writing (clean room) classes in the
> java.sql and javax.sql name spaces. (e.g. java.sql.Driver &
> javax.sql.DataSource).
> 
> Derby is not doing that, Derby is providing an implementation of a JDBC
> driver, not an implementation of JSR221 itself. Implementing JSR221 is
> something that Harmony would (might) do.
> 
> My point is that the JCP rules for implementation of the spec itself do
> not apply here.

It's true that you aren't doing what Harmony will do - you aren't
providing a jar with "java.sql.Connection" (for example), but you are
writing software that _uses_ the spec and, formally speaking, that
requires the spec to do.

Writing software to use the spec is allowed by JSR final spec licenses
(thank goodness), but not in the draft spec license for 221.

> 
>>> I'm sure other rules apply that say one cannot ship a JDBC 4.0 driver
>>> until Mustang goes GA, but that's not the JCP rules that Andrew provided
>>> references to. It would be good to get Lance to expand on what he meant
>>> when he said:
>>>
>>> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes final."
>>>
>>> Where does this restriction come from?
>>
>> I'm not sure.  The license on the public review draft gives no rights to
>> distribute an implementation under *any* label (more recent public
>> review draft licenses  do now after the whole EJB3 brouhaha)
> 
> My point is "implementation" of what exactly. I think Lance (JSR221 spec
> lead) has already said that the TCK for JSR221 just checks the the
> classes (java.sql etc.) exist and have the correct methods & signatures,
> that implies to me that implementing JSR221 means providing those classes.

Indeed.  You won't need a TCK for 221 for Derby (I clearly misspoke in
other messages on this thread) as you aren't implementing 221, just
software that uses it.

geir

Re: Proposed Solutoin ( Was Re: catch-22: Derby, Mustang, and JCP issue)

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>
> 1) Have Sun change the draft spec license for 221 from the current to
> the new one that allows distribution with appropriate warning markings.
>  I'm going to start working this line w/ the PMO and the JCP.
>
> 2) Reject Mark Reinhold's curious claim that a relational database is
> actually an implementation of JDBC.  Derby is a relational database that
> among a whole array of features and functionality, one just happens to
> be DBC4.
>
> 3)  Do your 10.2 GA on your own schedule. Serve your user community.
> Make people happy.  Bathe in the accolades, get good press coverage,
> treat yourself to a nice bottle of wine.  (And put a note in the release
> notes that state the the JDBC4 functionality is "pre-spec" or whatever
> the new draft spec license requires.)
>
> 4) Release 10.2.1 when Mustang goes GA and remove that sentence in the
> release notes.
>
> I think in this way, everyone is satisfied.  The ASF is following the
> letter of the law wrt JCP specs (as we always have done), a GA release
> can be made with the functionality needed by Mustang, Sun doesn't have
> to fork Derby, users have no upgrade issues with their "production
> version", etc.
>
> Comments?

+1

Added bonus that Sun can put a version of JavaDB/Derby into the JDK
that is midstream between the two releases, and which is unmodified
from the Derby source in Subversion.

andrew

Re: Proposed Solutoin ( Was Re: catch-22: Derby, Mustang, and JCP issue)

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Geir Magnusson Jr wrote:
> Update - a slight modification to #1 would be for the upcoming proposed
> final draft to be under the new license.  (Source of this suggestion
> shall remain nameless)  That way no unnatural acts have to be done to an
> already-released draft.

Whatever works for both Apache and Sun is terrific.

 -jean

> geir
> 
> 
> Jean T. Anderson wrote:
> 
>>Geir Magnusson Jr wrote:
>>
>><snipped interesting stuff to show just the end solution>
>>
>>>While I'd just chuck JDBC4 myself, that doesn't work for Sun.  So I did
>>>come up with one solution :
>>>
>>>1) Have Sun change the draft spec license for 221 from the current to
>>>the new one that allows distribution with appropriate warning markings.
>>> I'm going to start working this line w/ the PMO and the JCP.
>>>
>>>2) Reject Mark Reinhold's curious claim that a relational database is
>>>actually an implementation of JDBC.  Derby is a relational database that
>>>among a whole array of features and functionality, one just happens to
>>>be DBC4.
>>>
>>>3)  Do your 10.2 GA on your own schedule. Serve your user community.
>>>Make people happy.  Bathe in the accolades, get good press coverage,
>>>treat yourself to a nice bottle of wine.  (And put a note in the release
>>>notes that state the the JDBC4 functionality is "pre-spec" or whatever
>>>the new draft spec license requires.)
>>>
>>>4) Release 10.2.1 when Mustang goes GA and remove that sentence in the
>>>release notes.
>>>
>>>I think in this way, everyone is satisfied.  The ASF is following the
>>>letter of the law wrt JCP specs (as we always have done), a GA release
>>>can be made with the functionality needed by Mustang, Sun doesn't have
>>>to fork Derby, users have no upgrade issues with their "production
>>>version", etc.
>>>
>>>Comments?
>>
>>+1
>>
>> -jean
>>
>>


Re: Proposed Solutoin ( Was Re: catch-22: Derby, Mustang, and JCP issue)

Posted by Geir Magnusson Jr <ge...@pobox.com>.
Update - a slight modification to #1 would be for the upcoming proposed
final draft to be under the new license.  (Source of this suggestion
shall remain nameless)  That way no unnatural acts have to be done to an
already-released draft.

geir


Jean T. Anderson wrote:
> Geir Magnusson Jr wrote:
> 
> <snipped interesting stuff to show just the end solution>
> 
>> While I'd just chuck JDBC4 myself, that doesn't work for Sun.  So I did
>> come up with one solution :
>>
>> 1) Have Sun change the draft spec license for 221 from the current to
>> the new one that allows distribution with appropriate warning markings.
>>  I'm going to start working this line w/ the PMO and the JCP.
>>
>> 2) Reject Mark Reinhold's curious claim that a relational database is
>> actually an implementation of JDBC.  Derby is a relational database that
>> among a whole array of features and functionality, one just happens to
>> be DBC4.
>>
>> 3)  Do your 10.2 GA on your own schedule. Serve your user community.
>> Make people happy.  Bathe in the accolades, get good press coverage,
>> treat yourself to a nice bottle of wine.  (And put a note in the release
>> notes that state the the JDBC4 functionality is "pre-spec" or whatever
>> the new draft spec license requires.)
>>
>> 4) Release 10.2.1 when Mustang goes GA and remove that sentence in the
>> release notes.
>>
>> I think in this way, everyone is satisfied.  The ASF is following the
>> letter of the law wrt JCP specs (as we always have done), a GA release
>> can be made with the functionality needed by Mustang, Sun doesn't have
>> to fork Derby, users have no upgrade issues with their "production
>> version", etc.
>>
>> Comments?
> 
> +1
> 
>  -jean
> 
> 

Re: Proposed Solutoin ( Was Re: catch-22: Derby, Mustang, and JCP issue)

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Geir Magnusson Jr wrote:

<snipped interesting stuff to show just the end solution>

> While I'd just chuck JDBC4 myself, that doesn't work for Sun.  So I did
> come up with one solution :
> 
> 1) Have Sun change the draft spec license for 221 from the current to
> the new one that allows distribution with appropriate warning markings.
>  I'm going to start working this line w/ the PMO and the JCP.
> 
> 2) Reject Mark Reinhold's curious claim that a relational database is
> actually an implementation of JDBC.  Derby is a relational database that
> among a whole array of features and functionality, one just happens to
> be DBC4.
> 
> 3)  Do your 10.2 GA on your own schedule. Serve your user community.
> Make people happy.  Bathe in the accolades, get good press coverage,
> treat yourself to a nice bottle of wine.  (And put a note in the release
> notes that state the the JDBC4 functionality is "pre-spec" or whatever
> the new draft spec license requires.)
> 
> 4) Release 10.2.1 when Mustang goes GA and remove that sentence in the
> release notes.
> 
> I think in this way, everyone is satisfied.  The ASF is following the
> letter of the law wrt JCP specs (as we always have done), a GA release
> can be made with the functionality needed by Mustang, Sun doesn't have
> to fork Derby, users have no upgrade issues with their "production
> version", etc.
> 
> Comments?

+1

 -jean

Proposed Solutoin ( Was Re: catch-22: Derby, Mustang, and JCP issue)

Posted by Geir Magnusson Jr <ge...@pobox.com>.
Andrew McIntyre wrote:
> On 6/23/06, David Van Couvering <Da...@sun.com> wrote:
>>
>> What happened to the proposal to vote and approve a GA-enabled release,
>> but not make it actually available until Java SE 6 goes GA?  Did we
>> decide this was not feasible?  I may have missed it, but I don't think I
>> saw that discussion anywhere.
> 
> This point may have been lost in the discussion, but its still
> important to address this, since this in fact was the original
> proposal and I don't believe it's been directly addressed. From the
> discussion, though, what I'm beginning to ascertain is that we
> shouldn't do anything until the spec is final. But I'm curious what
> Geir thinks about this particular option (vote + hold the release till
> the spec is final).

My personal opinion is that you shouldn't let the Mustang tail wag the
Derby dog.

This was one of the fears I had with Sun bundling Derby into the JDK,
and the irony is that it took a whole 3 days for my fears to be
realized.  I was imagining a problem down the road in some patchlevel
update of the JDK.  Never did it occur to me that Sun couldn't even get
out of the starting gate with this.

  - o -

However, putting that aside, it's really up to the community and I want
 to find a solution that satisfied everyone, including Sun.

I also don't think that you want Mustang to "steal your thunder".  Derby
 should get the attention and kudos it deserves for 10.2, with the
Apache Derby project being the "primary distributor" rather than
Mustang, with Derby doing a "me too" release.

