Working towards graduation

[James M Snell <ja...@gmail.com>]

Hey all,

I've been thinking that we need to start thinking about getting Abdera
off the incubator and into its own top level project.  As a TLP, the
Abdera core could provide the foundation for the server, client and
extensions modules to flourish as sub-projects in their own right.

Thoughts?

- James

Re: Working towards graduation

[Brian Moseley <bc...@osafoundation.org>]

On 8/23/07, Garrett Rooney <ro...@electricjellyfish.net> wrote:

> Anyway, just my thoughts, I'd be curious what others think on the subject.

i agree with everything you said and am pretty pleased with what's
been accomplished.

i haven't been active yet due to my recent vacation and because my
work on cosmo has been in a different area for the last several weeks
(focused on webdav rather than atom). i'll be rolling back around to
atom work pretty soon tho, and i'll have to update cosmo to the new
server apis, which means i'll probably have some enhancements to roll
back in.

for what it's worth, there is a lot of server infrastructure in abdera
that could very easily be extracted into a general rest-protocol
framework. i've copied a lot of it in my recent dav work. that same
infrastructure probably heavily overlaps with the jsr 311 work too,
and i'll be looking at whether or not that's a direction we want to
go.

Re: Working towards graduation

[Garrett Rooney <ro...@electricjellyfish.net>]

On 8/23/07, Paul Querna <pq...@apache.org> wrote:
> James M Snell wrote:
> > Hey all,
> >
> > I've been thinking that we need to start thinking about getting Abdera
> > off the incubator and into its own top level project.  As a TLP, the
> > Abdera core could provide the foundation for the server, client and
> > extensions modules to flourish as sub-projects in their own right.
>
> My biggest concerns with going TLP is the lack of committers.
>
> In the last 2 months, there have been three people committing:
> James, Ugo and Stephen.
>
> But 98% of the commits have been by James.
>
> Traditionally the bar for projects has been 3-4 active committers.  And
> while at some level we do have 3, I'm not sure that is a good
> characterization.
>
> The simple question you have to ask is: If any single person was hit by
> a bus today, would the project still be able to thrive.
>
> Personally, I think the answer is no, right now.
>
> However, In every other regard, I think Abdera is ready to graduate.

I certainly agree that the lack of other committers who are as active
as James is a concern, but I'm not sure what the proper solution is,
and I'm also not sure if the "hit by a bus" issue is entirely correct.

We're not in a situation where James is the only person who can dive
into the code and fix problems.  Ugo has committed fixes to the core
code in the past (I recall one off the top of my head, there have
probably been more), and Brian has shown the ability to deal with the
server side code.  Would the loss of James slow the project down and
make dealing with issues in the core of the code difficult?  Of course
it would.  Would it kill the project?  I doubt it, the code has proven
itself too useful for that to happen.

I'm also not sure what the solution would be if we agree that this is
the last remaining problem.  The project is functionally close to
completion.  Aside from refactoring and bug fixing it does what it's
intended to do.  I expect the primary work going forward to be things
like making the server side code do new and interesting things (spring
integration, for example), creating extensions and fixing bugs.  It's
unlikely that we're going to see a situation where a number of people
will need to dive into the guts of the code (well, maybe if we had a
new non-axiom back end, but that's all I can think of).

With that in mind, I think that the project is on the right track.
We've got a number of committers from different companies with fairly
good distribution of expertise throughout the codebase.  There are
other contributors who certainly have the potential to become
committers over time.  I'd say that we're clearly on the right track,
and while waiting a few months would potentially make some people feel
better, I'm not convinced that the project has anything big left to
prove.  We're already doing things right, IMO.

Also, as far as number of active committers, we've got probably 3
active ones, one new one that hasn't had time to get active, and
several inactive ones depending on how you count.  That seems fairly
reasonable to me considering that the group has proven to be wiling to
allow newcomers who exhibit suitable clue to gain commit access
relatively quickly.

Anyway, just my thoughts, I'd be curious what others think on the subject.

-garrett

Re: Working towards graduation

[James M Snell <ja...@gmail.com>]

Yep, I agree 100%...

- James

Paul Querna wrote:
> James M Snell wrote:
>> Hey all,
>>
>> I've been thinking that we need to start thinking about getting Abdera
>> off the incubator and into its own top level project.  As a TLP, the
>> Abdera core could provide the foundation for the server, client and
>> extensions modules to flourish as sub-projects in their own right.
> 
> My biggest concerns with going TLP is the lack of committers.
> 
> In the last 2 months, there have been three people committing:
> James, Ugo and Stephen.
> 
> But 98% of the commits have been by James.
> 
> Traditionally the bar for projects has been 3-4 active committers.  And
> while at some level we do have 3, I'm not sure that is a good
> characterization.
> 
> The simple question you have to ask is: If any single person was hit by
> a bus today, would the project still be able to thrive.
> 
> Personally, I think the answer is no, right now.
> 
> However, In every other regard, I think Abdera is ready to graduate.
> 
> -Paul
> 

Re: Working towards graduation

[Paul Querna <pq...@apache.org>]

James M Snell wrote:
> Hey all,
> 
> I've been thinking that we need to start thinking about getting Abdera
> off the incubator and into its own top level project.  As a TLP, the
> Abdera core could provide the foundation for the server, client and
> extensions modules to flourish as sub-projects in their own right.

My biggest concerns with going TLP is the lack of committers.

In the last 2 months, there have been three people committing:
James, Ugo and Stephen.

But 98% of the commits have been by James.

Traditionally the bar for projects has been 3-4 active committers.  And
while at some level we do have 3, I'm not sure that is a good
characterization.

The simple question you have to ask is: If any single person was hit by
a bus today, would the project still be able to thrive.

Personally, I think the answer is no, right now.

However, In every other regard, I think Abdera is ready to graduate.

-Paul