On 09/05/2008, at 6:03 PM, Jason van Zyl wrote:
Ah, hold on there. 1) Since when did accepting new bodies of code
be decided between two people.
It shouldn't be, and IMO this discussion should continue and get
some more opinions (as I said "if there is support here"). After
that, the sandbox is the best starting point and if anything goes
in that is more than James' own work, then the IP clearance papers
should of course be filled in too.
But one standard for everyone, please. You checked in a more
significant contribution without discussion *at all* just a few
days ago (http://svn.apache.org/viewvc?view=rev&revision=653572).
I'm a committer and Ivan did not get access to our repository, and
it was no different then the GIT provider.
The difference was that the GIT provider was discussed and contributed
via JIRA, like all other contributions. The other provider appeared,
on trunk, completely unannounced. It's clearly a double standard to
then turn around and say things like "Since when did accepting new
bodies of code be decided between two people."
But to be absolutely clear, I think the Accurev contribution is great
in the log run, and I'm glad to see Ivan on the lists supporting it. I
certainly have no complaint with them. Let's move on.
Whereas the difference is access to our repository for what is
honestly a duplication of much code that exists, and far less then
something like an SCM provider. There is a stark difference because
I don't freely hand out access. There are alternatives for plugins,
especially given our model is fully distributed, and we already have
a ton of orphaned plugins.
What I checked in was written by the only people in the world with
clear authority to write that SCM provider. Whereas these plugins
largely duplicate what exists. So I think there is a stark difference.
Firstly, no access has been handed out that wasn't already present. If
you now have a problem with the sandbox openness we instituted as a
group, please raise that separately.
I wasn't debating the relative merits of the two contributions. As far
as I'm concerned, the discussion is ongoing - as I said, the wagon
plugin as is needs some work to be suitable for Wagon, if at all. I
don't want to maintain something that is only duplication either. The
license plugin in particular needs to take a look at ways to interact
with the existing techs rather than being a new thing (there is also
IANAL at Mojo that I just saw), but I think better license handling is
worth pursuing if James has ideas.
Let's just continue that discussion, as two separate threads for each
thing.
Thanks,
Brett
--
Brett Porter
[EMAIL PROTECTED]
http://blogs.exist.com/bporter/
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
