[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: Re: [tm-pubsubj] Points from psdoc post 29/01
Peter > How do topic maps deal with subjects? First they represent subjects > formally as abstract "topics". In XTM documents, topics are > represented by a <topic> XML element. A topic should represent an > unique, well-defined, non-ambiguous subject. > ******* > This equates 1 topic with 1 subject, whereas the rest of the text talks > about "topics" and "subjects" (plural) which may confuse the reader. Exact. Using singular form all along would be clearer. I will correct that. > 4. > A published subjects documentation shall include a formal statement > from its publisher, expliciting its conformance to this > recommendation, and its intention to maintain the documentation > trustable, and the PSIs stable. BTW it should be "published subject documentation set" ... that one escaped the rewording. > Having just had a discussion with a colleague about the venality of > publishers :-) I wonder if Requirements #4 is enough to stop the > indiscriminate publisher from simply lying about the trustworthiness and > stability? Once this takes off and the marketing people get into the > act, will we see XTM spam masquerading as useful information? I don't think there is a real risk here. Like everywhere else, trust in that matter will be a bootstrapping process. In fact, I figure efficient use of PSIs will occur not in the open wild web, but inside communities of practice/knowledge, big entreprises intranet portals etc. For PSIs to make sense, their semantics are somehow to be shared by their community of users. Very technical PSIs will be published by well-identified publishers for well-identified communities of users. Having garbage popping around is likely, but I don't figure what would be the business case for them. > 4.3.2: is it our job to go as far as writing a *canonical* DTD/schema > for PSIs (as hinted at in Requirements #3), like Bernard's pubsubj.dtd? Nope. We are now over that stage of reflection. As said yesterday, our job is to say: "If you use such syntax/format, we recommend such structure". The DTD I provided was in that spirit (if you publish XML subject indicators, they could look like that). We are moving now towards a "cross-syntax meta-model". Thanks for your input. Bernard
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC