Grouper Developers Forum

["Cramton, James" <>]

Setting functional priorities is important, but I would argue that we
need to take the opportunity now to spend an appropriate amount of time
thinking about the entire platform that encompasses Grouper, Signet,
LDAPpc, Subject API, Grouper Shell, and possibly other projects in this
space. If we focus on the individual functional requirements, we may
address some short term needs, but at the risk of missing out on a key
opportunity to develop a more robust product in the long run. 

I suggest that we start actively using Jira to manage these feature
requests, their popularity, and their dependencies. Jira does a great
job of documenting these issues in a roadmap, and it will be a useful
tool to track these big-picture issues. This smells a lot like product
management to me.

Depending on the design we choose, the items on the list today are not
necessarily independent of each other, particularly if we are thinking
in product management terms. In our discussions in San Diego, we did
some quick back-of-envelope system design that established some initial
dependencies that look something like this:

A1: Hooks is a prerequisite for
        A2: Notification of Changes
        A5: History and Audit
        (Not itemized) Transactions/Rollback support

A3: Web Service Interface is a prerequisite for 
        A4: Web service interface facades
        A7: I2MI integration
        A8: XML: import/export-Signet

A6: Rule-based actions is probably dependent on some number of the
others.

I'd be particularly wary of defining a web service interface before more
low-level changes go into a hooks interface and its dependents. 


And from Brown's perspective, we'd prioritize more along the lines of
A1, A2, A5, A3. Provisioning in real-time is a higher priority for us
than UI improvements.

James Cramton
Lead Programmer/Analyst
Brown University

401 863-7324



-----Original Message-----
From: Shilen Patel 
[mailto:]
 
Sent: Monday, October 15, 2007 1:00 PM
To: Michael R. Gettes
Cc: Grouper Dev; Signet; klara jelinkova
Subject: Re: [grouper-dev] updated grouper/signet roadmap

Hi Tom,

Based on Duke's requirements of what needs to be implemented in order
for Grouper to be more widely used at the university, we agree that A3
should be at the top of the list.  Our priorities are the following:

A3, A1, A2, A5, A6.

Thanks,

-- Shilen


Michael R. Gettes wrote:
> As you might expect... I don't agree with the order.  I don't recall 
> there being discussion at the meeting about specific ordering.  A3 
> should be first on the list, IMHO.  I think there should be discussion

> about timelines and not just ordering.
>
> /mrg
>
> On Oct 9, 2007, at 18:57, Tom Barton wrote:
>
>> I've updated the grouper roadmap page in the grouper project wiki to 
>> reflect the discussion of the *combined signet-grouper* roadmap that 
>> occurred in the combined signet-grouper working groups session at the
>> Internet2 meeting yesterday (Monday, October 8, 2007).
>>
>> https://wiki.internet2.edu/confluence/display/GrouperWG/Priorities+fo
>> r+Functional+Enhancements
>>
>>
>> This document is editable for grouper working group members (ie, as 
>> defined in the wiki). If there's something substantial not there that

>> should be, please add it, or write to the list and either Jessica or 
>> I will add the substance of your message to the roadmap.
>>
>> Tom
>>
>> PS. I'd like to add the signet working group to the ACL for that 
>> page, but apparently this wiki is not able to have more than one 
>> group in the ACL.
>>
>