TUNES vs the WWW (was: cliki changes + opinions + XML)
Sun Jan 5 07:57:01 2003
> [omitted details about logging issues with CLiki]
This seems a puzzle for ours great CLiki administrators. :-)
Apart from jokes, they are doing a good work, debugging and improving
CLiki (see for example the new syntax for external links).
> > Unfortunately it is not so. First of all a (meta-)language without
> > semantic is useless. As Erik Naggum, on comp.lang.lisp, noted:
> > Structure is nothing if it is all you got. Skeletons spook
> > people if [they] try to walk around on their own. I really
> > wonder why XML does not.
> I completely agree. In facts, I was speaking mainly about DTDs.
> The fact is that, if I got right, XML is used as a boilerplate term
> 1- the language with which you write DTDs;
> 2- all Your Own Special Purpose languages you can define using
> the language at point 1, i.e. a language whose syntax is described
> in a specific DTD document; Examples: RDF, XHTML, XSLT,
> XQuery, XPath, SOAP ...
> 3- last but not least, all the single phrases (i.e. the files with
> extension .rdf, .xhtml, .whatever) of all the languages at point 2.
> Of course 1 has a semantics, but with 1 you just express
> the syntax of a language at point 2, not its semantics.
> The DTD of XSLT does not say what a XSLT document
> computes. You must specify its semantics in some other way.
> Said that, I read somewhere that DTDs are bad both syntactically
> and semantically, that many of the most popular languages of point
> 2 (XSLT, XQuery and such) also are, and so on, so I reckon you
> are right in many respects when you speak about crappy semantics.
> What still I did not get is what XML is, 1, 2, 3 or all of above.
> In facts, I think it is a good example of _(buzzword).
You got it. We (as others in this project) share the impression
that this is another confused fad which periodically takes possession
of the whole IT industry (and, worse, academy), but with very little
I forgot to mention the relevant (inflammatory) link:
"TUNES vs the WWW"
As I noted on this CLiki node, W3C seems entranched in a mania
of reinventing everything from scratch, in contrast with TUNES
great ambition to synthesize the best of different CS fields
and then evolve from there (which indeed is a very difficult
result to achieve).