While I'd just chuck JDBC4 myself, that doesn't work for Sun.  So I did
come up with one solution :

1) Have Sun change the draft spec license for 221 from the current to
the new one that allows distribution with appropriate warning markings.
 I'm going to start working this line w/ the PMO and the JCP.

2) Reject Mark Reinhold's curious claim that a relational database is
actually an implementation of JDBC.  Derby is a relational database that
among a whole array of features and functionality, one just happens to
be DBC4.

3)  Do your 10.2 GA on your own schedule. Serve your user community.
Make people happy.  Bathe in the accolades, get good press coverage,
treat yourself to a nice bottle of wine.  (And put a note in the release
notes that state the the JDBC4 functionality is "pre-spec" or whatever
the new draft spec license requires.)

4) Release 10.2.1 when Mustang goes GA and remove that sentence in the
release notes.

I think in this way, everyone is satisfied.  The ASF is following the
letter of the law wrt JCP specs (as we always have done), a GA release
can be made with the functionality needed by Mustang, Sun doesn't have
to fork Derby, users have no upgrade issues with their "production
version", etc.

Comments?

geir

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, David Van Couvering <Da...@sun.com> wrote:
>
> What happened to the proposal to vote and approve a GA-enabled release,
> but not make it actually available until Java SE 6 goes GA?  Did we
> decide this was not feasible?  I may have missed it, but I don't think I
> saw that discussion anywhere.

This point may have been lost in the discussion, but its still
important to address this, since this in fact was the original
proposal and I don't believe it's been directly addressed. From the
discussion, though, what I'm beginning to ascertain is that we
shouldn't do anything until the spec is final. But I'm curious what
Geir thinks about this particular option (vote + hold the release till
the spec is final).

The obvious problem here is if the spec changes between the date the
voting starts on a  release candidate and the date the spec is
finalized, you'll need to start over with a new RC. Not a terrible
option, but plenty of work for the RM, who will need to track changes
to the spec down to the last minute.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by David Van Couvering <Da...@Sun.COM>.
Hi, Kathey.  I understand your concerns, and I think they're important, 
but I do want to make sure we distinguish two separate issues.

The first issue is around the JCP.  This has nothing to do with a 
company's release schedule.  It has to do with the fact that we as a 
community wanted to ship a product supporting JDBC4.  This holds us to 
certain legal constraints.  Yes, we could ship a feature release "on 
time" without JDBC4 support, but it would be another six months before 
we could do a second feature release that supports JDBC4.  I think it's 
a tough call to decide to hold off on the release a bit or drop an 
interim release.  I thought we as a community decided to hold off on the 
10.2 release, and I thought it was a good compromise to get 10.1.3 out 
in the meantime.

The second issue is around Derby (Java DB) into Sun's JDK.  I think this 
is where there is a bit of a struggle.  I think we all see the value of 
having a redistribution of Derby in the JDK, especially one that allows 
users to upgrade from a database created in this JDK to Derby or to 
other distributions based on Derby.  But it does mean perhaps a *little* 
cooperation, in terms of holding off the release for a few weeks 
(assuming we have that legal right -- let's assume that for now).  I for 
one believe that this little cooperation is worth the benefit.  The 
alternate result is Java DB will be a fork, potentially not upgradeable, 
and there will be a lot of confusion in the marketplace.

I think that folks looking at this list would probably shake their heads 
at all the legal stuff we have to go through, but hopefully will see 
that we are doing our best to make the right decisions.

Thanks,

David

Kathey Marsden wrote:
> David Van Couvering wrote:
> 
>>
>> What happened to the proposal to vote and approve a GA-enabled 
>> release, but not make it actually available until Java SE 6 goes GA?  
>> Did we decide this was not feasible?  I may have missed it, but I 
>> don't think I saw that discussion anywhere.
>>
> This was my comment regarding my concerns about this dependency. 
> http://www.nabble.com/Re%3A-catch-22%3A-Derby%2C-Mustang%2C-and-JCP-issue-p5012231.html 
> 
> 
> I think from a community perspective we would like other new projects to 
> look at us as a good example.  Will they say Yes! Let's do it just like 
> Derby:
> - Wait more than a year for our first feature release,
> -  Make  pseudo GA release and ask our Release Manager to hold onto it 
> until  the company he works for releases their comercial product that 
> uses it.
> -  Set a precedent for external comercial dependencies on our release 
> criteria.
> 
> I think/hope  they won't.
> 
> Kathey
> 

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Kathey Marsden <km...@sbcglobal.net>.
David Van Couvering wrote:

>
> What happened to the proposal to vote and approve a GA-enabled 
> release, but not make it actually available until Java SE 6 goes GA?  
> Did we decide this was not feasible?  I may have missed it, but I 
> don't think I saw that discussion anywhere.
>
This was my comment regarding my concerns about this dependency. 
http://www.nabble.com/Re%3A-catch-22%3A-Derby%2C-Mustang%2C-and-JCP-issue-p5012231.html

I think from a community perspective we would like other new projects to 
look at us as a good example.  Will they say Yes! Let's do it just like 
Derby:
- Wait more than a year for our first feature release,
-  Make  pseudo GA release and ask our Release Manager to hold onto it 
until  the company he works for releases their comercial product that 
uses it.
-  Set a precedent for external comercial dependencies on our release 
criteria.

I think/hope  they won't.

Kathey


Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
David Van Couvering wrote:
> Andrew, thanks for putting in your time and research for this discussion.
> 
> Andrew McIntyre wrote:
...
> What happened to the proposal to vote and approve a GA-enabled release,
> but not make it actually available until Java SE 6 goes GA?  Did we
> decide this was not feasible?  I may have missed it, but I don't think I
> saw that discussion anywhere.

Assuming that the legal issues are resolved, I have a problem with
sitting on the RC for 5 weeks. Here's the schedule from Rick's original
post [1]:

> August 10 - Feature work committed. 10.2 branch created.
> August 24 - Last day to commit changes for 10.2
> August 25 - Begin vetting 10.2 release candidate
> September 15 - Target date for finishing the voting on Derby 10.2
> End of October - Expected GA of JDBC4 with Mustang
> End of October - GA of Derby 10.2. Release promoted to Apache mirrors.

So, Derby approves a RC by Sep 15 -- then sits on it for 5 weeks.

Once Derby approves a release, we should be able to release it. If we
legally can't release it, then we shouldn't even be producing a release
candidate and voting on it.

Perhaps if the legal issues are resolved in a way that permits producing
the release candidate, maybe it can also cover the actual release.

 -jean


[1]
http://mail-archives.apache.org/mod_mbox/db-derby-dev/200606.mbox/%3c449AEAA7.8030109@sun.com%3e

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
David Van Couvering wrote:

> What happened to the proposal to vote and approve a GA-enabled release,
> but not make it actually available until Java SE 6 goes GA?  Did we
> decide this was not feasible?  I may have missed it, but I don't think I
> saw that discussion anywhere.

That is the discussion we are having: Can we legally mark something as
GA before the JSR221 spec is GA.

Obviously we would be making it available since the point of this early
release candidate vote would be for the Mustang folks to get their hands
on it.

Dan.



Re: catch-22: Derby, Mustang, and JCP issue

Posted by David Van Couvering <Da...@Sun.COM>.
Andrew, thanks for putting in your time and research for this discussion.


Andrew McIntyre wrote:
[snip]
> 
> So, I can see any of the following things as possibilities:
> 
> 1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
> whenever it likes. Maybe even next week. :-)
> 2) Sun releases a version of Derby with JDK 1.6 that reports as
> 10.2.1.x and has upgrade enabled, however they want to do that.
> Hopefully they get this right and it upgrades to the official 10.2.
> 3) Once the JDBC 4 spec is final, the Derby community could release
> 10.2.(1|2).x + JDBC 4 bits pretty much any time it feels it is ready.

> 
> Regarding (1) and (3), I believe I've seen comments to the effect that
> there are members of the Derby community that would object to adding
> the JDBC 4 feature set in a maintenance release (which would be the
> 10.1.2 in (3) ) which supposedly would not contain any new features.
> 
> That's the way I see it at the moment. Opinions? Did I miss anything?
> (Probably. :-) )

I am pretty sure we have opinions saying no new features in a 
maintenance release.  So I am not sure your proposed approach will work. 
  Derby would have to wait until 10.3 to support JDBC4.

I guess this is an option, but it, I believe, would create a lot of 
confusion in the market, as Java DB in Mustang would be "almost Derby" 
but actually not based on any release of Derby, and would have to stay 
that way until 10.3.  It would also mean that Derby would not be able to 
provide valuable functionality (JDBC4) which is ready and working for 
another six months.

What happened to the proposal to vote and approve a GA-enabled release, 
but not make it actually available until Java SE 6 goes GA?  Did we 
decide this was not feasible?  I may have missed it, but I don't think I 
saw that discussion anywhere.

David

> 
> andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
Errata from my last post, changes marked between ** ;-)

On 6/22/06, Andrew McIntyre <mc...@gmail.com> wrote:
> ...
> Presumably this is where the idea that you can't ship something that
> implements an interface in a *non-final* JSR comes from
>  ...
> Sun wants to release a version of *JavaDB based on Derby* with the JDK
> that ... includes javadoc *from* Derby that includes material copied from
> the JDBC 4 spec.
> ...
> so that *the Derby engine in JavaDB* reports itself as a *10.2.1.x*
> ...
> 2) Sun *could* release a version of Derby *as JavaDB* with JDK 1.6
> ...
> Hopefully they get this right and it upgrades *correctly* to the
> official *Derby 10.2.x release that includes the JDBC 4 interfaces*.

