Scope

["Daniel F. Savarese" <df...@savarese.org>]

I'm a loss to understand the problem.  People assert that projects are
out of scope yet also assert how much they like them.  Is it not better
to generate good code that lots of people use than to nitpick about
staying in scope?  If Jakarta is home to a bunch of projects that people
think are crap then that's a problem.  The ability to manage the collection
of projects and whether projects meet a high standard of quality is much
more of an issue than whether projects are in scope.  If it becomes too
difficult to manage the collection of projects, then we split off another
Apache subproject.

I can see that duplication can be a problem.  For example, jakarta-oro
and regexp overlap in the functionality they provide.  The groups have
already discussed this and agreed on an integration plan.  Identify a
problem, discuss possible solutions, come to an agreement, and execute
on it.  Failing to do that is a problem, so if this scope thing is really
a problem, let's take the next step and discuss possible solutions
(refer me to the archives if it's already been done).

Poll a set of server-side Java developers and you'll probably find that
most of them will tell you that regular expressions, logging, and efficient
build systems are all essential to their server side development projects.
(I base that prediction on empirical evidence I've gathered over time
supporting it for the first two items).  Yes, these things are more general
in nature.  Is it time to spawn off a new Java-oriented Apache subproject?
What is the motivating factor?  A subjective view of what's in scope and
out of scope, or the more concrete issues of project management and
product quality?  I'd honestly like to understand the views of the project
members a lot better.  I have companies breathing down my neck about why I
haven't open sourced NetComponents yet.  Aside from recently resolved time
availability issues, I'm reluctant to revisit the possibility of releasing it 
through Jakarta, as original suggested by Jon, because, unlike the projects
mentioned so far, it is unambiguously out of scope, being a set of _client_
libraries for IETF protocols.

In any case, Java developers are starting to look at Jakarta as a
supplier of de facto standard Java class libraries that provide an
alternative to waiting for things they need to show up through the Sun
Community Process.  If this is not a good thing for Jakarta, then
should we start a project which assumes that role?  I don't know
the answer because I think things are working pretty well right now.
Talking about scope is not very useful (to me at least) unless you
address the consequences of being out of scope.  Please help me
understand the consequences.  If there's a real problem here, I'd 
like to solve it sooner rather than later.  If it's time for a split
of some kind, or if the time is approaching, let's analyze the situation
and take care of it proactively.

daniel



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@jakarta.apache.org
For additional commands, e-mail: general-help@jakarta.apache.org

Re: Scope

[Ted Husted <hu...@apache.org>]

I would say that the only the scope that matters is whether the product
will be supported by an Apache meritocracy. Everything else is a
decision we, or the Board, can make by fiat. 

The real problem with Hypersonic was that the team didn't seem
interested in meritocratic development. I don't believe that we should
lobby people to change their development style. If they can't see it for
themselves, then we will just have to wait until they do. (Jakarta !=
cult.) By now, our products should be the proof. 

I don't think we can bootstrap an Apache product by chatting about it.
Someone with vision and the wherewithal to do the work ~and~ build a
community, has to bring something to the table. Something that they
would do here, or anywhere, just to have it done -- with or without our
help. Someone has to have the itch before we can help them scratch.

An Apache Roundtable should not create search committees. Those who wish
to join us should come here and join us. Otherwise, in their hearts,
they will never truly belong. 

-Ted.

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@jakarta.apache.org
For additional commands, e-mail: general-help@jakarta.apache.org

Re: Scope

[Peter Donald <do...@apache.org>]

Hi,

On Sun, 29 Jul 2001 11:44, Daniel F. Savarese wrote:
> I'm a loss to understand the problem.  People assert that projects are
> out of scope yet also assert how much they like them.  Is it not better
> to generate good code that lots of people use than to nitpick about
> staying in scope?  If Jakarta is home to a bunch of projects that people
> think are crap then that's a problem.  The ability to manage the collection
> of projects and whether projects meet a high standard of quality is much
> more of an issue than whether projects are in scope.  If it becomes too
> difficult to manage the collection of projects, then we split off another
> Apache subproject.

Heres my (somewhat unpopular it seems) opinion. If it is high quality, has a 
good (apache-ish) community, is (mostly) written in java and they want to 
become involved in Apache ... then +10000 ;)

I would have no problem with a Jakarta-forge style setup - as long as the 
projects satisfied the above conditions before being "published" under Apache 
name. Basically I see the main question is potential for success and 
community. If those two are available (even if it is not currently the most 
high quality product) then Apache as a whole will benefit.

Ages ago I heard someone mention the phrase "A brave jakarta world", 
obviously a play on "a brave GNU world" and kinda indicative of how I think 
apache/jakarta should go. However instead of a political organization I think 
Apache should endorse project for technical qualitys. So jakarta could be THE 
place where quality java free software/opensource is available.

I think this will happen just as a natural consequence of our evolution. Ages 
ago, last time round I posted a bunch of projects that I think it would 
benefit Apache to have under it's wings. Some of them (Lucene and BSF) are in 
the process of arriving. Others don't "qualify" (dnsjava while a high quality 
java project only has one developer ;/) and others no one has pursued (ie 
openejb or virtually any of the exolab projects). I think growth is 
inevitable and beneficial (as long as we maintain our standards).

BTW +1 to bringing NetComponents here if thats what you decide ... ;)

Cheers,

Pete

*-----------------------------------------------------*
| "Faced with the choice between changing one's mind, |
| and proving that there is no need to do so - almost |
| everyone gets busy on the proof."                   |
|              - John Kenneth Galbraith               |
*-----------------------------------------------------*

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@jakarta.apache.org
For additional commands, e-mail: general-help@jakarta.apache.org