dita — archive

[Christopher Nitchie]

Eliot, I think I see your point. There are two (at least, but let's work with two) approaches when applying key scopes to a map structure: 1. Logical key scope names, reused across multiple maps (e.g. "maintenance", "support", "specifications", etc.), in a consistent key scope structure. 2. Specific key scope names for specific collections of content (e.g. "TractorX", "TractorY", "Calculus101", "TheWorksOfShakespeare"), nested more-or-less arbitrarily, where cross-scope linking is relatively common. The only thing I'm sure of is that there *will* be both, sometimes within the same map. With a global key scope namespace, it would be impossible to combine publications authored with approach 1 into an omnibus map, because the first instance of a scope name would trump all others (thus kind of undermining the whole point of confining keys to a scope). But without a global key scope namespace, it becomes a major challenge to reuse topics authored with approach 2, because the scope paths will vary from publication to publication. Eliot, it sounds like you think case 2 will be more common. I honestly have no idea, though I have my doubts, and I'm troubled by the implications for case 1. The whole point of this proposal is to eliminate the global namespace for key names; it rankles - considerably - to introduce a new global namespace for key scopes. But I don't have any first-hand experience of situations where cross-scope linking would be necessary, much less the types of cases you're concerned about. Either way, explicitly-defined keys can still be cordoned off into their own scopes without polluting other parts of the map, which is the primary purpose of the proposal. So I'll take your word for it, but I'd be curious what other folks think. Chris