xml:id RE: [sc34wg3] Compact syntax requirement question
Wed, 20 Jul 2005 10:35:32 +0100
Lars Marius Garshol wrote:
> * Robert Barta
> | And this was exactly my counterargument Re: Bertrand. If he claims
> | that an XML editor 'guides' an author appropriately and there is no
> | need for tools to _really_ check whether the map is TMDM conformant
> | (what did we talk about in the last years?)...
> Ah, OK. I didn't see that that's what you were arguing. I certainly
> agree that XML editors, even though they can use the schema to guide
> you, don't really provide full guidance. They don't know all the rules
> for what is allowed, nor can they help you much with references, etc
> And that, I guess, highlights another point in this discussion. We're
> not creating CTM to make life easier for end-users. End-users should
> use a proper editor. We're standardizing what's already a commonly
> used tool among developers (LTM/AsTMa=), and we're doing it because
> it's needed for TMQL, anyway.
I think it is important to point out that just as there are levels
of "validation" it is *always* possible to make a nonsensical
statement at the next level up. If the ultimate level is essentially
a form of thought, this only makes sense: one can make nonsensical
statements. From this perspective, I've always found arguments about
validation that operate from misplaced levels (e.g., trying to
validate syntax at a lexical level, higher-level data structures at
a syntax level, etc. and particularly "semantics" anywhere within a
syntax expression) to be rather specious. There is *no* guarantee
that any statement made by human or computer, makes "sense," as the
problem is significantly more complex than that. Only an engineer
whose focus remains solely with syntax and parsing could make such
errors -- though this is rather rife nowadays with the "Semantic
Web" meme running its course.
XML does have substantial benefits over many other base syntaxes,
and these have little to do with its structure per se: there are
ISO and W3C standards and specifications (resp.) for it; it is
widely known and has a very long history of use, coming from the
GML and SGML work back into the 1960s; there are thousands upon
thousands of tools -- proprietary and open source -- that can
parse, process, edit and validate the syntax; extensive support
for XML is either built into or available as libraries for Java
and most of the most popular programming languages; there is
substantial expertise and knowledge amongst the various people
working on the standard; the network effect is currently in XML's
favour; lastly, both ISO 13250 and XTM are expressed in SGML and
Lars Marius Garschol wrote more recently (and was answered by
>>Well, a plain XML editor gives you a bit more guidance, but in
>>principle the situation is the same.
> ...this is exactly my point. You need tools in _any_ case. Using XTM
> helps to 90% to get a TMDM-sane map. But not 100%, and so, yes, the
> situation in principle is the same.
Only in the most bare sense. In practice it is hugely different.
You still need "tools" (how else?) but with an XML syntax these
tools (and the knowledge and support behind them) are vast.
I'm not arguing against a CTM syntax -- indeed, I've been an
extensive user of LTM over the past three years -- but I find
the arguments comparing an XML syntax with a new one extremely
uncompelling. XML has enormous advantages over almost any other
syntax *at this point in time*.
Murray Altheim http://www.altheim.com/murray/
Strategic and Services Development
The Open University Library
The Open University, Milton Keynes, Bucks, MK7 6AA, UK .
believe that everything is for
you until you discover
that you are for it
"The Robin and the Worm" by Don Marquis.