Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [science-iwg] Science release, target platform and Orbit version

Christopher,

You know, it might be interesting to see if Mike and the IP team need a guinea pig for a type A release...

Jay


On Aug 18, 2016 11:25 AM, "Christopher Brooks" <cxh@xxxxxxxxxxxxxxxxx> wrote:
I agree that there is a reasonable chance that the Triquetrum IP CQs will not be completed by October.  The list of Triquetrum CQs is at https://github.com/eclipse/triquetrum/issues/70

In that list, we have 3424K Jar files total that have not yet completed review.  These were submitted at the end of November, 2015 and updated in June.

Of those jars, 1459K have completed preliminary review, were returned to devel and then returned to IP.

I realize that we can't release without the IP CQs being completed, but I also realize that I have a bunch of work ahead of me to get to a having a release.

If we can't release because the IP CQs are not completed, then this suggests having a conversation about alternatives.  See https://mmilinkov.wordpress.com/2016/06/29/overhauling-ip-management-at-the-eclipse-foundation/.  It might make sense for Triquetrum to be released without extensive review of its dependencies.

In other news, mixing Orbits seems reasonable to me.

_Christopher

On Thu, Aug 18, 2016 at 4:37 AM, UOMo <uomo@xxxxxxxxxxx> wrote:
Jay/all,

Glad to see others are having "fun" with Orbit, too;-)

Judging from the experience of not so small Deutsche Telekom with its CQs filed by SmartHome/Kai:
From all I know these are both still unresolved. After Kai filed them on Feb 5, 2016.
He asked what could make the IP process and department work faster, and I recall the main answer was "if Telekom was a Platinum Eclipse member" or similar. Which it isn't.

I can't say, if those who filed the Ptolomy CQs are or have other ways to make the IP team faster, otherwise I would not bet they can be approved by October 1st 2016 or even the end of the year based on our experience with others.

Those CQs by SmartHome will if approved allow DAWN to use JSR 363, too. Instead of the current pre-predecessor JSR 275.

Regards,


On Thu, Aug 18, 2016 at 1:15 PM, <science-iwg-request@eclipse.org> wrote:
Send science-iwg mailing list submissions to
        science-iwg@xxxxxxxxxxx

To subscribe or unsubscribe via the World Wide Web, visit
        https://dev.eclipse.org/mailman/listinfo/science-iwg
or, via email, send a message with subject or body 'help' to
        science-iwg-request@xxxxxxxxxxg

You can reach the person managing the list at
        science-iwg-owner@xxxxxxxxxxx

When replying, please edit your Subject line so it is more specific
than "Re: Contents of science-iwg digest..."


Today's Topics:

   1. Science release, target platform and Orbit version (Erwin de Ley)
   2. Re: Science release, target platform and Orbit version
      (Christopher Brooks)
   3. Re: Science release, target platform and Orbit version
      (Jay Jay Billings)
   4. Re: Science release, target platform and Orbit version
      (Erwin de Ley)


----------------------------------------------------------------------

Message: 1
Date: Wed, 17 Aug 2016 22:50:15 +0200
From: Erwin de Ley <erwin.de.ley@xxxxxxxxxx>
To: Science Industry Working Group <science-iwg@xxxxxxxxxxx>
Subject: [science-iwg] Science release, target platform and Orbit
        version
Message-ID: <4a48c1cd-b7d0-6503-48e3-db474f00a446@xxxxxxxxxx>
Content-Type: text/plain; charset="utf-8"; Format="flowed"

Dear all,

I've been spending some time (learning how to) adding the Apache XML RPC
and WS Common Utils to the Orbit project, and finally seem to be ready
with that. The gerrit review should be ok now and the merge will be done
in the coming days

This is a prerequisite for a clean Triquetrum target platform that also
must support the Java->Python integration via an XML-RPC bridge
(extracted from DAWN and contributed by Jonah).

