[sakai2-tcc] Roster tool changes

[Daniel Robinson]

Hi Steve,

That's great news.

We plan on sending an email out to the developer list soon to see if  
there are any others interested.

Best wishes,

Daniel

On 19 Aug 2010, at 23:29, Steve Swinsburg wrote:

> We (ANU) will indeed be dropping it in as a replacement soon.
>
> cheers,
> Steve
>
>
>
> On 20/08/2010, at 12:55 AM, Anthony Whyte wrote:
>
>> Matt Jones plans to propose to the TCC that future maintenance  
>> releases be permitted to include new features (determined on a case- 
>> by-case basis) so that compelling improvements do not need to  
>> remain outside the core build for long periods of time before they  
>> are added.  If such a proposal was approved and the new roster tool  
>> picked up some production street cred between now and a future 2.8  
>> maintenance release, perhaps we would not have to wait until 2.9 to  
>> work it into the release.
>>
>> I do think roster tool should earn a bit of street cred before it's  
>> accepted into core.  Perhaps ANU will run it soon.
>>
>> David has it building in hudson now and we rejig the poms in order  
>> to provide snapshot and release binaries.  Looking at the trimpath  
>> branch poms, rejigging this project as an indie should take very  
>> little time.  I can work on it this afternoon, if you have no  
>> objections.  Then we could add it the trunk experimental profile (- 
>> Pexperimental) where it would replace the existing roster tool.   
>> It's any easy way for people to have a look at it.
>>
>> Anth
>>
>>
>>
>> On Aug 19, 2010, at 10:16 AM, Daniel Robinson wrote:
>>
>>> Hi Chuck,
>>>
>>> Thanks for the response.
>>>
>>> You're correct, the Roster tool is a "read-only" tool and doesn't
>>> interact with the database directly
>>>
>>> Our intention is that the refactored Roster tool will work with a
>>> number of recent releases of Sakai. As far as I know, it doesn't
>>> depend on any recent changes to the services used, but does require
>>> EntityBroker 1.3.13 or higher.
>>>
>>> We have paid interest to accessibility concerns and hope we have not
>>> reduced the accessibility of the tool in any way. We would welcome
>>> feedback on this area from those with the experience.
>>>
>>> It was our original intention to have this new version of Roster in
>>> the 2.8 release, but understand that this may be pushing it a bit.  
>>> We
>>> do have this running in production, but are not using it for course
>>> sites with enrolments as we don't use those at Lancaster. We have
>>> tested the new version using the Sakai sample course data, but  
>>> that is
>>> no substitute for real world tests.
>>>
>>> Does anyone else have a comment about the feasibility of including
>>> this in the 2.8 release?
>>>
>>> Best wishes,
>>>
>>> Daniel
>>>
>>> On 19 Aug 2010, at 14:20, csev wrote:
>>>
>>>> Dan/Adrian,
>>>>
>>>> Looks like good stuff a few questions.
>>>>
>>>> -  It appears as thought this makes no changes to any database
>>>> information so there is no conversion necessary - cool
>>>>
>>>> -  The "data model" is SakaiProxyImpl.java - and this just calls a
>>>> bunch of existing Sakai services to produce a service which gives a
>>>> roster-view of existing Sakai data - cool.   I am curious how  
>>>> tricky
>>>> this code is - would it work with a number of recent Sakai releases
>>>> or does it depend on recent changes to the services it calls?   
>>>> There
>>>> is no wrong answer - more of a curiosity.
>>>>
>>>> - With the switch to trimpath, it would be good to get an
>>>> accessibility review by someone outside the team just to make sure
>>>> that it is as good or better than the old tool - I see you have
>>>> fixed some accessibility things - so I would guess this is already
>>>> on your minds - but a third-party look at it when you are ready
>>>> would make everyone more comfortable, I think.  For me  
>>>> personally, I
>>>> would like to understand better the impact of using trimpath in
>>>> general on accessibility as per the experts.
>>>>
>>>> We are looking at code freeze in the third week of September for  
>>>> 2.8
>>>> - we wondered if Roster might be trying to squeeze into 2.8 before
>>>> the freeze.    At this point with 4-5 weeks remaining, it would  
>>>> seem
>>>> a bit of a stretch to put this into 2.8 on that time scale unless
>>>> someone was running it in production somewhere.    Feel free to
>>>> clarify if I misunderstood.
>>>>
>>>> Again, thanks - this sounds like good stuff we all will benefit  
>>>> from
>>>> and in my short review the code looks nicely self-contained that
>>>> will drop in pretty painlessly.
>>>>
>>>> /Chuck
>>>
>>> _______________________________________________
>>> sakai2-tcc mailing list
>>> sakai2-tcc at collab.sakaiproject.org
>>> http://collab.sakaiproject.org/mailman/listinfo/sakai2-tcc
>>>
>>>
>>
>> _______________________________________________
>> sakai2-tcc mailing list
>> sakai2-tcc at collab.sakaiproject.org
>> http://collab.sakaiproject.org/mailman/listinfo/sakai2-tcc
>