[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: reason for application/iotp-xml (was RE: Registration of MIME med ia type APPLICATION/IOTP)



At 01:55 PM 3/11/00 -0800, Paul Hoffman / IMC wrote:
>IOTP is a trading protocol. It involves talking about money and 
...
>To date, we have not seen anything formatted in XML that would be useful to 
>automatically hand to an XML parser.

I also am troubled by the ad-hoc-ness of the -xml suffix; but unlike Paul,
I do see a potential benefit.  First of all, is there an existence proof
of an application that does generic XML processing (whatever that means)
of an XML vocabulary it doesn't understand, the example here being IOTP?
I have some:

1. A web crawler building a full-text index
2. An XLink processor building a topological map a la Google
3. A rules-driven firewall filter deciding whether to let something
   through
4. A schema-driven datatype-checker

There's also an argument from principle.  One of the core ideas of XML is to 
move past the notion that data should be encoded in a format proprietary 
to a product or application.  The idea is that the door should always be 
left open for future processing of data in ways that are unpredictable at 
the time of creation.  The generalized -xml convention certainly is 
consistent with that spirit.

Having said that, I still have the feeling that if we do all agree that
this is a good effect to achieve, there has to be a better way than
this -xml convention.  Media types have been working well in a 2-part
structure for a long time, going to a 2-and-a-half parts smells funny.
But I haven't thought of anything better. -Tim