Now the main issue :
- the analysis RPC stuff to provide a bridge to Python needs those (4)
bundles as dependencies
- it should be part of our Triquetrum release, within the Science release
- Orbit will only offer those bundles in an Oxygen M2 milestone build,
sometime in september (or I can request to promote a nightly build sooner)
- we'll be on the Mars.x (or Neon if multiplatform builds can be made to
work) as principal target platform


So this could be a problem both in :
- /timing/, as we would like to finish our work sometime mid September
to allow sufficient time for all reviews etc
- and in "/version/"? I.e. do we allow to depend on Orbit for Oxygen for
our builds/releases?

What would be the best approach?

regards

erwin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://dev.eclipse.org/mailman/private/science-iwg/attachments/20160817/3e7327af/attachment.html>

------------------------------

Message: 2
Date: Wed, 17 Aug 2016 13:55:47 -0700
From: Christopher Brooks <cxh@xxxxxxxxxxxxxxxxx>
To: Science Industry Working Group <science-iwg@xxxxxxxxxxx>
Subject: Re: [science-iwg] Science release, target platform and Orbit
        version
Message-ID:
        <CAC5j0dyjh7-d8Th26hr1uUTxXGPct+EYCwORqdj9gEAJQ4nD=A@xxxxxxxail.com>
Content-Type: text/plain; charset="utf-8"


I think the Python work is really important.  I have a strong distaste for
Python's whitespace policy, but I realize that Python has excellent mind
share in the scientific community.  Supporting the Python interface would
increase the usefulness and visibility of our work.

I'd like to see Triquetrum include this work, and if that means that
Triquetrum needs to build on the Oxygen M2 Orbit, then that's ok with me.

I'm not sure about the timing, but I would not mind keeping our target
dates as is and slipping them as necessary as we get closer.

_Christopher

On Wed, Aug 17, 2016 at 1:50 PM, Erwin de Ley <erwin.de.ley@xxxxxxxxxx>
wrote:

> Dear all,
>
> I've been spending some time (learning how to) adding the Apache XML RPC
> and WS Common Utils to the Orbit project, and finally seem to be ready with
> that. The gerrit review should be ok now and the merge will be done in the
> coming days
>
> This is a prerequisite for a clean Triquetrum target platform that also
> must support the Java->Python integration via an XML-RPC bridge (extracted
> from DAWN and contributed by Jonah).
>
> Now the main issue :
> - the analysis RPC stuff to provide a bridge to Python needs those (4)
> bundles as dependencies
> - it should be part of our Triquetrum release, within the Science release
> - Orbit will only offer those bundles in an Oxygen M2 milestone build,
> sometime in september (or I can request to promote a nightly build sooner)
> - we'll be on the Mars.x (or Neon if multiplatform builds can be made to
> work)  as principal target platform
>
>
> So this could be a problem both in :
> - *timing*, as we would like to finish our work sometime mid September to
> allow sufficient time for all reviews etc
> - and in "*version*"? I.e. do we allow to depend on Orbit for Oxygen for
> our builds/releases?
>
> What would be the best approach?
>
> regards
>
> erwin
>
> _______________________________________________
> science-iwg mailing list
> science-iwg@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/science-iwg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://dev.eclipse.org/mailman/private/science-iwg/attachments/20160817/4fa76bea/attachment.html>

------------------------------

Message: 3
Date: Thu, 18 Aug 2016 07:05:47 -0400
From: Jay Jay Billings <jayjaybillings@xxxxxxxxx>
To: Science Industry Working Group <science-iwg@xxxxxxxxxxx>,
        Christopher Brooks <cxh@xxxxxxxxxxxxxxxxx>
Subject: Re: [science-iwg] Science release, target platform and Orbit
        version
Message-ID:
        <CAE3ybv6S0Tco6jeFVXihCDe0VYoCWh6bQOEs+1UPKDHY5obB7w@xxxxxxxail.com>
Content-Type: text/plain; charset="utf-8"


A couple of thoughts...

You don't have to build solely on Oxygen just because you are pulling from
a new drop of Orbit. We pull from multiple releases of Orbit for ICE to get
the bundles that we need. So I would say release primarily on Neon, but
with your extra few bundles from the new repo too.

As far as timing goes, the reviews take only a couple of weeks, so your
freeze date couple be as late as October 1st or so. The bigger problem at
the moment is that the Ptolemy dependencies still haven't been approved
through the CQ process, right? We can't even do the reviews if we don't
make it through that process.

