Fair enough. When I made the comment I was thinking about a user that was interested in helping out or working on Geronimo. I would expect them to look at the Wiki as opposed to mining in JIRAs.

Sachin Patel wrote:
Yes we should have both but I'm not sure if I agree with your second statement. :) I think JIRA should be the first place users should go to request a feature. The wiki will never encapsulate all the work that needs to be done for a project. JIRA tells us this. Thus the wiki should be used to bundle the wish-list and associate them to the themes and priorities for a release. I think trying to keep jiras and the wiki synchronized is invaluable and is unnecessary work. The wiki should be the place where to look for what is planned for that release, and the discussions around and links to jiras/threads for those items. So the first step I think should be to open jira's and from there these can be sorted and prioritized and pushed up into the wiki, I don't think it should be the other way around. There is alot of things that need to be cleaned up in current in JIRA and by using it as the first place for work items this forces us to maintain and constantly scub JIRA to provide an accurate indication of project status.

thats by 2 cents :)


- sachin



On Apr 5, 2006, at 6:12 PM, Matt Hogstrom wrote:

I think you need both. The Wiki is a place to look at what there is to do...JIRA is the way to tell people your doing it.

Dave Colasurdo wrote:

Wasn't suggesting the wiki as the final resting spot, just a temporary holding spot while the user requirements were being gathered, tallied, discussed and prioritized by the community. It works better than email threads passing in the night. No objection to moving the output to JIRAs..
-Dave-
Sachin Patel wrote:

I don't think they should go on the wiki. Why can't we add them as Wish List Jira's with #votes and have the wish list query exported and posted? This way further discussion and progress on each of the items could be tracked as well.

- sachin



On Apr 5, 2006, at 4:07 PM, Dave Colasurdo wrote:

I tried to order them by priority (based on number of times mentioned) within each category/sub-category.. The ones that were mentioned the most are first on the list and so forth.. Items that were mentioned only once are listed in random order..

Anyway, I've updated the list below with the number of requests for each item. I've denoted with *(number)*. Absence of a number indicates one request for an item.

Of course the community needs to digest the input and decide on priorities.

BTW, I have added Global JNDI ENC to the list..

If folks agree with the format, I will post to the wiki..

-Dave-


Matt Hogstrom wrote:

This is great Dave...I think we need to prioritze them as well. Can you translate the priorty from the other e-mails to this?
Matt
Dave Colasurdo wrote:

Excellent feedback from all..

Here is an attempt to consolidate the feedback into one list. (Hope I'm not stepping on anyone's toes.) I've grouped by a few high level categories.. Of course, one could argue that some of the items could fall into multiple categories... Any glaring omissions?


Specifications/Functionality
****************************
-JDK 5 for Geronimo *(11)*
-JEE 5 *(3)*
  Enterprise JavaBeans 3.0 *(5)*
  Java Servlet 2.5
  Java ServerPages 2.1
  Java ServerPages Standard Tag Library
  Java Persistence Architecture (is this part of Java EE?)
  Java Transaction API (JTA)
  J2EE Management
  J2EE Connector Architecture
-JAX-WS support *(4)*
-GBean improvements (doc, lifecycle, dependencies, dep. injection, ..) *(2)*
-Dynamic Queries *(2)*
-javax.persistence
-annotated session beans.
-Support for JSR-168 (Portlet API)
-ServiceMix
-Maven2 support
-service/daemon wrapper
-Remove the requirement of the openejb-jar.xml??
-Better db tools in the admin console
-Configuration management, possibility to make a production version without some current modules (e.g. OpenEJB or ActiveMQ), there are end users who don't have the resources (memory, disk) to run Geronimo fully equipped and they don't need every
  feature of the J2EE stack
- Continued support of Jetty
-First class HttpSession clustering.
-More integration with Apache mod_JK/mod_ajp . It would be nice if a request would continue through a pool until it landed
on a server with that webapp.
-Mass deployment tools that allow the server 'cloning' and rollout mentioned during Geronimos initial days.
-Global JNDI ENC

Tools
*****
-NetBeans support *(3)*
-JDK 5 for launching and running the Geronimo Eclipse plugin *(2)*
-Eclipse plugin improvement (it is really good but think it  could be
better)
-Eclipse mini-roadmap (from Sachin)
- run resources directly from the workspace, so the ear isn't built and re-packaged on every publish - More control over the runtime/server wizards, publish process, and server management - Continue to build up the UI so we have a complete set of "form editors" for the G deployment plans - ability to see changes reflected live in something like the Common Navigator as you modify your plans - full synchronization between the source view and views using the model
  - copy/paste, undo/redo support
-There's also plans in WTP to improve the Server Tools Framework to make it easier to have more control over what defines your runtime.


Usability
*********
-Deployment *(5)*
  Less deployment requirements (simpler plans, more defaults, etc.)
Simplifying deployment (some means to generate geronimo deployment plans?
  Easier  way to  deploy  EAR Files
  More application validation at deployment
Better redeployment to prevent requests from failing if they hit that server during the redeploy.
-More powerful text configuration
-Migration path from Tomcat to Geronimo
-Shortcuts for building web services


Process
*******
-More frequent releases incorporating more user feedback (small releases more often vs. big releases only 4 times a year) *(7)* -Geronimo Certified Partner Program (including Jetspeed-2 as a member). *(2)*


Documentation *(10)*
*************
-Tutorials *(4)*
  How to use Geronimo with: Apache Axis, WSS4J, ActiveMQ
-Cookbooks
-Better documentation on deployment descriptors
-More detailed documentation about the architecture and Gbeans * (2)*
-A browsable table describing where to find all plans, etc. for each deployed component or service. -More documentaion on deploying EAR's, WAR's, EJB's, RAR's, classloading and dependencies with that apps)

Examples *(4)*
********
-Examples for everything
-More documentation/examples for me should be more explicative models of the basic openejb-jar.xml and ejb-jar.xml,
   explaining which tag points where or what
  Session beans, entity beans(BMP and CMP)
  Message-Driven-Bean.



-Dave-











Reply via email to