thank you all for your replies on this query.... if i might take the liberty
to get a little more specific, and shed some light in relation to my actual
project, given that this is the only place i know where to turn regarding
the project itself is related to developing a useable set of XML to enable
use of nationwide railway timetable queries in a plethora of
devices/media... i have been thinking along the lines of using XML to define
both the query, and the result set.... the current requirements
specifications, outline extensibility in order to support future use in
PDA's, Mobile Phones, Digital TV, third party data-users and others.
this in itself is not too great a concern - or is that an assumption I
shouldn't be making?...
added to this, is the complex nature of queries/resultsets that might be
required over the duration....
this is quite in depth, so rather than cover it here, i have posted an
sample of the requirements, at the following URL:
what i would like to find out, is if there is a metadata framework (is this
right?) that might apply to my application/data, which has a pre-existing
DOM that i could possibly make use of, and conform to, for openness, rather
than taking my own initiative and developing some hybrid/bespoke monster,
which can't interoperate when i most want it to.... !
if there does not exist any pre-existing stuff, if there are anything anyone
would like to say to ensure it is designed right, i equally am open to
failing that, i am still going to the XML technical seminars in London at
the end of the month, but then deadlines are always hectic....
> -----Original Message-----
> From: Tim Bray [mailto:[log in to unmask]]
> Sent: Wednesday, November 04, 1998 4:07 PM
> To: [log in to unmask]
> Subject: Re: basic xml questions
> At the moment you have to generate the XML yourself. It's
> actually not
> that hard, many people end up with a lot of statements along
> the lines of
> out.println("<"+ type + " " + attr1 + "='" + val1 "'>");
> or more efficient equivalent. There's some talk in perl-land
> of including
> an efficient XML generation API just to avoid sweating the
> syntax... but
> it's not that much sweat.
> BTW, if this is an environment where you control all the
> computers, you
> can probably skip the validation, it adds significant cost. If it's
> not, and you're going to be accepting XML instances from non-trusted
> outsiders, you're going to have to a lot more sanity-checking
> that just
> DTD validation. -Tim