Jay

On Aug 17, 2016 4:55 PM, "Christopher Brooks" <cxh@xxxxxxxxxxxxxxxxx> wrote:

> I think the Python work is really important.  I have a strong distaste for
> Python's whitespace policy, but I realize that Python has excellent mind
> share in the scientific community.  Supporting the Python interface would
> increase the usefulness and visibility of our work.
>
> I'd like to see Triquetrum include this work, and if that means that
> Triquetrum needs to build on the Oxygen M2 Orbit, then that's ok with me.
>
> I'm not sure about the timing, but I would not mind keeping our target
> dates as is and slipping them as necessary as we get closer.
>
> _Christopher
>
> On Wed, Aug 17, 2016 at 1:50 PM, Erwin de Ley <erwin.de.ley@xxxxxxxxxx>
> wrote:
>
>> Dear all,
>>
>> I've been spending some time (learning how to) adding the Apache XML RPC
>> and WS Common Utils to the Orbit project, and finally seem to be ready with
>> that. The gerrit review should be ok now and the merge will be done in the
>> coming days
>>
>> This is a prerequisite for a clean Triquetrum target platform that also
>> must support the Java->Python integration via an XML-RPC bridge (extracted
>> from DAWN and contributed by Jonah).
>>
>> Now the main issue :
>> - the analysis RPC stuff to provide a bridge to Python needs those (4)
>> bundles as dependencies
>> - it should be part of our Triquetrum release, within the Science release
>> - Orbit will only offer those bundles in an Oxygen M2 milestone build,
>> sometime in september (or I can request to promote a nightly build sooner)
>> - we'll be on the Mars.x (or Neon if multiplatform builds can be made to
>> work)  as principal target platform
>>
>>
>> So this could be a problem both in :
>> - *timing*, as we would like to finish our work sometime mid September
>> to allow sufficient time for all reviews etc
>> - and in "*version*"? I.e. do we allow to depend on Orbit for Oxygen for
>> our builds/releases?
>>
>> What would be the best approach?
>>
>> regards
>>
>> erwin
>>
>> _______________________________________________
>> science-iwg mailing list
>> science-iwg@xxxxxxxxxxx
>> To change your delivery options, retrieve your password, or unsubscribe
>> from this list, visit
>> https://dev.eclipse.org/mailman/listinfo/science-iwg
>>
>
>
> _______________________________________________
> science-iwg mailing list
> science-iwg@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe
> from this list, visit
> https://dev.eclipse.org/mailman/listinfo/science-iwg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://dev.eclipse.org/mailman/private/science-iwg/attachments/20160818/e50ebc3e/attachment.html>

------------------------------

Message: 4
Date: Thu, 18 Aug 2016 13:15:05 +0200
From: Erwin de Ley <erwin.de.ley@xxxxxxxxxx>
To: Science Industry Working Group <science-iwg@xxxxxxxxxxx>
Subject: Re: [science-iwg] Science release, target platform and Orbit
        version
Message-ID: <8fcdba84-f94b-32c8-d09b-4c46a8613f25@xxxxxxxxxx>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"


Hi Jay,

Ok. For me the main question was about platform "consistency". So we'll
mix Orbits as needed.

About the Neon vs Mars choice : are you able to get the multi-platform
builds working for ICE with Tycho on Neon? Christopher's work to migrate
Triquetrum to Neon hit an issue with multi-platform builds which for the
moment has forced us to switch back to Mars (cfr
https://bugs.eclipse.org/bugs/show_bug.cgi?id=491951, and it appears
that Gunnar's p2 repo can not be loaded successfully from our HIPP)

For the Ptolemy CQs, we should not be far away from the approvals. I
think the waiting time until now was more about the reviews not being
started yet, and now the good news is that they are ongoing, with only
very limited remarks.

