As I mentioned a few weeks ago, I've recently had to spend some time coping with the complexities of Microsoft Word based document management - and came away horrified at the inefficiencies this imposes on users.
As a zdnet blogger I see the various bug and action notices flying around as a result of the recent re-design. Now, personally, I like the new layout much better than the old one but that's not the point here: the point is that the change subjected the editor, some of the bloggers, and numerous support people to a whole bunch of stress and work.
And the question, of course, is exactly the same one that came up when dealing with Microsoft Word documents: how much of this hassle is really necessary, and how much is human effort being abused to sweat together a workflow compiled out of anachronistic bits?
As a Drupal/Cocoon afficionado my intuitive reaction is that implementing an imagined ideal system in which both template and backend changes can be safely pioneered on the editor's desktop and pushed out to one property after another without too much concern about blowing up other things faces the usual competence and co-operation challenges, but no technical barriers - and would be a lot cheaper to run afterwards.
Unfortunately I'm not going to get to try it and therefore can never really know for sure - but, I think, it generally true that where you see complex workflows supported by systems that "kinda just growed" in place, you'll also see human time and commitment being used as the glue that keeps things together.
And the corolary, of course, is that most such systems acquire near sacred status among their victims but should logically be seen as just so much steak on the hoof for non IT employees - almost all of whom would be better off, if only they knew it, with fully integrated workflow solutions.
Summary: I am becoming increasingly intrigued by the opportunities for organizational efficiency improvement through better user workflow support - specifically through work aimed at taking the human effort out of keeping lash-ups working.