> Well, I don't think #HID2.15:5 is completely unreadable. In "real life" we > say thinks like "Take a look at the fifth paragraph in chapter 2.15". If > you'd say, "Take a look at the fifth paragraph in the sub-chapter about > dogs, which is a in the mammals chapter." won't let you find anything > quickly (unless you know the book by heart). ;-) > The only unreadable bit in the HID is that you don't necessarily know what > 'HID' means. Oh of course, I was thinking of numerical indexing as preferred method as well. I just re-read the conversation and I see what you're doing now, it's very cool. I actually think (contrary to others so far) that HIDs of the form #HID2.1.12:5 would work better for citation purposes than say something like #mammal.canine.poodle:5. Having hierarchical enumerated ids I'd say are also very intuitive; any end user seeing them on a page at the end of a paragraph could pretty quickly deduce what the numbers mean in relation to what they mean. I see now that the purple page linked in the first post uses letters and NIDs... neither of those I like, as I feel they add too much complication to a simpler concept. The spec is eight years old now though I see (and really never took off, right)? Perhaps instead of Purple, would you not just rename them "Hierarchical IDs" based on the HID subset you've got? Anyway, excited to see this when it's in plugin form, and congratulations so far, I like the idea! -- --Terence J. Grant -- DokuWiki mailing list - more info at http://wiki.splitbrain.org/wiki:mailinglist