Eugene Eric Kim wrote:
> On Sat, Aug 07, 2004 at 11:03:08PM +0100, Murray Altheim wrote:
>
>
>>I've also been thinking about this kind of thing. My approach was
>>to consider that one of the problems with wikis is that any given
>>page is not really about *one* subject (i.e., the title) but about
>>potentially many different subjects. It occurs to me to use the
>>existing Dublin Core DC.subject approach to embed <meta> keyword
>>content within the wiki pages, then have a harvester build up the
>>wiki index from that, not simply from the page titles. This would
>>work pretty well and be easily brought into a Topic Map navigator.
>>My only problem in implementing such a beast is that I'm already
>>busy with an entirely different project. The two projects could
>>meet in the middle, but I don't have the time to begin hacking a
>>wiki tool right now...
>
> How would you propose the subject tags be authored? As additional
> text fields in the Wiki pages? And would subjects correspond to other
> Wiki pages?
>
> If we can come up with a reasonable spec that makes sense to both of
> us, then we can hack this functionality into PurpleWiki for viewing
> from Ceryle. (01)
Eugene, (02)
No need to come up with a spec, it's already written: (03)
Expressing Dublin Core in HTML/XHTML meta and link elements
http://www.dublincore.org/documents/dcq-html/ (04)
I'd recommend just glomming on to OCLC's way of doing things. The
semantics are already there, and syntax too. An example of the
above document in practice might look like: (05)
<html>
<head>
<title>Ceryle Tutorial</title>
<link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"; />
<link rel="schema.DCTERMS" href="http://purl.org/dc/terms/"; />
<meta name="DC.title" content="Ceryle Tutorial" />
<meta name="DC.type" scheme="DCTERMS.DCMIType" content="Text" />
<meta name="DC.format" scheme="DCTERMS.IMT"
content="application/xhtml+xml" />
<meta name="DC.date" content="2004-07-05T11:25:47" />
<meta name="DC.creator" content="Murray Altheim" />
</head>
<body>
... (06)
The big question is how to connect up the declared subjects of
each wiki page with some kind of controlled vocabulary. But then
again, maybe this isn't necessary at all given that wikis at any
moment in time have a "controlled vocabulary" built into the
structure of the wiki and its wiki words. So maybe just populating
a <meta name="DC.subject" content="" /> element with wiki words
might be enough, at least as a first crack at it. A spider could
harvest all that metadata from the web site, generate an LTM or
XTM document that could then be fed into Ceryle. Or somebody
could build a spider directly into Ceryle... (07)
Murray (08)
......................................................................
Murray Altheim http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK . (09)
It may have been the provocative title of a book and then a movie
adaptation, but Prozac Nation is no longer just a notion within
the realm of entertainment, nor even a living reality only among
those actually taking the drug. The 24 million prescriptions in
the UK (in an adult population of 42.5 million) have now entered
the nation's water supply in significant but undisclosed amounts.
http://news.bbc.co.uk/1/hi/health/3545684.stm (010)
--
This message is archived at: (011)
http://collab.blueoxen.net/forums/cgi-bin/mesg.cgi?a=tools-yak&i=41170107.1090204@open.ac.uk (012)
|