Custom schemas and semantics

[Patrick Durusau <patrick@durusau.net>]

Greetings!

Something Rob said recently about custom schemas impairing 
interoperability made me start thinking about why we have custom schemas 
(read various DTDs/schemas generally as well as the ability to have them 
in DIS 29500) at all?

One reason is that markup elements and attributes have semantics that we 
wish to express and then apply to some set of data.

Another reason is that we want to process some data based upon the 
semantics that have been imposed by markup upon some set of data.

But in either case the semantics are simply implied by the markup and 
isn't easily discoverable from the markup per se.

Which of course means that a custom application must be written with the 
usual functions, save, copy/paste, etc., for the custom schema.

Seems to me that if one has a sufficiently general markup structure that 
can have custom *semantics* applied to its elements, that should reduce 
if not eliminate the need for custom schemas.

While it would depend upon the effort put into the metadata but with 
sufficiently detailed metadata it would be easier to discover the 
semantics of a document because it would be explicit rather than simply 
implied by the markup structure of a custom schema.

So, the advantages of using metadata seemed to be two-fold: 1) It 
becomes unnecessary to write custom applications for customary 
functions, 2) The semantics of the document would be explicit and not 
implied.

I don't think this requires any changes by the metadata SC but I thought 
it is an interesting way to think about using metadata with ODF documents.

Hope everyone is having a great day!

Patrick

-- 
Patrick Durusau
patrick@durusau.net
Chair, V1 - US TAG to JTC 1/SC 34
Acting Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
Co-Editor, OpenDocument Format (OASIS, ISO/IEC 26300)