Software thoughts

First off, it'd be really nice to have categories on here... I haven't tried to find out it Blogger does that or not, but it's yet one more reason I'm thinking of moving on to my own host.

Anyway, at work these past few weeks I've been working on Gibraltar, a content management system that will run all of the Center for Family History and Genealogy's websites, including the Immigrant Ancestors Project and the Irish Mormon History site. Why? The main reason was that we needed to create the Irish site, and in the future we'll inevitably have more sites in a similar vein, and it would be much easier to have a framework which we could quickly and conveniently put the new site into. Like cookie-cutter websites, except extensible so they wouldn't be boring replicas of each other (CSS and other things would help with that). So I've designed the whole thing and have written a good part of it, but lately I've gotten bogged down. And the powers that be are anxious (which is quite understandable, since the project has taken far longer than I anticipated).

So, I'm going to switch my focus to getting the prototype up with editing functionality as soon as possible. Which leads me to the real point of this post: design-wise, is it better to build the whole system up first and then release it to the users (thus giving them a polished product), or is it better to focus on one bit at a time and release it incrementally (thus allowing them to use it sooner, albeit at reduced functionality)? There are pros and cons for both, of course, and I suppose it depends on the situation.