There are a couple of other places where I meant to say JavaDB instead
of Derby and vice versa, but hey, it can be confusing. (Derby !=
Cloudscape) most of the time either. :-)

cheers,
andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/22/06, Andrew McIntyre <mc...@gmail.com> wrote:
> (which would be the 10.1.2 in (3) )

one last errata - this should read:

(which would be the 10.2.2.x in (3) )

a

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
> Andrew McIntyre wrote:
>
>
> > So, I can see any of the following things as possibilities:
> >
> > 1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
> > whenever it likes. Maybe even next week. :-)
>
> -1 The plan of record has been (and continues to be) shipping a 10.2
> Derby GA after Mustang releases.

Right. I just said it *could* happen. Not that it's likely. That's why
I said I didn't see the combination of 1 and 3 as something the
community was interested in. Given the circustances, what I really see
happening is 2 + 3, not all three. I guess that wasn't entirely clear
from the language.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Andrew McIntyre wrote:


> So, I can see any of the following things as possibilities:
> 
> 1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
> whenever it likes. Maybe even next week. :-)

-1 The plan of record has been (and continues to be) shipping a 10.2
Derby GA after Mustang releases. That's what I've been working to. I did
suggest (and some people agreed I think) that if Mustang slipped beyond
the initial timeframe of Sep/Oct 2006 we should consider shipping 10.2
in the Oct 2006 timeframe. Currently it seems Mustang still is in the
initial timeframe (though at the end of it).

Dan.




Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Geir Magnusson Jr wrote:
> Nice post.  Comments inline :
> 
> Andrew McIntyre wrote:
...
>>So, I can see any of the following things as possibilities:
>>
>>1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
>>whenever it likes. Maybe even next week. :-)
>>2) Sun releases a version of Derby with JDK 1.6 that reports as
>>10.2.1.x and has upgrade enabled, however they want to do that.
>>Hopefully they get this right and it upgrades to the official 10.2.
> 
> 
> What does it mean "upgrade enabled"?

I didn't spot that anybody answered this question.

The derby development code line has an alpha version tag that disables
upgrade in Derby.

If a user creates a "production" database using software built from the
development code line, then installs the official Derby release, the
official Derby release can't access that database, can't upgrade it to
the official release.

 -jean

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>
> > So, I can see any of the following things as possibilities:
> >
> > 1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
> > whenever it likes. Maybe even next week. :-)
> > 2) Sun releases a version of Derby with JDK 1.6 that reports as
> > 10.2.1.x and has upgrade enabled, however they want to do that.
> > Hopefully they get this right and it upgrades to the official 10.2.
>
> What does it mean "upgrade enabled"?

Derby's versioning mechanism prevents upgrading from anything with the
beta flag enabled or from a prerelease (x.x.0.x) version to a release
version (x.x.1.x). See
http://db.apache.org/derby/papers/versionupgrade.html

> > 3) Once the JDBC 4 spec is final, the Derby community could release
> > 10.2.(1|2).x + JDBC 4 bits pretty much any time it feels it is ready.
> >
> > Regarding (1) and (3), I believe I've seen comments to the effect that
> > there are members of the Derby community that would object to adding
> > the JDBC 4 feature set in a maintenance release (which would be the
> > 10.1.2 in (3) ) which supposedly would not contain any new features.
>
> Could you simply uptick the version as appropriate when JDBC4 is finally
> added?

I think that would be the 1 + 3 option that Dan objects to further
down and which I mentioned already has negative support in the
community. If you mean go to 10.3, well, yes, that is also an option,
but there was some negative feedback there as well about having two
major feature releases within a span of a couple of months.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.
Nice post.  Comments inline :

Andrew McIntyre wrote:
> On 6/22/06, Daniel John Debrunner <dj...@apache.org> wrote:
>>
>> Isn't an implementation of JSR221 writing (clean room) classes in the
>> java.sql and javax.sql name spaces. (e.g. java.sql.Driver &
>> javax.sql.DataSource).
>>
>> Derby is not doing that, Derby is providing an implementation of a JDBC
>> driver, not an implementation of JSR221 itself. Implementing JSR221 is
>> something that Harmony would (might) do.
>>
>> My point is that the JCP rules for implementation of the spec itself do
>> not apply here.
> 
> +1

I just can't agree that this is true as much as I want to, because the
draft license is clear, and the spec itself claims that any
implementation of a driver is a JDBC implementation.

However, it may not matter.

> 
> I agree, on rereading the JSPA agreement, the text I quoted in my
> previous email refers to independent implementations of the core
> classes covered by JSR-221. i.e. the java.sql.* classes, etc.
> Applications which happen to implement interfaces in those packages in
> their own namespace don't seem to be covered by anything in the JSPA.
> 
> But, looking at the evaluation license you need to agree to download
> the spec, it contains this:
> 
[snip spec license blurb]

> 
> Presumably this is where the idea that you can't ship something that
> implements an interface in a JSR comes from, because you would have
> needed to copy material which is copyrighted by Sun, and to which
> you've agreed to this license to have knowledge of. Let's assume that
> the people who implemented the Derby implementations of the JDBC 4
> interfaces are subject to this agreement. And, that this means that
> the method signatures described in the spec and any javadoc comments
> in the JDBC 4.0 spec are the IP in question here that we have
> 'acquired no right' to use. The method signatures in the Derby classes
> were clearly copied from the spec, and I believe perhaps some javadoc
> comments as well, although I would need to check that.
> 
> Now, there are two things involved in our catch-22:
> 
> * Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
> 
> Sun wants to release a version of Derby with the JDK that (I would
> assume) includes javadoc for Derby that includes material copied from
> the JDBC 4 spec. But if Sun shipping the JDK is the event which
> satisifes ' (ii) the date on which the final version of the
> Specification is publicly released' then the license above expires and
> the people who were reading the spec and copied parts of it into Derby
> are no longer bound by this agreement either. It's not clear what
> license is then in effect, but I would like to think that by virtue of
> the contributors in question, the ASL is the license in effect on that
> code, since it was contributed to Derby by employees of Sun under its
> CCLA and the various ICLAS in effect for the individual contributors.
> But then, IANAL.

The Apache License is the only copyright license on the code for
downstream licensees.  Nothing else - Sun's weird theories about spec
copyright notwithstanding.

Patents are another story, of course, but unless Sun or other
contributors to the 221 EG assert that there are necessary patents, we
should ignore them here.  (Which brings up an interesting situation for
discussion at another tim that if there are necessary patents held by an
EG member, and there is no TCK for which to certify compatibility of an
implementation, which the spec claims a driver is....)

> 
> As far as the Derby bits that Sun ships in the JDK, well, it's not
> really Derby they've committed to shipping but JavaDB. So they can
> twiddle the bits as they see fit I suppose, as long as they don't call
> it an official Derby release anywhere in the JDK. I could imagine a
> situation where they simply flip the beta flag and update the version
> so that Derby reports itself as a 10.2.0.0 database at whatever
> revision it happens to be the day the JDK ships, along with the
> modified 'M' flag in the version string. If Sun says that all that did
> was ship the Derby code at that revision level with the necessary
> version bits modified, then anyone using the Derby bundled with the
> JDK will just have to believe them.

This is where things start to smell like a fork.

If JavaDB behaves differently than Apache Derby for the same version
number, that's a problem.

> 
> If, say, that version happens to not upgrade to official Derby
> releases for some bizarre reason, then suddenly there are lots of new,
> irate users complaining that Derby is broken. This would make the
> Derby community look bad through no fault of its own (and Sun has a
> sticky problem to boot). Hopefully this won't be an issue, though.
> 
> * Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
> 
> One modification here: Derby could ship a GA 10.2 any time it likes.
> Just not with the JDBC 4 bits. Nor would it want to, anyway, since
> besides the evaluation license probably covering material that Derby
> would be releaseing, Derby could possibly ship incorrect
> implementations of interfaces defined in the spec, and nobody in the
> community is interested in that. :-)
> 
> But, once the spec is final, then the Derby community can release its
> JDBC 4 bits at its leisure, since the evaluation license has expired
> and presumably the bits in question (i hope) are then covered by the
> ASL.

Don't worry about the "expiration" mechanism - once the spec is final,
it's clear Derby can ship an implementation of it.

BTW, those "bits in question" are *always* covered by the Apache License
(it's not the "ASL" by the way - that was v1.x)  The issue is if the
project can distribute them.  The issue really is if it's legal to
implement, but Sun has acknowledged this problem with a new version of
the spec license this last May.

> 
> And, hopefully whatever Derby bits Sun releases as part of the JDK
> will be compatible with future Derby releases. The Derby community
> obviously has no responsibility for anything Sun releases, but it can
> definitely be affected, negatively or positively, by whether or not
> they get this 'right'. Let's hope they do.

Exactly.

> 
> So, I can see any of the following things as possibilities:
> 
> 1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
> whenever it likes. Maybe even next week. :-)
> 2) Sun releases a version of Derby with JDK 1.6 that reports as
> 10.2.1.x and has upgrade enabled, however they want to do that.
> Hopefully they get this right and it upgrades to the official 10.2.

What does it mean "upgrade enabled"?

> 3) Once the JDBC 4 spec is final, the Derby community could release
> 10.2.(1|2).x + JDBC 4 bits pretty much any time it feels it is ready.
> 
> Regarding (1) and (3), I believe I've seen comments to the effect that
> there are members of the Derby community that would object to adding
> the JDBC 4 feature set in a maintenance release (which would be the
> 10.1.2 in (3) ) which supposedly would not contain any new features.

Could you simply uptick the version as appropriate when JDBC4 is finally
added?

geir

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/22/06, Daniel John Debrunner <dj...@apache.org> wrote:
>
> Isn't an implementation of JSR221 writing (clean room) classes in the
> java.sql and javax.sql name spaces. (e.g. java.sql.Driver &
> javax.sql.DataSource).
>
> Derby is not doing that, Derby is providing an implementation of a JDBC
> driver, not an implementation of JSR221 itself. Implementing JSR221 is
> something that Harmony would (might) do.
>
> My point is that the JCP rules for implementation of the spec itself do
> not apply here.

