[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [sequoyah-dev] Re: [mobile-iwg] Pulsar SDK discovery work for Helios
- From: Christian Kurzke <christian@xxxxxxxxxxxx>
- Date: Wed, 17 Mar 2010 10:48:26 -0700
- Delivered-to: email@example.com
My biggest concern in this is the fact that the code would have to pass the Eclipse Foundation Contribution process, which includes the IP review and code scanning.
I recommend to contact Sharon and see if there is any chance this can get reviewed as a contribution so late in the game.
While i agree that it would be very nice to have the Mylin integration for Helios, i am very concerned that this may introduce a huge risk of not hitting the milestone at all.
As an alternative, I would suggest to participate in the Helios release with a minimum risk (aka plain P2 update) code, and then add the Mylin code at the first Maintainance Release.
One concern i have for this is also the change of the SDK provider meta-data.
Marcel, can you please explain next week at the meeting how your proposed code impacts the P2 Meta Data of the SDK providers?
I hope those changes are minimal (or none at all), which would be another reason for not rushing the Mylin code into the Helios release.
On Wed, Mar 17, 2010 at 10:30 AM, <Ronnie.King@xxxxxxxxx>
Yes, I will be at the meeting on Wed. I agree that we should
plan on getting an agreement on what exactly we believe can be accomplished in
the time that we have remaining for Helios and hopefully get commitments from
interested parties as to what areas we can expect support on. This is very similar
to what happened with Galileo last year, we had to limit what was done due to
time, but we were able to pull in the targets with support from everyone who
participated in the project.
And yes, I fully understand the additional areas that need to be
factored in, documentation, SDK vendor coordination, testing, IP review, etc, which
is exactly why we need to make sure that there is a strong push to do
everything out in the open so we can solicit help where needed.
On the feature risk, I can certainly appreciate the concern
there but I do think that we should make a concerted effort to align with the platform
changes that will be supported in Helios, meaning the discovery UI updates. From
my understanding the integration of these api’s to P2 are complete, which
means we should be able to align the SDK view with the new model. This will of
course also require changes to the metadata for including icons, data, etc, but
this should be capable of completing in the timeframe if we get commitments
from the others on the project.
Will you and/or David be at the Pulsar meeting on Wednesday morning (10 am -
noon)? We should discuss this effort from a logistical viewpoint.
My biggest concern is that we are introducing a significant volume of code
immediately prior to the last milestone before release candidates start rolling
out. In addition to the risk of destabilizing the build and running into test
incompatibilities, there is also the issue of getting the IP review done on
this code. As we're approaching the release the IP reviews tend to take more
In addition to the code contribution and the things you mention below, there is
also the test effort to be defined and executed.
On Mon, Mar 15, 2010 at 2:08 PM, <Ronnie.King@xxxxxxxxx> wrote:
We(Nokia) will be providing development
effort for getting the discovery UI model hooked up to the Pulsar SDK view, so
what I would suggest is that we approach this in a collaborative way and put a
plan together on how we can get this support in by Apr30 (M7). From our team,
David Dubrow is researching/tracking the latest changes in the Equinox project,
and will be looking into hooking up the discovery UI model.
Here’s a quick pass at the items
that we see are needed to be done:
- Confirmation of metadata model changes
and or compatibility : David/Nokia
- Integration of latest changes to the P2
api updates from M6 : Daniel(?)/Motorola
- Design discussion on SDK view changes
needed for discovery UI update, what additional extensions may be needed for
SDK vendors : David, Daniel, ?/Nokia, Mot, any others?
- Other changes that may be needed for the
engine : David/Nokia, other?
- Changes required for metadata generator
tool : ?
- Test cycles with update servers :
It would be good to get your feedback on
these, or if you have additional items to add.
we didn't have time to plan carefully what is going to be the Mylyn based
solution due our Helios integration deadlines, that's why we went for a P2
based solution. As soon as we have time to explore the Mylyn solution, we'll
provide you more details.
Sat, Mar 13, 2010 at 4:11 PM, <Ronnie.King@xxxxxxxxx> wrote:
Could you provide more details on what you are planning to contribute after you
complete the Helios integration builds? If the Mylyn discovery UI is going to
be included, it would be good to know if this will include backward
compatibility with the current metadata model or will it require changes by all
of the SDK vendors?
yes, for now we have SDK Discovery fixed based only on P2 API. We are not
integrating anything from Mylyn due to time constraints. As soon as we have
Helios train ok, we'll restart this work.
On Thu, Mar 11, 2010 at 7:03 PM, ERIC CLONINGER <ericc@xxxxxxxxxxxx>
This work is providing only enough P2 work to discover and handle the packages.
There is not, to my knowledge, anything being done with the Mylyn discovery UI.
I don't know the details of how it's being done, so I'm including Daniel on
On Thu, Mar 11, 2010 at 3:33 PM, <David.Dubrow@xxxxxxxxx>
Hello Pulsar members,
We made a decision at the last Sequoyah call regarding the SDK discovery in
Pulsar. The Sequoyah team was on the call, as was Jon from RIM. We
discussed the options and agreed that this solution is the safest path.
The Pulsar package does not currently build against Helios. We need to get the
Pulsar package building and operational before Friday's M6 milestone, so
Eldorado volunteered the effort. They proposed a plan that will get the
package working using the existing p2 connector with the Helios APIs. As of
today, they've implemented the changes, tested them, and have them in a branch,
ready to integrate into the main trunk. The Mylyn connector strategy is the
right choice for the future, but without anyone committed to the effort, it
will have to wait until Helios SR1.
If anyone has issues with this plan, please respond ASAP.
mobile-iwg mailing list
sequoyah-dev mailing list
mobile-iwg mailing list
Eric Cloninger (ericc@xxxxxxxxxxxx)
Senior Product Manager, MOTODEV Studio for Android
Eclipse Sequoyah Project Lead
sequoyah-dev mailing list