Depot

[Adam Jack <aj...@mric.coop>]

Folks

Sorry I've been offline a while, I've had other priorities to take care of.

I will be sad to let Depot go, but I suspect it is time. I am starting to
believe that without a lot more community drive (or just outright individual
stubbornness to continue) that something like a version/repository idea is
too much to pull off. Further, I suspect that Javasoft ought be handling it,
not other groups. Also, I do suspect we tried to do too much (and yes, I
probably was over enthusiastic and helped over engineer some parts) so
couldn't keep up with it. That, and (for me) the transition into incubation
caused too much environment awkwardness, and I lost momentum.

Whatever the case, long and the short is that we didn't seem to build on a
simple use case, we didn't use it first and code it second. Basically it
stagnated underneath us. I don't think I'll ever quite understand why, but
despite technical merits (IMHO) something just didn't spark others. Maybe
marketing, maybe star alignment, definitely community.

That all said, I really really struggle with this. So much effort, so much
good stuff, so close. Gak! I hate letting go.

regards,

Adam

Test

[Nicola Ken Barozzi <ni...@apache.org>]

Test if list closed.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------

Re: Depot

[Niclas Hedhman <ni...@hedhman.org>]

On Sunday 07 November 2004 17:48, Niclas Hedhman wrote:
> are intersecting with Depot becomes very generic. In fact, we can now
> artifact enable any application with changing the app(!). 

Should of course be "without changing the app"...

Cheers
Niclas
-- 
   +------//-------------------+
  / http://www.bali.ac        /
 / http://niclas.hedhman.org / 
+------//-------------------+

Re: Depot

[Niclas Hedhman <ni...@hedhman.org>]

On Sunday 31 October 2004 21:26, Adam Jack wrote:

> Whatever the case, long and the short is that we didn't seem to build on a
> simple use case, we didn't use it first and code it second. Basically it
> stagnated underneath us. I don't think I'll ever quite understand why, but
> despite technical merits (IMHO) something just didn't spark others. Maybe
> marketing, maybe star alignment, definitely community.

I think you are close to the target when you say; "Didn't use it."

Since we have constantly refined the equivalent that started as Avalon 
Repository (and now is at www.dpml.net as Transit), some of the parts that 
are intersecting with Depot becomes very generic. In fact, we can now 
artifact enable any application with changing the app(!). I.e. add a Jar to 
the classpath and set a single system property and you use Maven remote 
artifacts and local caching. No API, no code changes, 'just use it'...

That is IMHO, real value. OTOH, this effort will stabilze fairly soon, and 
that is something that ASF doesn't like. Perfect apps that has no 
evolutionary path are doomed in this setting, so perhaps even Depot didn't 
have any long term prospect within ASF, due to the same principle.


Cheers
Niclas
-- 
   +------//-------------------+
  / http://www.bali.ac        /
 / http://niclas.hedhman.org / 
+------//-------------------+