+1

I agree, on rereading the JSPA agreement, the text I quoted in my
previous email refers to independent implementations of the core
classes covered by JSR-221. i.e. the java.sql.* classes, etc.
Applications which happen to implement interfaces in those packages in
their own namespace don't seem to be covered by anything in the JSPA.

But, looking at the evaluation license you need to agree to download
the spec, it contains this:

Subject to the terms and conditions of this license,
Sun hereby grants you a fully-paid, non-exclusive,
non-transferable, limited license (without the right
to sublicense) under Sun's intellectual property
rights to review the Specification only for the
purposes of evaluation. This license includes the
right to discuss the Specification (including the
right to provide limited excerpts of text to the
extent relevant to the point[s] under discussion) with
other licensees (under this or a substantially similar
version of this Agreement) of the Specification. *Other
than this limited license, you acquire no right, title
or interest in or to the Specification or any other
Sun intellectual property*, and the Specification may
only be used in accordance with the license terms set
forth herein. This license will expire on the earlier
of: ... (ii) the date on which the final version of the
Specification is publicly released;

Presumably this is where the idea that you can't ship something that
implements an interface in a JSR comes from, because you would have
needed to copy material which is copyrighted by Sun, and to which
you've agreed to this license to have knowledge of. Let's assume that
the people who implemented the Derby implementations of the JDBC 4
interfaces are subject to this agreement. And, that this means that
the method signatures described in the spec and any javadoc comments
in the JDBC 4.0 spec are the IP in question here that we have
'acquired no right' to use. The method signatures in the Derby classes
were clearly copied from the spec, and I believe perhaps some javadoc
comments as well, although I would need to check that.

Now, there are two things involved in our catch-22:

* Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.

Sun wants to release a version of Derby with the JDK that (I would
assume) includes javadoc for Derby that includes material copied from
the JDBC 4 spec. But if Sun shipping the JDK is the event which
satisifes ' (ii) the date on which the final version of the
Specification is publicly released' then the license above expires and
the people who were reading the spec and copied parts of it into Derby
are no longer bound by this agreement either. It's not clear what
license is then in effect, but I would like to think that by virtue of
the contributors in question, the ASL is the license in effect on that
code, since it was contributed to Derby by employees of Sun under its
CCLA and the various ICLAS in effect for the individual contributors.
But then, IANAL.

As far as the Derby bits that Sun ships in the JDK, well, it's not
really Derby they've committed to shipping but JavaDB. So they can
twiddle the bits as they see fit I suppose, as long as they don't call
it an official Derby release anywhere in the JDK. I could imagine a
situation where they simply flip the beta flag and update the version
so that Derby reports itself as a 10.2.0.0 database at whatever
revision it happens to be the day the JDK ships, along with the
modified 'M' flag in the version string. If Sun says that all that did
was ship the Derby code at that revision level with the necessary
version bits modified, then anyone using the Derby bundled with the
JDK will just have to believe them.

If, say, that version happens to not upgrade to official Derby
releases for some bizarre reason, then suddenly there are lots of new,
irate users complaining that Derby is broken. This would make the
Derby community look bad through no fault of its own (and Sun has a
sticky problem to boot). Hopefully this won't be an issue, though.

* Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.

One modification here: Derby could ship a GA 10.2 any time it likes.
Just not with the JDBC 4 bits. Nor would it want to, anyway, since
besides the evaluation license probably covering material that Derby
would be releaseing, Derby could possibly ship incorrect
implementations of interfaces defined in the spec, and nobody in the
community is interested in that. :-)

But, once the spec is final, then the Derby community can release its
JDBC 4 bits at its leisure, since the evaluation license has expired
and presumably the bits in question (i hope) are then covered by the
ASL.

And, hopefully whatever Derby bits Sun releases as part of the JDK
will be compatible with future Derby releases. The Derby community
obviously has no responsibility for anything Sun releases, but it can
definitely be affected, negatively or positively, by whether or not
they get this 'right'. Let's hope they do.

So, I can see any of the following things as possibilities:

1) The Derby community could release a 10.2.1.x minus the JDBC 4 bits
whenever it likes. Maybe even next week. :-)
2) Sun releases a version of Derby with JDK 1.6 that reports as
10.2.1.x and has upgrade enabled, however they want to do that.
Hopefully they get this right and it upgrades to the official 10.2.
3) Once the JDBC 4 spec is final, the Derby community could release
10.2.(1|2).x + JDBC 4 bits pretty much any time it feels it is ready.

Regarding (1) and (3), I believe I've seen comments to the effect that
there are members of the Derby community that would object to adding
the JDBC 4 feature set in a maintenance release (which would be the
10.1.2 in (3) ) which supposedly would not contain any new features.

That's the way I see it at the moment. Opinions? Did I miss anything?
(Probably. :-) )

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Geir Magnusson Jr wrote:
> Daniel John Debrunner wrote:
> 
>>Jean T. Anderson wrote:
>>
>>
>>>David posted a good summary of the legal catch-22 at [1]. But the
>>>shortest story is:
>>>
>>> + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>>> + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
>>>
>>>Let's keep this thread confined to the JCP issue Andrew raised that to
>>>roll a release candidate qualifies as "creation".[2] And those release
>>>candidates will be generally available.
>>
>>I don't think the JCP rules apply since we are not creating an
>>implementation of JSR221.
> 
> 
> How could that be?  Where did the information for the APIs come from?

Isn't an implementation of JSR221 writing (clean room) classes in the
java.sql and javax.sql name spaces. (e.g. java.sql.Driver &
javax.sql.DataSource).

Derby is not doing that, Derby is providing an implementation of a JDBC
driver, not an implementation of JSR221 itself. Implementing JSR221 is
something that Harmony would (might) do.

My point is that the JCP rules for implementation of the spec itself do
not apply here.

>>I'm sure other rules apply that say one cannot ship a JDBC 4.0 driver
>>until Mustang goes GA, but that's not the JCP rules that Andrew provided
>>references to. It would be good to get Lance to expand on what he meant
>>when he said:
>>
>>"You cannot have a GA version of a JDBC 4 driver until JSR 221 goes final."
>>
>>Where does this restriction come from?
> 
> 
> I'm not sure.  The license on the public review draft gives no rights to
> distribute an implementation under *any* label (more recent public
> review draft licenses  do now after the whole EJB3 brouhaha)

My point is "implementation" of what exactly. I think Lance (JSR221 spec
lead) has already said that the TCK for JSR221 just checks the the
classes (java.sql etc.) exist and have the correct methods & signatures,
that implies to me that implementing JSR221 means providing those classes.

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.
Daniel John Debrunner wrote:
> Jean T. Anderson wrote:
> 
>> David posted a good summary of the legal catch-22 at [1]. But the
>> shortest story is:
>>
>>  + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>>  + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
>>
>> Let's keep this thread confined to the JCP issue Andrew raised that to
>> roll a release candidate qualifies as "creation".[2] And those release
>> candidates will be generally available.
> 
> I don't think the JCP rules apply since we are not creating an
> implementation of JSR221.

How could that be?  Where did the information for the APIs come from?

> 
> I'm sure other rules apply that say one cannot ship a JDBC 4.0 driver
> until Mustang goes GA, but that's not the JCP rules that Andrew provided
> references to. It would be good to get Lance to expand on what he meant
> when he said:
> 
> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes final."
> 
> Where does this restriction come from?

I'm not sure.  The license on the public review draft gives no rights to
distribute an implementation under *any* label (more recent public
review draft licenses  do now after the whole EJB3 brouhaha)

> 
> Dan.
> 
>> How do we handle this cleanly?

I would assume that you just don't ship a JDBC 4 driver with the release
of Derby, and do a release when the spec is final and a TCK can be
obtained from the spec lead to ensure that the implementation is
compatible with the spec.

It's just a driver.  The database is fully functional w/o it, right?

Seems to be fairly simple.

geir

>>
>> -jean
>>
>>
>> [1]
>> http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
>> [2] http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
>>
> 
> 
> 
> 

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Kathey Marsden <km...@sbcglobal.net>.
[snip lot's of interesting legal conversation that I can't keep up with]

To me,  the core question for the Derby community is do we want to  
interject an  extended commercial external dependency  into our release 
cycle.   To that I vote -1.   It seems contrary to the Apache Way and 
all that makes it work:  incremental development, contributor 
independence,  release early often.  It also seems to make this  a legal 
rather than technical forum. 

There are perfectly reasonable technical release alternatives but there 
seem to be legal issue's getting in the way.    I suggest the lawyers  
take this off line, and find a solution so Derby can release with the 
implementation based on the draft and subject to change.  The lawyers on 
both sides of the table  seem to work at the same place, so  I am sure  
negotiations can't be too difficult.  If they  are,  then I think Java 
DB will just need to fork of temporarily and sync back up to us when 
ready to play again the Apache Way.

10.2 has already been far too long of a release cycle.   In large part 
the delays already are related to this legal issue and I think we need 
to stop it.  I seen 10.2  as our *real* graduation from incubation:  
Will we as a community be able push out a real  Apache feature release 
together and feel like we did it the Apache Way?   Whatever precedents 
we set this time we should be willing to carry forward and  this seems 
like a bad one to me.

Kathey





Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Daniel John Debrunner wrote:
> Brian McCallister wrote:
> 
>> On Jun 22, 2006, at 7:09 PM, Daniel John Debrunner wrote:
>>
>>> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes 
>>> final."
>>>
>>> Where does this restriction come from?
>>
>> Until a spec is final I don't see how you can have a certified compliant
>> implementation of that spec. 
> 
> Agreed, and I'm not sure that there is any requirement for Derby to have
> a certified compliant JDBC driver. Sure it would be great to have a
> certified driver, but it's not a requirement for a Derby release.

