Re: Software strings in documents
- Hi Leigh,
actually we are using a RDB CMS that tracks all dependencies. We are thinking of implementing statuses as you describe and your post makes me aware of the fact that indeed changing one topic's status can have a huge effect on other topics. This will not only challenge the developers how to tackle this but it also affects the management of content by the authors. We could start a different post on this subject, the more I think of it the more complex it becomes (what happens when you 'approve' a topic wile it's dependencies have not been approved yet?).
--- In email@example.com, "Leigh White" <lwwhite5@...> wrote:
> All excellent points, Kris. I was, as you point out, thinking of native XML CMSs. RDB CMSs probably would not track the dependencies so closely, but then one has to ask whether that's a good thing or not...a whole different post! :-)
> But as to your point about basing conref or keyref usage on how easy the tool makes adding one vs. the other, the purist in me cringes to think of groups making architectural decisions based on a tool, because we do tend to change tools. But of course, I realize that's a daily reality. Balancing architecture with tool and resource availability is always tricky. My preference is to start with a golden ideal and back off of that in increments only as necessary to accommodate reality and only when there's a reasonable business case for not doing the best thing. So when creating references to small, minimally-formatted phrases, I'll really root, root, root for the keyref team :-)