Clarity Blogs - now with Community Server 2.0
So it took a little time, but I've finally moved the Clarity blogs over to a customized version of
Community Server 2.0. We have a decent number of customizations, particularly around our global categories that aggregage content across all our blogs. I didn't want to start moving my changes until I had the final RTM codebase. Unfortunately, I had started a change log to keep track of my changes to the source code, but it wasn't super complete. So this time I wanted to both document the code and also have a document listing out at least all the changed files. Also wanted a sql script that would make any necessary changes to the data layer. So all that took some time, but now I feel like I know where our customizations are. I'm still interested in writing about how we implemented the global categories and that should be easier now that I have better documentation.
I did run into one issue that I hadn't seen before and it took me awhile to track down. I was deleting a blog and when I did all the global category stuff broke. Turned out the table that links posts to categories was losing all global category data. The reason: when a user's account is deleted, the code loops through that user's categories and deletes all rows in the table for a particular user category and post. Because of the way our global categories work (to allow any user to post to them), it was also looping through all the global categories and deleting rows for those as well. Took awhile to figure out, but makes perfect sense now. Kinda glad I ran into this now versus later because now I won't have a surprise if I have to delete a blog. Turnover at Clarity is so low it's not something that I need to think about very often, which is a good problem to have.
Telligent has announced they will introduce tags with
version 2.1, ala
Del.icio.us or
Flickr (or a myriad of other sites), but I still think we are going to want to keep our global category approach. It just works better for us. I'll have to wait and see how it is implemented though. It's enough work to get our customized code into a new codebase that I don't think I can upgrade with every minor version, unless there is something very compelling. and even then, I may start trying to do a diff on the source code trees and just grab the stuff I want.