Exactly.

> 
>> It might be as easy as not claiming JDBC  4.0
>> compliance until *after* JSR-221 is final, than saying, "huh, gee whiz,
>> we pass the TCK!" 
> 
> Except there is no TCK for compliance of a JDBC 4.0 driver. I believe
> the compliance for a driver is only described by text in the JDBC 4.0
> spec. The JSR221 TCK is for compliance of the JDBC 4.0 apis.

You're right - I just read that part of the spec. That's interesting.

The document does state that a JDBC implementation includes the driver
and its underlying data source.

How odd.  So, in my last msg on this thread, I agreed with you that it
wasn't an implementation, but according to the spec itself, it is.

> 
>> If the interfaces happen to exist in a release before the spec is final,
>> well, cool. Folks using them are at risk of the spec changing at the  last
>> minute, so I would put bright red warnings around them if they are event
>> documented before the official release of the spec (not of Derby).
> 
> +1

In this case, we can argue :

- that it's moot as Derby doesn't offer any code in the java.* or
javax.* namespace defined by JDBC4 spec

yet

- the JDBC4 spec itself claims that a JDBC4 driver *is* an
implementation of the spec

How fun.

geir


> 
> Dan.
> 
> 
> 
> 

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Brian McCallister wrote:

> 
> On Jun 22, 2006, at 7:09 PM, Daniel John Debrunner wrote:
> 
>> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes 
>> final."
>>
>> Where does this restriction come from?
> 
> 
> Until a spec is final I don't see how you can have a certified compliant
> implementation of that spec. 

Agreed, and I'm not sure that there is any requirement for Derby to have
a certified compliant JDBC driver. Sure it would be great to have a
certified driver, but it's not a requirement for a Derby release.

> It might be as easy as not claiming JDBC  4.0
> compliance until *after* JSR-221 is final, than saying, "huh, gee whiz,
> we pass the TCK!" 

Except there is no TCK for compliance of a JDBC 4.0 driver. I believe
the compliance for a driver is only described by text in the JDBC 4.0
spec. The JSR221 TCK is for compliance of the JDBC 4.0 apis.

> If the interfaces happen to exist in a release before the spec is final,
> well, cool. Folks using them are at risk of the spec changing at the  last
> minute, so I would put bright red warnings around them if they are event
> documented before the official release of the spec (not of Derby).

+1

Dan.



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.
Brian McCallister wrote:
> 
> On Jun 22, 2006, at 7:09 PM, Daniel John Debrunner wrote:
> 
>> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes
>> final."
>>
>> Where does this restriction come from?
> 
> Until a spec is final I don't see how you can have a certified compliant
> implementation of that spec. It might be as easy as not claiming JDBC 4.0
> compliance until *after* JSR-221 is final, than saying, "huh, gee whiz,
> we pass the TCK!"

The only problem I see is that by the letter of the law, there is no
right to implement and distribute JDBC4's public draft.

I guess the first thing to figure out is what from the JDBC4 expert
group has actually been implemented.

> 
> If the interfaces happen to exist in a release before the spec is final,
> well, cool.

Sun used to have some strange arguments about spec API copyrights which
made that uncool unless you passed the TCK (which implied completion of
the JSR....)

> Folks using them are at risk of the spec changing at the last
> minute, so I would put bright red warnings around them if they are event
> documented before the official release of the spec (not of Derby).

Or just leave JDBC4 out of the GA release, and given it's only a
database driver, just offer it as a second jar when it's possible to do so.

Why tie the DB release cycle to an updated driver?  Won't a JDBC3 driver
work well too?

geir


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Kathey Marsden wrote:
> Brian McCallister wrote:
> 
>> If the interfaces happen to exist in a release before the spec is final,
>> well, cool. Folks using them are at risk of the spec changing at the 
>> last
>> minute, so I would put bright red warnings around them if they are event
>> documented before the official release of the spec (not of Derby).
>>
> +1
> - We have a Derby 10.2 GA release with any JDBC 4.0 functionality marked
> as USE AT YOUR OWN RISK, just part of incremental devlopment and totally
> subject to change if the spec does.
> - Mustang takes that and runs with it.
> - Then we have a follow up release, yes even two months later where we
> are certified compliant.
> 
> A nice clean 2 phase commit.  I haven't read the legal stuff, but does
> this not work?

It works, but if you want to be technical about it, not really.  Because
of the license Sun chose for the public draft for JDBC4, you can't
actually implement and distribute it.

Of course, as the licensor, Sun could offer a 'clarification' of that
license and allow implementations of the public draft, and that would at
least let us legally ship it.

The cleanest way seem to just remove the JDBC4 stuff from the upcoming
GA release, and then release a JDBC4 driver as fast as you can once you
have the final spec and a TCK.

geir


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Kathey Marsden <km...@sbcglobal.net>.
Brian McCallister wrote:

> If the interfaces happen to exist in a release before the spec is final,
> well, cool. Folks using them are at risk of the spec changing at the  
> last
> minute, so I would put bright red warnings around them if they are event
> documented before the official release of the spec (not of Derby).
>
+1
- We have a Derby 10.2 GA release with any JDBC 4.0 functionality marked 
as USE AT YOUR OWN RISK, just part of incremental devlopment and totally 
subject to change if the spec does.
- Mustang takes that and runs with it.
- Then we have a follow up release, yes even two months later where we 
are certified compliant.

A nice clean 2 phase commit.  I haven't read the legal stuff, but does 
this not work?

Kathey







Re: catch-22: Derby, Mustang, and JCP issue

Posted by Brian McCallister <br...@apache.org>.
On Jun 22, 2006, at 7:09 PM, Daniel John Debrunner wrote:

> "You cannot have a GA version of a JDBC 4 driver until JSR 221 goes  
> final."
>
> Where does this restriction come from?

Until a spec is final I don't see how you can have a certified compliant
implementation of that spec. It might be as easy as not claiming JDBC  
4.0
compliance until *after* JSR-221 is final, than saying, "huh, gee whiz,
we pass the TCK!"

If the interfaces happen to exist in a release before the spec is final,
well, cool. Folks using them are at risk of the spec changing at the  
last
minute, so I would put bright red warnings around them if they are event
documented before the official release of the spec (not of Derby).

-Brian

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Jean T. Anderson wrote:

> David posted a good summary of the legal catch-22 at [1]. But the
> shortest story is:
> 
>  + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>  + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
> 
> Let's keep this thread confined to the JCP issue Andrew raised that to
> roll a release candidate qualifies as "creation".[2] And those release
> candidates will be generally available.

I don't think the JCP rules apply since we are not creating an
implementation of JSR221.

I'm sure other rules apply that say one cannot ship a JDBC 4.0 driver
until Mustang goes GA, but that's not the JCP rules that Andrew provided
references to. It would be good to get Lance to expand on what he meant
when he said:

"You cannot have a GA version of a JDBC 4 driver until JSR 221 goes final."

Where does this restriction come from?

Dan.

> 
> How do we handle this cleanly?
> 
> -jean
> 
> 
> [1]
> http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
> [2] http://www.nabble.com/Proposal-for-10.2-release-schedule-t1832249.html
> 



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andreas Korneliussen <an...@sun.com>.
> and does the close of the successful vote mark the change to GA? Seems
> it's the only logical point, the announce and distribution via the
> mirrors are just action items after the GA decision is made.
> 
The vote could be to go GA on a specific date.

Andreas

> Dan.
> 


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Jean T. Anderson wrote:

> Andreas Korneliussen wrote:

>>As for alpha/beta bits: they should not be set on a release candidate.
>>The vetted bits are the final bits if the community votes for the
>>release.  As far as I know, they were not set on the recently rejected
>>10.1.3 RC.

The 10.1 (in fact any) branch is always marked GA since its first
official release.

> Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
> GA? --Understanding that the RC will be readily available to anyone to
> download?

and does the close of the successful vote mark the change to GA? Seems
it's the only logical point, the announce and distribution via the
mirrors are just action items after the GA decision is made.

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Kathey Marsden wrote:


> I know I do and I think it is good to facilitate it with a plan like
> this, but not to be held hostage by it by creating  a special "not GA
> for anyone but Sun" release.

Well put.

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Kathey Marsden <km...@sbcglobal.net>.
Daniel John Debrunner wrote:

>  
>
>>Looks like, barring clarification by Sun what the spec license means
>>concerning applications that implement interfaces in a non-final JSR,
>>even in their own namespace, the Derby community shouldn't do anything
>>regarding a release with JDBC 4 - i would say not even post a release
>>candidate for a vote - until the JDBC 4 spec is final.
>>    
>>
>
>+1 and this means no catch-22 for Apache Derby.
>
>Sun's plans to have Sun's JavaDB in Sun's Java SE 6 Developers Kit are
>Sun's problem.
>
>I do think the community can work with Sun to help the goal of a
>distribution of Derby based technology in Sun's JDK. I think the
>required steps are something like:
>   - 10.2 branch creation as Rick indicated (timing wise)
>   - testing on the 10.2 branch by everyone with snapshots and beta
>release candidate(s)
>   - minor fork by Sun to produce a GA release of *Java DB* to be
>released with Mustang GA
>   - Sun assumes risk of any disk format/api changes between their
>release of JavaDB and the Derby 10.2 release, lots of early testing here
>by Sun (& others) can help reduce this risk
>
>  
>
Sounds good to me and I think even we could interject a separate vote to 
finalize disk format  on the 10.2 branch
before JavaDB forks off to help the reduce the risk further.

>Of course I assuming that the community thinks it is good for Derby
>technology to be shipped with Sun's Java 6 Developers Kit, but there
>actually hasn't been any discussion either way on that point.
>
>  
>
I know I do and I think it is good to facilitate it with a plan like 
this, but not to be held hostage by it by creating  a special "not GA 
for anyone but Sun" release.


