> Now, I might have this wrong - but you're suggesting that 
> there should be a standard way of... describing data.... 
> suggested by the BBC, so that all systems structure their 
> data in the same way?

Not quite. There should be one or more standards for appropriate
applications suggested by a wider community (broadcasters - of which the
BBC is but one) so that all systems structure their data in a way that
is able to be widely understood. For example, agreeing a common ontology
for programme/schedule data. The partners don't even have to publish the
data in the same formats, just agree the ontologies and data formats to
allow apps to do transforms and comparisons more easily between sets of
data. That's more what I'm getting it.

Call me a dreamer...

> 
> I think the BBC has given up trying to do that even 
> internally, and instead relies on being able to map piece of 
> data A in system X on to piece of data B in systemY, and be 
> reasonably sure they're the same thing.

Isn't that what an ontology is supposed to do? If so, why not just write
it down? 
But yes, I understand why even the simplest thing might turn into a
major nightmare.

> 
> I really get the feeling that 95% accurate mappings between 
> different ways of describing stuff is the best we can hope 
> for. The suggestion of gentle harmonisation is preferable to 
> 'having to do it X way always or else' in any big 
> organisation. And, in fact, any system involving fallible 
> meatbags doing data entry.
> 
> This is coloured by having spent some time up to my elbows in 
> SMEF and datamodelling at the BBC, and also by trying to 
> persuade editorial teams to enter their HTML metadata vaguely 
> consistently.
> http://www.bbc.co.uk/guidelines/smef/

I agree, the problems of the real world do get in the way, and stuff
needs to get done. The only thing is that, great though mashups are,
they tend to be incredibly ad hoc, so when the slightest thing changes,
lots of developers pipe up aying things like, "oh look my feed
aggregator or somesuch has just died a death because they've changed the
feed format".

Some things never change, do they?
*****************************************************************************************
To view the terms under which this email is distributed, please go to 
http://www.hull.ac.uk/legal/email_disclaimer.html
*****************************************************************************************

Reply via email to