[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
[higgins-dev] Notes from August 14th Higgins developers call
|
Notes from the
Higgins Developers call on Thursday, August 14
Attendees
=========
* Greg Byrd - NCSU/IBM
* Tom Doman -
Novell
* Andy Hodgkinson - Novell
* Ranjan Kumar - IBM
* Mike McIntosh - IBM
* Dale
Olds -
Novell
* Ernst Plassmann - IBM
* Mary
Ruddy
- Meristic/SocialPhysics
* Jim Sermersheim - Novell
* George Stanchev -
Serena
* Daniel Sanders
* Paul
Trevithick -
Parity/SocialPhysics
* Brian
Walker -
Parity
* Tom Caroll - Parity
* Mohamad Raji
Gani - Oracle
Time: noon
EDT (1700
London; 1800
Vienna,
Paris, Berlin)
Dial-in:
1-866-362-7064 /
892048#
Agenda
1. [Brian] Now working on 1.1M4.
2. [Mary] Status
update on Eclipse SVN situation.
- [Mike] Will we need to
switch our repository locations?
Once we do, it may not be as easy to commit. Want to know the repercussions. We may not be able to easily re-commit
the changes made to a different repository.
- [Mary] I will clarify
that.
- [Mary] Any other issues?
- [Paul] Understand that we
can’t write to the repository.
- [Mary] Eclipse has frozen
it. No news as of an hour a go.
- [Andy] The last commit I
show is august 1st. I had been committing almost the entire week
until they locked down the repository.
I lost about a week’s work of changes. I still have them on my hard
drive. It looks like they had
restored a back up that was about a week old.
- [Drummond] I
joined.
3. [Brian]
Nightly Auto-Test
- Next step is
getting Mike's STS to build using headless
Buckminster.
- [Mary] Mike any update on
the build?
- [Mike] In order to
support other stuff. I need to
spend some focused time on other things.
Once I get to a stopping point, I will get back to work on
this.
- [Brian] The next item is
internationalization.
- [Mary] David is on the
call too.
4. [Brian & David]
Internationalization
- See http://wiki.eclipse.org/Internationalization_Support
- Ipzilla item 2515 has
been approved. Waiting for SVN to open up to create Jar
file.
- Need to identify next
components to work on.
- [Brian] This centers on
what we want to tackle next. Have
engaged in an offline dialogue with David and Ranjan to sort this out and
propose that to the Higgins folks.
- [David] We don’t have a
priority after the IdAS and JNDI context provider. It would be really up to the team
about how they would approach it in terms of the priorities we have for this
year.
- [Brian] Does anyone else
have any priority? For us to repeat what we did for IdAS and the JNDI CP? Can send out message to the list for
people.
- [Brian] I will take an
action to send a note to the dev list to see who else would be interested in
going next and keep the ball rolling.
- [Paul] Do you know if Mohamad is on the
call?
- [Mary]
Yes.
5. [Mohamad] New
Oracle-provided Google Contacts Context Provider
- Overview of functionality
- Does it implement each
contact as a separate Entity?
- (Mohamad has
confirmed.)
- [Paul] Can you walk us
though the functionality and the model and any issues you
have?
- [Mohamad] A brief
overview of what this really is and how if fits with the IdAS
API.
- [Mohamad]It is basically
a syndication format of the ?? protocol. It is just to create contacts, and
delete contacts. It supports all
the operations, authentication, and sending back the xml format. Simply create context using http
post. This is internally done by
the Google java client library.
- [Mohamad]It just allows
you to access and edit the context.
You can also query the context. [For example] Give me all the updates after a
certain date.
- [Mohamad]Once have legal
approval, plan to send it over to validate our code. Each context is mapped to a unique ID
a URI for that particular
contact. We use that
URI as a unique ID. The thing not completed until now is auditing or batch updates. Google supports batch, but we don’t really do it
now. On the legal approval part,
we are almost through. Are waiting for an official response from Oracle legal,
and then we will be sending the code.
- [Paul] Thank you. This is great. Does anyone else on the call have a
question?
- [?] It sounds like each
contact is implemented as a separate entity id.
- [Paul] Is this the right
way?
- [Drummond] This is
definitely the right way. It
should definitely be entity.
- [Mohamad] We used the
JNDI base and mapped it from there.
- [Paul] Ah I see. I’m not facile with the terms of
service on the Google API. Have we looked at that?
Are there any restrictions?
- [David] I wanted to. I
thought you need to have a develop key to use the
API? Not really just need a Google
account.
- [Paul] So there is no
restrictions.
