> David, I think you're doing this the wrong way round
Hi Martin
Thanks for the message. You've done a great job with your scripts.
I think you misunderstood me, or more likely I didn't make myself
clear.
Of course the data is stored as plain text. There's no markup stored
with the questions. The XML RSS file you get with that URL I posted
isn't how the data is stored. That'd be crazy! No, it's made up on the fly
in request to a query. It's no different that suggesting that your data is
stored marked up as HTML 'cos that's how it's delivered.
Think of the XML as a wrapper. Data can be wrapped in any way you
like. I happened to have used the RSS example as it's a useful way of
'discovering' the data in a format that's widely used on the web.
Because it's XML, it could just as easily be an IMS XML wrapper. Or
anything else you like. You can now transform relatively easily to RTF
or even PDF.
You're examples are a great credit to the work you've put into your
system. but don't confuse the functionality with the storage or the
wrapping. I think there's lot of potential debate about functionality of
delivery. After all that's what everyone who run thier own system is
interested in.
However, what I was hoping to prompt was a discussion of these
wrapper issues, IMS or otherwise. There are a great number of
databases and methods of querying them. Trying to create a CGI that
knows about all of these might be tricky. Maybe a structured query
using XML, maybe using SOAP for example, might lead to interesting
ways of linking resources or finding them?
Keep up the good work. Offline, let's talk about linking up our
databases?
Best regards,
David.
|