regards

erwin

Op 8/18/2016 om 1:05 PM schreef Jay Jay Billings:
>
> A couple of thoughts...
>
> You don't have to build solely on Oxygen just because you are pulling
> from a new drop of Orbit. We pull from multiple releases of Orbit for
> ICE to get the bundles that we need. So I would say release primarily
> on Neon, but with your extra few bundles from the new repo too.
>
> As far as timing goes, the reviews take only a couple of weeks, so
> your freeze date couple be as late as October 1st or so. The bigger
> problem at the moment is that the Ptolemy dependencies still haven't
> been approved through the CQ process, right? We can't even do the
> reviews if we don't make it through that process.
>
> Jay
>
>
> On Aug 17, 2016 4:55 PM, "Christopher Brooks" <cxh@xxxxxxxxxxxxxxxxx
> <mailto:cxh@xxxxxxxxxxxxxxxxx>> wrote:
>
>     I think the Python work is really important.  I have a strong
>     distaste for Python's whitespace policy, but I realize that Python
>     has excellent mind share in the scientific community.  Supporting
>     the Python interface would increase the usefulness and visibility
>     of our work.
>
>     I'd like to see Triquetrum include this work, and if that means
>     that Triquetrum needs to build on the Oxygen M2 Orbit, then that's
>     ok with me.
>
>     I'm not sure about the timing, but I would not mind keeping our
>     target dates as is and slipping them as necessary as we get closer.
>
>     _Christopher
>
>     On Wed, Aug 17, 2016 at 1:50 PM, Erwin de Ley
>     <erwin.de.ley@xxxxxxxxxx <mailto:erwin.de.ley@isencia.be>> wrote:
>
>         Dear all,
>
>         I've been spending some time (learning how to) adding the
>         Apache XML RPC and WS Common Utils to the Orbit project, and
>         finally seem to be ready with that. The gerrit review should
>         be ok now and the merge will be done in the coming days
>
>         This is a prerequisite for a clean Triquetrum target platform
>         that also must support the Java->Python integration via an
>         XML-RPC bridge (extracted from DAWN and contributed by Jonah).
>
>         Now the main issue :
>         - the analysis RPC stuff to provide a bridge to Python needs
>         those (4) bundles as dependencies
>         - it should be part of our Triquetrum release, within the
>         Science release
>         - Orbit will only offer those bundles in an Oxygen M2
>         milestone build, sometime in september (or I can request to
>         promote a nightly build sooner)
>         - we'll be on the Mars.x (or Neon if multiplatform builds can
>         be made to work)  as principal target platform
>
>
>         So this could be a problem both in :
>         - /timing/, as we would like to finish our work sometime mid
>         September to allow sufficient time for all reviews etc
>         - and in "/version/"? I.e. do we allow to depend on Orbit for
>         Oxygen for our builds/releases?
>
>         What would be the best approach?
>
>         regards
>
>         erwin
>
>         _______________________________________________
>         science-iwg mailing list
>         science-iwg@xxxxxxxxxxx <mailto:science-iwg@xxxxxxxxxxg>
>         To change your delivery options, retrieve your password, or
>         unsubscribe from this list, visit
>         https://dev.eclipse.org/mailman/listinfo/science-iwg
>         <https://dev.eclipse.org/mailman/listinfo/science-iwg>
>
>
>
>     _______________________________________________
>     science-iwg mailing list
>     science-iwg@xxxxxxxxxxx <mailto:science-iwg@xxxxxxxxxxg>
>     To change your delivery options, retrieve your password, or
>     unsubscribe from this list, visit
>     https://dev.eclipse.org/mailman/listinfo/science-iwg
>     <https://dev.eclipse.org/mailman/listinfo/science-iwg>
>
>
>
> _______________________________________________
> science-iwg mailing list
> science-iwg@xxxxxxxxxxx
> To change your delivery options, retrieve your password, or unsubscribe from this list, visit
> https://dev.eclipse.org/mailman/listinfo/science-iwg

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://dev.eclipse.org/mailman/private/science-iwg/attachments/20160818/10048ffc/attachment.html>

------------------------------

_______________________________________________
science-iwg mailing list
science-iwg@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/science-iwg

End of science-iwg Digest, Vol 43, Issue 16
*******************************************


_______________________________________________
science-iwg mailing list
science-iwg@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/science-iwg


_______________________________________________
science-iwg mailing list
science-iwg@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/science-iwg

Back to the top