Raphaël Luta wrote:
Excellent post! It is nice to see somebody take the time to review the
actual proposal.
Overall, there is clearly strong interest in AJAX at the ASF, whether it
be based on Zimbra or Dojo or whatever. Furthermore, the proposal needs
to be revised, particularly to incorporate the people who have expressed
an interest in participating and creating ties to other projects.
-------------------------------------------------------------------------------
Criteria
========
* Meritocracy:
I don't believe it looking at the committer list.
Who's going to argue with his VP of engineering ? To create a real
meritocracy, you can't have an established hierarchy in the committership.
Valid concern - needs to be worked.
* Community:
none
A bit overstated. There is a community, but it has yet to be incubated
and certified as following the Apache way.
* Core developers:
no existing Apache committer or Apache member
That's the initial proposal. The proposal was immediately was met with
several volunteers.
* Alignment:
no simple mission statement but trying two roll out 2 complementary
sub projects into a single community.
My fault for not catching that one. I have stated that the goal is to
build a single community. See below.
Warning signs
=============
* Orphaned products:
Apparently no
I'm not certain what you are trying to say here.
* Inexperience with open-source:
Limited experience if I judge by the number of OSS related hits tied to the
proposed committer names on Google. Only 3 names get some hits.
You can also how Zimbra as a corp currently gets it here:
http://www.zimbra.com/community/
Valid concern.
* Homogenous developers/salaried developers:
Definitely yes, all work for 2 companies with strong hierarchical ties in
the proposed committer base
Again, the proposal was immediately met with volunteers. I will state
that everybody involved fully understands that the current level of
diversity certainly would not meet the incubator's exit criteria for a
project - and everybody supports the goal of building a diverse community.
* No ties to Apache products:
True
Again, immediately upon seeing the initial post, several people
suggested a number of possible ties.
* Fascination with Apache brand:
True, just see prc@ activity.
I understand how you could see it that way. For those not on the PRC, I
sent a draft email yesterday which essentially said that discussions
were underway with the ASF and gave an overview of what AJAX is.
To help you see it another way, take a look at the following link:
http://ajaxian.com/archives/2005/12/apache_ajax_too.html
AJAX is hot. People outside are watching. IBM and Zimbra will
undoubtably get a lot of press people asking questions. My experience
has been that such people are well trained in saying "no comment", but
the fact is that there is interest, and at some point it makes sense to
meet such interest with facts.
As is, I can't see a single reason to support the proposal ans see several to
vote a strong -1 on it in its current form :
- The proposal is too large to incubate, it's hard enough to create a community
from scratch around a single well-defined goal and codebase, rolling 2
together is suicide in my book.
I don't mean to minimize the concern, but we have incubated larger. As
we have seen in this and other donations - IP lawyers are very
interested in clearly delineating the precise origins of each component.
As such, we've overstressed the separate nature of these pieces.
Just to be clear: the goal is to build one community.
- I don't see any benefit for the ASF and several drawbacks (more
hard work and strain on resources, possible PR complications, additionnal
strain on friendly relations with other OSS groups like Eclipse)
There definitely is interest in AJAX at the ASF. If not Zimbra, then
Dojo. And as Sanjiva and Dims have eloquently put it
I have no patience for any kind of "this space is mine, you keep to
yours" type nonsense. I totally agree that the only discussion here
should be does ASF want to take this on or not, not on whether
Eclipse folks feel it "rightfully" belongs there or not.
and
I really don't mind if Apache gets into Eclipse tools/plugins. We do
have Eclipse plugins in Axis2 project. We also have another plugin
for running Geronimo inside WTP. So it's not a new thing and the
proposal has my +1. Please pardon me for being blunt, I don't
really care about what happens inside IBM/Eclipse or who said
what/when. All i know is that we have a proposal in front of us and
as a community we take it or leave it or ask for changes if we think
they are needed.
- There's no mentor yet ! Bad sign...
Again, two volunteers within moments of posting alone.
- The odds of this project of successfully exiting the Incubator based on the
diversity of community criteria seem very low to me: there are too many
initial committers and most of them will have strong internal communication
channels which will be invisible from the community.
This will all move to mailing lists.
- I don't believe most of the proposed committers would get committership
on their own merit and I would hate the Incubator to become an easy way to
bypass the meritocratic model of the ASF: work at IBM and get a free
committership when they donate the codebase to the ASF ! Most of the time
you end up with paid-for-committers that only last as long as they're told
to work on the project. (This is not pure paranoia ;) just look at Pluto if
you want to see it in effect)
Valid concern.
In summary I see this proposal as a high risk, low value offer to the ASF
and would definitely pass on it.
I don't want to minimize the risk, but I do think you have
underestimated the interest/value.
- Sam Ruby
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
