On Fri, May 28, 2010 at 11:25:39PM -0700, Ian MacGregor wrote: > [20:37:01] <mchua> but months later, the status is "wow, zikula is hard" > [20:37:08] <ardchoille> ah, ok > [20:37:11] <mchua> the thing we were trying to make work: > https://fedoraproject.org/wiki/Insight > [20:38:09] <mchua> and the issue is that it's taken months to try and > get the last 10% of the features in - zikula upstream has great people, > but they're tight on time to help > [20:38:36] <ardchoille> Ok, I'll get with Karsten and see what the > status is and if there's anything I can do to help > [20:38:59] <mchua> and the folks on the Fedora side don't have the > zikula knowledge, nor the time to acquire it given the state of "learn > zikula!" resources (very few) out there [...] > [20:39:56] <mchua> ardchoille: and for the CMS - I think there are two > things to look at in parallel > [20:40:09] <mchua> ardchoille: the first is "does websites still want to > examine a CMS solution?" which I think is worth asking Karsten and the > list > [20:40:39] <mchua> (and if the answer is yes, "do we want to look at > options other than zikula for it?") > [20:41:12] <mchua> ardchoille: the second is to look at the zikula > progress, which is what we said we were going to check out (but it's > been months, and things have changed since, so questioning that decision > again is good) > [20:41:16] <ardchoille> yeah, if zikula is taking too much time, perhaps > there is something else we can use because we'll still need to maintain > it From the perspecitve of hosting this in Fedora Infrastructure.... I'm wary of including another CMS. There's not enough sysadmins and developers to go around in Fedora Infrastructure so putting new apps up (especially in the CMS realm where there's a whole lot of security holes) is a large commitment. With zikula we were hoping for two things: 1) Everyone wanting a CMS could use it. 2) Docs/marketing/etc would bring people to infrastructure to manage the installation, update the software, etc. That way we might have an initial drain on resources to train people but the harder task of maintainance would not hurt us as much. These wishes have been a mixed bag. There's been upstream involvement on the setup and deploy aspect. There's also been upstream involvement on coding issues. However, there hasn't been the tight integration of sysadmins in Fedora Infrastructure working with sysadmins for zikula and eventually becoming an integrated team. Since we haven't deployed zikula yet, I'm not sure how much of an issue this will be (or whether it will all resolve itself) after we deploy. If you're going to look at a different CMS for fp.o, you'll need to address these issues as adding services without adding manpower to take care of it both from coding and from the sysadmin side is a losing proposition. Also, be careful about judging the slow rate of getting zikula into production for Insight as being due to coding for zikula being hard... You may find that the slow rate of getting something adopted has less to do with the software and more about what we're trying to do with it. -Toshio
Attachment:
pgp2Gju8GKTMB.pgp
Description: PGP signature
-- websites mailing list websites@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/websites
