jakarta-taglibs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Glenn Nielsen <gl...@mail.more.net>
Subject Re: Accepting new projects [was: LDAP taglib proposal]
Date Sat, 28 Sep 2002 13:10:07 GMT
Hi Pierre!

My comments are intermixed below.

Pierre Delisle wrote:
> I have not received any comments on my proposal for 
> opening up the way we accept new proposals for
> jakarta-taglibs. 
> I'd eventually like to call for a vote on that topic, 
> but before I do so, let me expand a little bit more 
> on how I'd see it happen for two cases we currently have pending:
> the JNDI/LDAP taglib (Orhan, Mark, Mike), as well
> as the PDF taglib (Roberto).
> Essentially, I think we could run the design/development of
> these new taglibs a bit like the JSR-52 expert group has functioned
> for the design/development of JSTL 1.0.

See my comments nested in step 4.

> We have experts on specific topics that have the interest as well as
> commitment to assume the responsibility to successfully design and
> develop new libraries. What they need is a stimulating environment to
> bring their project to fruition. jakarta-taglibs should be the right
> place to do so if we have the proper process in place.
> So, leveraging the experience of the JSTL expert group, as well
> as our current "new project submission" rules, how about something 
> like this:
> ---
> 1. Motivation
> First of all, the motivation for a new project should be clear.
> There is a multitude of topics that custom taglibs can address.
> I don't think we should accept any project for which the value
> proposition is not clear.
> So, if someone is interested in submitting a new project
> to jakarta-taglibs, it should start with a document that
> describes the motivation for the new taglib.
> This document clearly explains why this taglib is important/useful. 
> It tells what kind of problems it solves.
> It provides as many concrete use cases as possible 
> to support the fact that this will be a good addition to
> a page author's toolbox.
> For example:
>   It would be very important to learn at this
>   stage why people want to do JNDI/LDAP in JSP and not in the
>   business logic. Orhan seems to already have many customers
>   for his taglib; it would be great to hear why they find
>   it so useful in their development environment. Same with
>   Mark and Mike.
>   As for Roberto's PDF taglib, I must admit I know nothing
>   about FO, but would very much like to understand why
>   I should care, and why people would be interested in 
>   his taglib. How do they do it right now without the taglib?
>   What benefit will they gain with the taglib?
> It is the responsibility of the submitter(s) to generate
> interest for the taglib. Show to everyone that it is worth
> for jakarta-taglibs to pay attention to your proposal.
> ---
> 2. Comparative Analysis
> If applicable, a comparative analysis should accompany
> the motivation document (or follow shortly afterwards).
> It should answer the following questions:
>   - Other taglibs out there that do something similar?
>   - What are their strenghts/weaknesses
>   - Why should we have one at jakarta-taglibs?
>   - What will be the key design issues?

Issues 1 & 2 are important but we also don't want to raise
the barrier to high.  The documentation for what to submit
to meet 1 & 2 should be general in nature.  The vote is
the final arbiter.

> ---
> 3. Vote 
> The intent of (1) and (2) is to make sure that a submitter
> does his/her homework properly to motivate the need
> for that new taglib. A vote should not be called
> until that homework has been successfully completed.
> Details to be nailed down, but we probably need
> binding votes where whoever votes +1 will be committed
> in helping this project be successful (the submitter
> will need help to get the project started in the
> sandbox)
> Once the vote is positive, the project enters the sandbox,
> and design discussions can start on the list.

Here is a sample ballot:

[ ] +1 Add taglib to sandbox and I will help
[ ] +0 Add taglib but I can't help
[ ] -1 Don't add taglib, please explain why

For the taglib to be added it woulr require a majority vote
where there is at least 1 +1, and the combination of +1 and +0
is greater than the -1's.

> ---
> 4. Design / Development
> A design document for the taglib is a requirement. 
> I'd suggest using the same format as the one used 
> for the JSTL spec (e.g. see chapter 9 for the SQL actions).
> [I believe that format is clear, provide enough
> information to get someone up and running easily with the taglib, and leaves
> plenty of room for book/article authors to write more about them...]
> The process is iterative. First draft of spec; get comments,
> adjust spec, more comments, some implementation; second
> draft of the spec, more comments, more implementation, etc...

I wouldn't get into much detail here. It might just be enough
to state that the design, including that of each individual tag,
has to be completed (but not fully implemented) before it
can become an official jakarta taglib.  This is open source,
lets allow those working in the sandbox to self organize
their development in ways that work best for them.

I would like to strongly urge new taglibs in the sandbox to
use as much of the standardized build system as they can.
This is documented in addtaglib.html.

> The project is hosted in the sandbox, and eventually moves
> to the official repository once it is ready for prime time.

Once the taglib is no longer experimental it can become official when
the following three requirements are met:

   - It has well defined design goals for the tags
   - it is ready for its first milestone release
   - the jakarta-taglib committers conduct a vote where the majority approve

> [And by doing a great job at designing the best possible taglig
> for that specific topic, the taglib is eventually considered
> within the Java Community Process for standardization :-)]
> ---
> Call for Action
> Here are some things we need to do to help us move forward:
> - All: Is this the right direction?
>   We still need to refine the proposal made above, but how
>   does it sound in general? Is that the direction we should take to
>   be more flexible in handling new project submissions?
>   [the main difference with the current rules is that we would not
>    mandate a project to be fully cooked before being accepted;
>    we would simply require it to have clear motivation for its importance, 
>    clear commitment from the submitters, and clear interest from the community).
>   Orhan, Mark, Mike: Would this process work for all of you interested in a newly
>   designed JNDI/LDAP taglib? If yes, then I'd propose that you guys decide among
>   yourself how to get the motivation/comparative analysis documents ready for 
>   consideration by the list.
> - PDF
>   Roberto: Same question to you. Would this process work for you?
>   If yes, then please get us clear motivation / comparative analysis
>   documents...
> Sorry for another long email... 

Thats ok.  Things like this require it.

Thanks Pierre!


To unsubscribe, e-mail:   <mailto:taglibs-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:taglibs-dev-help@jakarta.apache.org>

View raw message