|
From: <ch...@op...> - 2004-10-20 20:34:29
|
First off, a huge apology for dropping off the irc session on monday. It wasn't meant to express my dislike of the new interfaces in the slightest, it was due to my phone line getting cut and my internet going down. I tried to get back on but was unable to. I'm in the process of thinking them over and writing up my feedback. Overall I like all the motivations behind it, very much what I would like to see too. I really don't like the use of the Catalog interface, as I see that as making a radical change drawing upon what is in my opinion a bad specification, that gains us little, except some sort of justification that we're closer to some specs. I'm in the process of writing this up, have taken a few pages of notes, and will try to synthesize tonight. I was in the process of talking over some of the issues with James, but we got stuck as he couldn't access the interfaces, since http://svn.geotools.org/geotools/trunk/spike/dzwiers/ appears to be down. Could we get those back? I think they provide a good discussion point for moving forward on this stuff. And I'm sorry you can't get cracking on head right away David, but these changes have serious reprecussions, and need everyone's full attention, even if that takes more than an irc session. I personally didn't get a chance to check out the interfaces until the day of the irc, which was far from sufficient time. And they are much more radical changes than Andrea's FID mapping stuff, which generated additional bugs for me for many weeks. But I'm very much for improving our interfaces and moving forward in the directions suggested in the Data Access wiki page. I think we can start to move forward on this now, I don't think waiting until every developer is ready is the right thing, as it will then never move forward. But we need more than one core developer driving it, even if that means we drive a bit slower. I for one am definitely willing to put some time in, but at the very least we need those interfaces back publically available. Chris ---------------------------------------------------------- This mail sent through IMP: https://webmail.limegroup.com/ |