Kathey



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Andrew McIntyre wrote:

> Looks like, barring clarification by Sun what the spec license means
> concerning applications that implement interfaces in a non-final JSR,
> even in their own namespace, the Derby community shouldn't do anything
> regarding a release with JDBC 4 - i would say not even post a release
> candidate for a vote - until the JDBC 4 spec is final.

+1 and this means no catch-22 for Apache Derby.

Sun's plans to have Sun's JavaDB in Sun's Java SE 6 Developers Kit are
Sun's problem.

I do think the community can work with Sun to help the goal of a
distribution of Derby based technology in Sun's JDK. I think the
required steps are something like:
   - 10.2 branch creation as Rick indicated (timing wise)
   - testing on the 10.2 branch by everyone with snapshots and beta
release candidate(s)
   - minor fork by Sun to produce a GA release of *Java DB* to be
released with Mustang GA
   - Sun assumes risk of any disk format/api changes between their
release of JavaDB and the Derby 10.2 release, lots of early testing here
by Sun (& others) can help reduce this risk

Of course I assuming that the community thinks it is good for Derby
technology to be shipped with Sun's Java 6 Developers Kit, but there
actually hasn't been any discussion either way on that point.

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Andreas Korneliussen wrote:
...
> Anyone can check out the Derby source tree, and build a "derby" with the
> GA bit set at any time. If anyone does that, it clearly does not mean
> that Derby has gone GA with JDBC4, and that the community is held
> legally responsible.

Anyone can build Derby, yes.

Anyone who builds Derby with the intent of producing a release candidate
that will voted on by the community to become an official Apache release
is now working within the Derby community.

As for who is held responsible ... the DB PMC has oversight over the
Derby project to make sure that legal bases are covered [1]:

> The role of the PMC from a Foundation perspective is oversight. The main role of the PMC is not code and not coding - but to ensure that all legal issues are addressed, that procedure is followed, and that each and every release is the product of the community as a whole. That is key to our litigation protection mechanisms.

Please don't take it amiss if some of us on the DB PMC are aggressively
jumping in regarding these legal issues. We know the intent of this
release candidate and it's our responsibility to make sure legal issues
get resolved correctly.

 -jean

[1] http://www.apache.org/foundation/how-it-works.html#structure






Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Andrew McIntyre wrote:
> On 6/23/06, Andreas Korneliussen <an...@sun.com> wrote:
...
>> I responded to this part of your post, and wanted to clarify that Derby
>> does not build the RC, it is the RM who does it.
> 
> Yes, because it is the RM as an individual that ultimately builds and
> publishes the release as an official Apache release on behalf of the
> ASF. Once it is published, though, the bits belong to the ASF, not the
> release manager or the Derby project. I'm pretty sure that's what Jean
> meant by 'Apache is ultimately responsible for the release'. 

One of the reasons the Apache Software Foundation was formed was to
provide a legal umbrella that protects individual software developers
from litigation [1]:

> provide a means for individual volunteers to be sheltered from legal suits directed at the Foundation's projects

I don't think it was anybody's intention to suggest throwing our RMs to
the lions .... but let's don't.  :-) We should always work on releases
within ASF guidelines.

 -jean


[1] http://www.apache.org/foundation/how-it-works.html, see "What is the
Apache Software Foundation?"



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andreas Korneliussen <an...@sun.com>.
>>True. But we're not talking about just anybody. If a member of the
>>Derby community does put together a release candidate (that would need
>>to report itself as a GA version) and publishes it on a website for
>>others to vote on, then the community could be held responsible.
> 
> 
> For what?
> 

As far as I can tell, the legal issues are about when a product goes GA. 
I.e Derby 10.2 could not go GA until JDBC4 spec is final.

If we follow Rick's proposal for a release schedule, I would argue that 
this would not be a problem, since Derby would go GA after the JDBC 4 
spec is final.

In this thread however, it seems like the meaning of GA is debated / 
redefined, as to mean "when someone builds the RC", or "someone set a GA 
flag in SVN", and therefore argues that there potential legal issues 
w.r.t to JCP. Please correct me if I have misunderstood, or if there 
were other things the community could be held responsible for.


Andreas

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Andrew McIntyre wrote:
> On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>>
>> Andrew McIntyre wrote:
>>
>> > Yes, because it is the RM as an individual that ultimately builds and
>> > publishes the release as an official Apache release on behalf of the
>> > ASF.
>>
>> Absolutely not.
> 
> Right, I know I should have put the PMC in there, too, sorry I left it
> out. I understand that this is a very important legal distinction and
> that the RM is just the messenger/agent of the PMC when it comes to
> releasing software for the ASF.

More like a go-fer :)

Until the PMC votes to release software, it's not an Apache release.


> 
>> > True. But we're not talking about just anybody. If a member of the
>> > Derby community does put together a release candidate (that would need
>> > to report itself as a GA version) and publishes it on a website for
>> > others to vote on, then the community could be held responsible.
>>
>> For what?
> 
> For violating the JSPA or the spec evaluation agreement as far as
> creating/distributing an implementation, which is where people got the
> idea that we couldn't publish a GA version of Derby that had JDBC 4.0
> bits in it. Although it appears we've now determined that Derby has no
> obligations with regard to these agreements as to when it can do a
> release.

Until it's an Apache action, the JSPA doesn't matter, because those
people aren't signatories of the JSPA.

Sure, there is some grey area here, such as because the ASF has a public
repository containing JSR implementations-in-progress - but as we work
under a bona fide open source methodology, everyone understands that
isn't "distributing"...

geir


Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Lance J. Andersen" <La...@Sun.COM>.

Geir Magnusson Jr wrote:
> Andrew McIntyre wrote:
>   
>> On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
>>     
>>>> In #2 of his proposed solution, Geir said he doesn't believe that
>>>> Derby qualifies as an implementation, and thus would not be affected
>>>> by the JSPA.
>>>>         
>>> I thought Geir's proposed solution was predicated on item 1)
>>>
>>> Geir wrote:
>>>       
>>>> 1) Have Sun change the draft spec license for 221 from the current to
>>>> the new one that allows distribution with appropriate warning markings.
>>>>  I'm going to start working this line w/ the PMO and the JCP.
>>>>         
>>> Until the licence is changed we cannot ship a GA version of Derby with
>>> JDBC 4.0 code.
>>>       
>> Then I'm confused, if we're not an implementation, thus not subject to
>> section 5 of the terms in the JSPA, and the copyright concerns w/r/t
>> the evaluation license are not an issue for us, then why does the spec
>> draft license need to change? Can somebody spell that out for me?
>>     
>
> Derby isn't an implementation, but there is a small piece that
> implements the JDBC4 spec.
>
>   
>> It certainly seems like changing the spec license is the right thing
>> to do to make everybody happy. So, can someone from Sun or JCP please
>> confirm that the draft spec license will in fact be changed?
>>     
>
> I've made the request formally.  As I said in a follow-up, the solution
> that will be easier will be a permissive license for the upcoming
> proposed final draft.
>   
Geir and I have spoken and i have also discussed it internally and we 
are going to look at updating the license for the PFD
>> I guess that, yes, we still cannot ship a GA version of Derby with the
>> JDBC 4 until another draft of the spec is posted with the new license
>> attached.
>>
>> andrew
>>
>>
>>     

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Andrew McIntyre wrote:
> On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
>>
>> > In #2 of his proposed solution, Geir said he doesn't believe that
>> > Derby qualifies as an implementation, and thus would not be affected
>> > by the JSPA.
>>
>> I thought Geir's proposed solution was predicated on item 1)
>>
>> Geir wrote:
>> > 1) Have Sun change the draft spec license for 221 from the current to
>> > the new one that allows distribution with appropriate warning markings.
>> >  I'm going to start working this line w/ the PMO and the JCP.
>>
>> Until the licence is changed we cannot ship a GA version of Derby with
>> JDBC 4.0 code.
> 
> Then I'm confused, if we're not an implementation, thus not subject to
> section 5 of the terms in the JSPA, and the copyright concerns w/r/t
> the evaluation license are not an issue for us, then why does the spec
> draft license need to change? Can somebody spell that out for me?

Derby isn't an implementation, but there is a small piece that
implements the JDBC4 spec.

> 
> It certainly seems like changing the spec license is the right thing
> to do to make everybody happy. So, can someone from Sun or JCP please
> confirm that the draft spec license will in fact be changed?

I've made the request formally.  As I said in a follow-up, the solution
that will be easier will be a permissive license for the upcoming
proposed final draft.
> 
> I guess that, yes, we still cannot ship a GA version of Derby with the
> JDBC 4 until another draft of the spec is posted with the new license
> attached.
> 
> andrew
> 
> 

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
>
> > In #2 of his proposed solution, Geir said he doesn't believe that
> > Derby qualifies as an implementation, and thus would not be affected
> > by the JSPA.
>
> I thought Geir's proposed solution was predicated on item 1)
>
> Geir wrote:
> > 1) Have Sun change the draft spec license for 221 from the current to
> > the new one that allows distribution with appropriate warning markings.
> >  I'm going to start working this line w/ the PMO and the JCP.
>
> Until the licence is changed we cannot ship a GA version of Derby with
> JDBC 4.0 code.

Then I'm confused, if we're not an implementation, thus not subject to
section 5 of the terms in the JSPA, and the copyright concerns w/r/t
the evaluation license are not an issue for us, then why does the spec
draft license need to change? Can somebody spell that out for me?

It certainly seems like changing the spec license is the right thing
to do to make everybody happy. So, can someone from Sun or JCP please
confirm that the draft spec license will in fact be changed?

