xml:id RE: [sc34wg3] Compact syntax requirement question

Patrick Durusau sc34wg3@isotopicmaps.org
Tue, 19 Jul 2005 07:53:50 -0400


Consider the age of HTML and the lack of any demand for a "compact" 
syntax to replace it.

Or that XQuery appears to be proceeding with an XML syntax.

I think part of the problem is not everyone shares the background that 
makes you say "XML is too painful."

Realizing that we all have other committments, perhaps making the case 
for a "compact" syntax along the lines of LTM or AsTMa=, rather than 
assuming everyone shares the same view of an XML syntax might ease us 
past this point.

Personally I think very abbreviated syntaxes are probably the favorites 
of heavy users while XML (but still compact) syntaxes would probably be 
favored by less heavy users.

Both are legitimate user communities.

Hope you are having a great day!


Lars Marius Garshol wrote:

>* Lars Marius Garshol
>| CTM is specifically meant *not* to be an XML syntax.
>* Bernard Vatant
>| Indeed? Where is this specified, [...]
>Well, that was the problem: it wasn't specified in writing anywhere.
>| and what is the rationale for not using XML?
>That CTM should be compact, etc, and usable within TMQL (because we
>need it for the INSERT support in TMQL updates). XML is just too
>verbose for that.
>| If I read the following requirements.
>|   (f) CTM needs to be compact, and easy to write.
>|   (g) CTM needs to be easy to read.
>Yep. I think it's quite clear that XML is going to have to struggle
>quite hard to compete with
>  [bernard : person = "Bernard Vatant"; "vatant, bernard"]
>(though this is of course not meant to imply that CTM == LTM, or
>anything like it). 
>The rationale for doing this is that a for a very large number of
>cases one needs to be able to hand-edit topic map content. In these
>cases XML is just too painful, and today people have to use LTM or
>AsTMa= (and in fact lots of people do). It would be better if there
>were something standardized they could use, and since we will need
>something standardized for TMQL updates anyway we figured it would be
>better to define this now, so that CTM, TMQL, and TMCL can have
>coherent syntaxes.
>| I would like to add :
>| 	CTM should be easy to validate
>That's an interesting requirement, but I'm not sure exactly what you
>mean by it. Validate on what level? Syntactically? Or against a
>schema? And easy for whom? The implementor or the user?
>| 	CTM should be easy to translate from/into other TM syntaxes
>| 	(e.g. XTM)
>Hmmmm. What do you mean by "translate" here? If you have a topic map
>engine with support for import/export of all the syntaxes then there's
>really no way to fail this requirement, which makes me suspect you
>mean something else.
>| So, if not XML, what kind of format do you figure? Plain text, with
>| specific grammar?  With specific editing tools? Specific validators
>| and parsers? Specific stylesheets?
>Specific grammar and parser, yes. I don't know that there is any need
>for the others. A simple editing mode for your favourite text editor
>is all that's needed for developers, and they are the only ones meant
>to use this anyway (real users should use real editing tools). Given a
>proper parser there is no need for a specific validator (the parser
>will do syntactic validation as part of parsing, anyway). What you
>mean by "specific stylesheets" I don't know.
>| This does not mean "easy" to me at all. If, just for sake of
>| compactness (compacity?), I lose all XML technology support, count
>| me out, I'll stick to XTM. Given existing tools, today : easy to
>| write, read, parse, validate and translate means to me XML, sorry.
>Why are you so negative to this? We're proposing to standardize a kind
>of syntax which is already very widely used, primarily because it's
>needed as a piece in another standard anyway. What's the big problem?
>Why get upset over that?

Patrick Durusau
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model
Member, Text Encoding Initiative Board of Directors, 2003-2005

Topic Maps: Human, not artificial, intelligence at work!