- [??] There is no
governor?
- [Mohamad] No, currently
restricted to 100 entries. I think in one shot.
- [Paul] In one shot?
- [Mohamad] I can get 100
entries in one shot.
- [Paul] Do you have any
performance data?
- [m] No I can do that next
week and send that to you
- [Paul] Just interested.
- [Mohamad] I can do that
next week.
- [Paul] Next topic.
6. [Mary] Next
Higgins F2F
- We had suggested
having a F2F just after DIDW, but there wasn’t a lot of energy around
that. So another idea might be having a F2F before or after IIW
(November, Mountain
View) or at some other
time
- ttp://www.doodle.ch/vbyw4anyc3ezurv7
- Poll on having it after IIW (November 10-12
in Mountain
View)
- http://www.doodle.ch/e29wa6a77ctg8ft5
- Poll on ideal length
- http://www.doodle.ch/pesa7yhbexbtm6n7
- Poll on ideal location
- [Mary] I created three new doodles. One for when people could meet after
IIW, and two for having a meeting at some other time, not connected with an
event. One is for the location and the other is for the length. After this feedback, we may need to
create a more fine grained Doodle.
- [Drummond]
I didn’t know quite how to
reflect. It seemed outside the scope of the doodle. One comment is to have the F2F meeting
with the OSIS meeting in
London.
- [Mary] What about IIW?
- [Drummond] [There is a TC meeting that
week.]
- [Drummond] Not sure they are also in the planning
stage.
- [Drummond] Know that Tony is head of that TC. There may be others in the same
boat.
- [Mary] Sounds like IIW isn’t a good time. So may net out to having [the F2F at]
a separate time and venue.
- [Paul] That is a good plan.
- [Dale] I think IIW is being overloaded in general. This one
especially.
- [Mary] So I think we will not try for
IIW.
- [Paul] We
could try some standalone times. And ask our generous Novel and
IBM friends to
provide a location.
- [Paul] Is Jim is on the
call?
- [Jim] I’m here.
- [Paul]
Different topic. I was wondering
whether or not or when we night want to discuss the overall model
API and
replacing that with direct modeling of classes as
entities.
- [Jim] That is a good idea.
Shall we just say we are going to do it next week so that I propose
something readable? I
really need to draw a picture. It will help people see and flesh out the
detail. Is a phone call the right
forum for that?
- [Paul] You could send to the list this as a potential Thursday topic
and see if there are others who are interested. I know I am and Brian. Some of our Ukrainian friends might be
very interested. So maybe you
could send a link to the wiki and see what other people think. It would also force me into a
deadline. Doesn’t need to be next
Thursday. Just picking a
time. So decide, think about
it.
- [Paul] Does anyone have any other topics?
- [Dale] I do.
If I put on my
OSIS hat, one of
the things for DIDW was the suggestion that every project have a 1 pager that
could be handed out at the event.
So as a Higgins participant, how do we get that done and get testing
done between now and then? And furthermore there is a possibility of an actual
presentation during the OSIS event. We
are trying to not have project presentations per se. I think Mary and Tom are also full. We
need to plan for that and at least get the one pager done.
- [Paul] I’m willing to contribute to the one pager. When does it need to be
done?
- [Dale] Before DIDW.
- [Paul] In terms of the testing. DigitalME and Higgins use cases are so
similar.
- [Dale] If tested on one selector, is effectively tested on
both.
- [Paul] Parity
is developing a selector based on the AIR based
selector, so we have a very similar situation to Novel. There is a Parity selector that is
largely the same as the Higgins one.
Brian if we are putting the resources
in...
- [Tom] We are
planning on testing the Parity AIR
selector.
- [Paul] That is ok. That is what I thought. I think the right way is to
put in a new column. We call this
the Azigo selector.
- [Tom] Should show up over this weekend.
- [Paul] What I’m hearing is that there aren’t enough resources to go
ahead to test both. Just as with Dale, they are so
similar.
- [Paul] The last thing Dale said is a
presentation.
- [Dale] At this point, lets drop this off, as we already have some
involvement from Mary and others.
I don’t think we have room for a Higgins presentation. So let’s drop
that one. If there is anyone
particularly interested talking about interoperability from a high
perspective, we can add a few slides.
- [Paul] That is great. Just
trying to understand.
- [Paul] Anything else?
- Silence.
- OK
- [Paul] I’m not on the call next week. So could I have a
volunteer?
- [Brian] Yes.
- [Paul] Thanks
everyone. See you in two weeks
(or for some one week.)
- End.