I guess that, yes, we still cannot ship a GA version of Derby with the
JDBC 4 until another draft of the spec is posted with the new license
attached.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Andrew McIntyre wrote:

> On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
> 
>> Andrew McIntyre wrote:
>>
>> >
>> > For violating the JSPA or the spec evaluation agreement as far as
>> > creating/distributing an implementation, which is where people got the
>> > idea that we couldn't publish a GA version of Derby that had JDBC 4.0
>> > bits in it. Although it appears we've now determined that Derby has no
>> > obligations with regard to these agreements as to when it can do a
>> > release.
>>
>> I must have missed that, when was that detemrined?
> 
> 
> In #2 of his proposed solution, Geir said he doesn't believe that
> Derby qualifies as an implementation, and thus would not be affected
> by the JSPA.

I thought Geir's proposed solution was predicated on item 1)

Geir wrote:
> 1) Have Sun change the draft spec license for 221 from the current to
> the new one that allows distribution with appropriate warning markings.
>  I'm going to start working this line w/ the PMO and the JCP.

Until the licence is changed we cannot ship a GA version of Derby with
JDBC 4.0 code.

Dan.




Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.
I'm _hoping_ there aren't holes here.  Seems like a reasonable solution...

Andrew McIntyre wrote:
> On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>>
>> The key is #1.  Get it so that you can actually distribute an impl w/
>> proper labeling, and then consider the JDBC4 functionality just one
>> minor feature in a much bigger codebase.
>>
>> Then, since Derby isn't an implementation of JDBC4, but rather it's
>> driver is, Derby doesn't have to be labeled according to the JDBC4 spec
>> license terms.  Just the driver does.
>>
>> So Derby is a GA, a minor feature (JDBC4) isn't, it's still a feature,
>> so it can be updated w/o a dot-version uptick...
> 
> Ah! Got it.
> 
> Thanks, Geir!
> 
> andrew
> 
> 

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>
> The key is #1.  Get it so that you can actually distribute an impl w/
> proper labeling, and then consider the JDBC4 functionality just one
> minor feature in a much bigger codebase.
>
> Then, since Derby isn't an implementation of JDBC4, but rather it's
> driver is, Derby doesn't have to be labeled according to the JDBC4 spec
> license terms.  Just the driver does.
>
> So Derby is a GA, a minor feature (JDBC4) isn't, it's still a feature,
> so it can be updated w/o a dot-version uptick...

Ah! Got it.

Thanks, Geir!

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Andrew McIntyre wrote:
> On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
>> Andrew McIntyre wrote:
>>
>> >
>> > For violating the JSPA or the spec evaluation agreement as far as
>> > creating/distributing an implementation, which is where people got the
>> > idea that we couldn't publish a GA version of Derby that had JDBC 4.0
>> > bits in it. Although it appears we've now determined that Derby has no
>> > obligations with regard to these agreements as to when it can do a
>> > release.
>>
>> I must have missed that, when was that detemrined?
> 
> In #2 of his proposed solution, Geir said he doesn't believe that
> Derby qualifies as an implementation, and thus would not be affected
> by the JSPA.

The key is #1.  Get it so that you can actually distribute an impl w/
proper labeling, and then consider the JDBC4 functionality just one
minor feature in a much bigger codebase.

Then, since Derby isn't an implementation of JDBC4, but rather it's
driver is, Derby doesn't have to be labeled according to the JDBC4 spec
license terms.  Just the driver does.

So Derby is a GA, a minor feature (JDBC4) isn't, it's still a feature,
so it can be updated w/o a dot-version uptick...

> 
> Geir also said that as far as the copyright claims in the spec
> evaluation license were not of consequence for us.
> 
> From:
> http://mail-archives.apache.org/mod_mbox/db-derby-dev/200606.mbox/%3c449BCE21.2000207@pobox.com%3e
> 
> 
> "The Apache License is the only copyright license on the code for
> downstream licensees.  Nothing else - Sun's weird theories about spec
> copyright notwithstanding."

I think I said they were of no consequence to downstream licensees -
they are only accepting the terms of the Apache license.

geir

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Daniel John Debrunner <dj...@apache.org> wrote:
> Andrew McIntyre wrote:
>
> >
> > For violating the JSPA or the spec evaluation agreement as far as
> > creating/distributing an implementation, which is where people got the
> > idea that we couldn't publish a GA version of Derby that had JDBC 4.0
> > bits in it. Although it appears we've now determined that Derby has no
> > obligations with regard to these agreements as to when it can do a
> > release.
>
> I must have missed that, when was that detemrined?

In #2 of his proposed solution, Geir said he doesn't believe that
Derby qualifies as an implementation, and thus would not be affected
by the JSPA.

Geir also said that as far as the copyright claims in the spec
evaluation license were not of consequence for us.

From: http://mail-archives.apache.org/mod_mbox/db-derby-dev/200606.mbox/%3c449BCE21.2000207@pobox.com%3e

"The Apache License is the only copyright license on the code for
downstream licensees.  Nothing else - Sun's weird theories about spec
copyright notwithstanding."

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Andrew McIntyre wrote:

> 
> For violating the JSPA or the spec evaluation agreement as far as
> creating/distributing an implementation, which is where people got the
> idea that we couldn't publish a GA version of Derby that had JDBC 4.0
> bits in it. Although it appears we've now determined that Derby has no
> obligations with regard to these agreements as to when it can do a
> release.

I must have missed that, when was that detemrined?

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Geir Magnusson Jr <ge...@pobox.com> wrote:
>
> Andrew McIntyre wrote:
>
> > Yes, because it is the RM as an individual that ultimately builds and
> > publishes the release as an official Apache release on behalf of the
> > ASF.
>
> Absolutely not.

Right, I know I should have put the PMC in there, too, sorry I left it
out. I understand that this is a very important legal distinction and
that the RM is just the messenger/agent of the PMC when it comes to
releasing software for the ASF.

> > True. But we're not talking about just anybody. If a member of the
> > Derby community does put together a release candidate (that would need
> > to report itself as a GA version) and publishes it on a website for
> > others to vote on, then the community could be held responsible.
>
> For what?

For violating the JSPA or the spec evaluation agreement as far as
creating/distributing an implementation, which is where people got the
idea that we couldn't publish a GA version of Derby that had JDBC 4.0
bits in it. Although it appears we've now determined that Derby has no
obligations with regard to these agreements as to when it can do a
release.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Andrew McIntyre wrote:
> On 6/23/06, Andreas Korneliussen <an...@sun.com> wrote:
>> Andrew McIntyre wrote:
>> > On 6/23/06, Jean T. Anderson <jt...@bristowhill.com> wrote:
>> >
>> >> Andreas Korneliussen wrote:
>> >> > Jean T. Anderson wrote:
>> >> >
>> >> > In this case, it is Rick (the release manager) who builds the RC,
>> and
>> >> > who makes it available before JDBC4 goes GA. I guess it would be
>> up to
>> >> > him to take the legal responsibility.
>> >
>> >
>> > The RM for any Apache release is responsible for evaluating the
>> > legalities of the situation surrounding a release and do what is
>> > necessary to protect the ASF from legal action. See sections 4 and 8
>> > of the ICLA.
>> >
>>
>> Thanks for confirming that it is the RM who is responsible for the
>> legalities.
>>
>> </snip>
>>
>> I responded to this part of your post, and wanted to clarify that Derby
>> does not build the RC, it is the RM who does it.
> 
> Yes, because it is the RM as an individual that ultimately builds and
> publishes the release as an official Apache release on behalf of the
> ASF.

Absolutely not.

The PMC acts on behalf of the ASF, not the RM.  The RM is just some
hapless schmoe that was out of the room getting the pizza when the
community voted on who herds the cats.


> Once it is published, though, the bits belong to the ASF, not the
> release manager or the Derby project. I'm pretty sure that's what Jean
> meant by 'Apache is ultimately responsible for the release'.
> 
>>  > Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
>>  > GA? --Understanding that the RC will be readily available to anyone to
>>  > download?
>>
>> Anyone can check out the Derby source tree, and build a "derby" with the
>> GA bit set at any time. If anyone does that, it clearly does not mean
>> that Derby has gone GA with JDBC4, and that the community is held
>> legally responsible.
> 
> True. But we're not talking about just anybody. If a member of the
> Derby community does put together a release candidate (that would need
> to report itself as a GA version) and publishes it on a website for
> others to vote on, then the community could be held responsible.

For what?

geir

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Andreas Korneliussen <an...@sun.com> wrote:
> Andrew McIntyre wrote:
> > On 6/23/06, Jean T. Anderson <jt...@bristowhill.com> wrote:
> >
> >> Andreas Korneliussen wrote:
> >> > Jean T. Anderson wrote:
> >> >
> >> > In this case, it is Rick (the release manager) who builds the RC, and
> >> > who makes it available before JDBC4 goes GA. I guess it would be up to
> >> > him to take the legal responsibility.
> >
> >
> > The RM for any Apache release is responsible for evaluating the
> > legalities of the situation surrounding a release and do what is
> > necessary to protect the ASF from legal action. See sections 4 and 8
> > of the ICLA.
> >
>
> Thanks for confirming that it is the RM who is responsible for the
> legalities.
>
> </snip>
>
> I responded to this part of your post, and wanted to clarify that Derby
> does not build the RC, it is the RM who does it.

Yes, because it is the RM as an individual that ultimately builds and
publishes the release as an official Apache release on behalf of the
ASF. Once it is published, though, the bits belong to the ASF, not the
release manager or the Derby project. I'm pretty sure that's what Jean
meant by 'Apache is ultimately responsible for the release'.

>  > Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
>  > GA? --Understanding that the RC will be readily available to anyone to
>  > download?
>
> Anyone can check out the Derby source tree, and build a "derby" with the
> GA bit set at any time. If anyone does that, it clearly does not mean
> that Derby has gone GA with JDBC4, and that the community is held
> legally responsible.

