Leigh Dodds ldodds at INGENTA.COM
Tue Dec 7 15:02:28 CET 1999

> Over the past couple of days I've partially implemented an export function
> to produce Leigh's XDELTA documents (as a simple example of a possible XML
> format) for the data in Lucid keys.

Would it be possible to see this code, or at least the resulting XML
files? I'd like to put a lengthier example in with the XDELTA
documentation than I currently have.

Note that some of the verbosity in XDELTA (i.e. the length of
the tag names) was derived from the requirement to make the
format as 'readable' as possible. So that it wasn't opaque
to a user how their data was stored, and also potentially
allowing hand-authoring.

> what could we do with such data as XML that we couldn't do with
> the data as a simple structured file as above?

At the basic data level I'd say nothing. A data format is a data format.
You could put the whole thing into a complicated CSV (comma separated
value) format if you wanted.

Apart from the lauded benefits of interoperability and the other XML
hype, I see one of the big advantages of XML is that because so much
of the information is explicitly tagged you are able to add additional
layers above that of processing the plain data.

For example an RDF application layer might allow one to make inferences
about a data set (or sets) that would be otherwise difficult with
a format that is more context based.

There are also advantages regarding extensibility (and again this is
just a personal view). Take the DELTA comment structure. It includes
all sorts of different information and has been used to encode different
types of information during the lifetime of the format. When it
comes to processing that data the application needs to be much more
aware of this type of legacy usage. Explicitly tagging data mitigates
some of these problems (but not all, XML isn't a magic bullet by any

As a user of an application it wouldn't matter to me whether my data
is in XML or some other format. Until I decide to use another application.
Can it use this format? Will I lose any data? Can my colleague
use my data in his/her application? Has all my data been rendered obsolete
because the latest version of this package has changed the file
format (i.e. the Microsoft Word syndrome). I believe careful application
of XML can help in these cases.

But you're right to question it at this stage - the initial focus
needs to be on design and data modelling...

> A final point. It seems to me that the discussion so far has been
> dominated by computer nuts (no offence intended, we need you!)

...which I hope to be able to help with, as well as providing any
XML and/or programming support. I'm really keen to hear from the
domain experts out there who can provide hard usage facts.

I'll duck back into my shell now! :)



More information about the tdwg-content mailing list