True. But we're not talking about just anybody. If a member of the
Derby community does put together a release candidate (that would need
to report itself as a GA version) and publishes it on a website for
others to vote on, then the community could be held responsible.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Daniel John Debrunner <dj...@apache.org>.
Andreas Korneliussen wrote:


> Anyone can check out the Derby source tree, and build a "derby" with the
> GA bit set at any time. If anyone does that, it clearly does not mean
> that Derby has gone GA with JDBC4, and that the community is held
> legally responsible.

It might depend on if they had to "fork" the code to set the GA bit, or
if the GA bit was already set in the Derby SVN. The model being proposed
is that the GA bit is already set in the Derby SVN, this seems like the
Derby community is saying the codeline is GA.

Dan.


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Geir Magnusson Jr <ge...@pobox.com>.

Andreas Korneliussen wrote:
> Andrew McIntyre wrote:
>> On 6/23/06, Jean T. Anderson <jt...@bristowhill.com> wrote:
>>
>>> Andreas Korneliussen wrote:
>>> > Jean T. Anderson wrote:
>>> >
>>> > In this case, it is Rick (the release manager) who builds the RC, and
>>> > who makes it available before JDBC4 goes GA. I guess it would be up to
>>> > him to take the legal responsibility.
>>
>>
>> The RM for any Apache release is responsible for evaluating the
>> legalities of the situation surrounding a release and do what is
>> necessary to protect the ASF from legal action. See sections 4 and 8
>> of the ICLA.
>>
> 
> Thanks for confirming that it is the RM who is responsible for the
> legalities.

That's not true.

The PMC is responsible for ensuring that any Release of Apache software
is appropriate.

The PMC.

> 
>>>
>  It's up to Rick? Excuse me but Apache is ultimately responsible for this
>>> release and, as a member of this community, I think it's irresponsible
>>> to not resolve any legal issues *before* a candidate is released. The
>>> last thing Derby needs is legal grief.
>>
> 
> Jean, I did not argue that we should not resolve legal issues before the
> candidate is released.
> 
> I responded to this part of your post, and wanted to clarify that Derby
> does not build the RC, it is the RM who does it.

And the ASF releases it via the action of the PMC.

> 
>> Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
>> GA? --Understanding that the RC will be readily available to anyone to
>> download?
> 
> Anyone can check out the Derby source tree, and build a "derby" with the
> GA bit set at any time. 

But they can't call it Derby.



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andreas Korneliussen <an...@sun.com>.
Andrew McIntyre wrote:
> On 6/23/06, Jean T. Anderson <jt...@bristowhill.com> wrote:
> 
>> Andreas Korneliussen wrote:
>> > Jean T. Anderson wrote:
>> >
>> > In this case, it is Rick (the release manager) who builds the RC, and
>> > who makes it available before JDBC4 goes GA. I guess it would be up to
>> > him to take the legal responsibility.
> 
> 
> The RM for any Apache release is responsible for evaluating the
> legalities of the situation surrounding a release and do what is
> necessary to protect the ASF from legal action. See sections 4 and 8
> of the ICLA.
> 

Thanks for confirming that it is the RM who is responsible for the 
legalities.

>>
  It's up to Rick? Excuse me but Apache is ultimately responsible for this
>> release and, as a member of this community, I think it's irresponsible
>> to not resolve any legal issues *before* a candidate is released. The
>> last thing Derby needs is legal grief.
> 

Jean, I did not argue that we should not resolve legal issues before the 
candidate is released.

I responded to this part of your post, and wanted to clarify that Derby 
does not build the RC, it is the RM who does it.

 > Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
 > GA? --Understanding that the RC will be readily available to anyone to
 > download?

Anyone can check out the Derby source tree, and build a "derby" with the 
GA bit set at any time. If anyone does that, it clearly does not mean 
that Derby has gone GA with JDBC4, and that the community is held 
legally responsible.

Andreas

Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andrew McIntyre <mc...@gmail.com>.
On 6/23/06, Jean T. Anderson <jt...@bristowhill.com> wrote:
> Andreas Korneliussen wrote:
> > Jean T. Anderson wrote:
> >
> > In this case, it is Rick (the release manager) who builds the RC, and
> > who makes it available before JDBC4 goes GA. I guess it would be up to
> > him to take the legal responsibility.

The RM for any Apache release is responsible for evaluating the
legalities of the situation surrounding a release and do what is
necessary to protect the ASF from legal action. See sections 4 and 8
of the ICLA.

> It's up to Rick? Excuse me but Apache is ultimately responsible for this
> release and, as a member of this community, I think it's irresponsible
> to not resolve any legal issues *before* a candidate is released. The
> last thing Derby needs is legal grief.

+1

Looks like, barring clarification by Sun what the spec license means
concerning applications that implement interfaces in a non-final JSR,
even in their own namespace, the Derby community shouldn't do anything
regarding a release with JDBC 4 - i would say not even post a release
candidate for a vote - until the JDBC 4 spec is final.

andrew

Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Andreas Korneliussen wrote:
> Jean T. Anderson wrote:
...
>> Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
>> GA? --Understanding that the RC will be readily available to anyone to
>> download?
>>
>> This legal issue needs to get resolved.
>>
> In this case, it is Rick (the release manager) who builds the RC, and
> who makes it available before JDBC4 goes GA. I guess it would be up to
> him to take the legal responsibility.

It's up to Rick? Excuse me but Apache is ultimately responsible for this
release and, as a member of this community, I think it's irresponsible
to not resolve any legal issues *before* a candidate is released. The
last thing Derby needs is legal grief.

 -jean


Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andreas Korneliussen <an...@sun.com>.
Jean T. Anderson wrote:
> Andreas Korneliussen wrote:
> 
>>Jean T. Anderson wrote:
>>
>>
>>>David posted a good summary of the legal catch-22 at [1]. But the
>>>shortest story is:
>>>
>>> + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>>> + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with
>>>Mustang.
>>>
>>>Let's keep this thread confined to the JCP issue Andrew raised that to
>>>roll a release candidate qualifies as "creation".[2] And those release
>>>candidates will be generally available.
>>>
>>>How do we handle this cleanly?
>>>
>>
>>I think, by following the schedule Rick has proposed, we will handle
>>this cleanly.
>>
>>August 25 - Begin vetting 10.2 release candidate
>>End of October - Expected GA of JDBC4 with Mustang
>>End of October - GA of Derby 10.2. Release promoted to Apache mirrors.
>>
>>As you can see from the schedule, Derby 10.2 does not go GA when we
>>start the vetting of the release candidate. Building a RC is not the
>>same as going GA.
>>
>>As for alpha/beta bits: they should not be set on a release candidate.
>>The vetted bits are the final bits if the community votes for the
>>release.  As far as I know, they were not set on the recently rejected
>>10.1.3 RC.
> 
> 
> Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
> GA? --Understanding that the RC will be readily available to anyone to
> download?
> 
> This legal issue needs to get resolved.
>

In this case, it is Rick (the release manager) who builds the RC, and 
who makes it available before JDBC4 goes GA. I guess it would be up to 
him to take the legal responsibility.

Andreas


Re: catch-22: Derby, Mustang, and JCP issue

Posted by "Jean T. Anderson" <jt...@bristowhill.com>.
Andreas Korneliussen wrote:
> Jean T. Anderson wrote:
> 
>> David posted a good summary of the legal catch-22 at [1]. But the
>> shortest story is:
>>
>>  + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>>  + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with
>> Mustang.
>>
>> Let's keep this thread confined to the JCP issue Andrew raised that to
>> roll a release candidate qualifies as "creation".[2] And those release
>> candidates will be generally available.
>>
>> How do we handle this cleanly?
>>
> 
> I think, by following the schedule Rick has proposed, we will handle
> this cleanly.
> 
> August 25 - Begin vetting 10.2 release candidate
> End of October - Expected GA of JDBC4 with Mustang
> End of October - GA of Derby 10.2. Release promoted to Apache mirrors.
> 
> As you can see from the schedule, Derby 10.2 does not go GA when we
> start the vetting of the release candidate. Building a RC is not the
> same as going GA.
> 
> As for alpha/beta bits: they should not be set on a release candidate.
> The vetted bits are the final bits if the community votes for the
> release.  As far as I know, they were not set on the recently rejected
> 10.1.3 RC.

Can Derby legally build an RC with the GA bits set before JDBC 4.0 is
GA? --Understanding that the RC will be readily available to anyone to
download?

This legal issue needs to get resolved.

 -jean



Re: catch-22: Derby, Mustang, and JCP issue

Posted by Andreas Korneliussen <An...@Sun.COM>.
Jean T. Anderson wrote:
> David posted a good summary of the legal catch-22 at [1]. But the
> shortest story is:
> 
>  + Mustang wants to ship a GA Derby 10.2, which supports JDBC 4.0.
>  + Derby can't ship a GA 10.2 until JDBC 4.0 is GA, which is with Mustang.
> 
> Let's keep this thread confined to the JCP issue Andrew raised that to
> roll a release candidate qualifies as "creation".[2] And those release
> candidates will be generally available.
> 
> How do we handle this cleanly?
> 

I think, by following the schedule Rick has proposed, we will handle 
this cleanly.

August 25 - Begin vetting 10.2 release candidate
End of October - Expected GA of JDBC4 with Mustang
End of October - GA of Derby 10.2. Release promoted to Apache mirrors.

As you can see from the schedule, Derby 10.2 does not go GA when we 
start the vetting of the release candidate. Building a RC is not the 
same as going GA.

As for alpha/beta bits: they should not be set on a release candidate. 
The vetted bits are the final bits if the community votes for the 
release.  As far as I know, they were not set on the recently rejected 
10.1.3 RC.

Andreas