This is very clear :-). It is exactly this sort of thing that concerns me.
It's not that it's intrinsically difficult, but:
- a it's easy for someone to overlook when writing an *application*.
(Remember that - at present - inheritance is implemented by the application.
- *every* application - potentially - has to consider whether
'inheritance' applies to it. It may decide it doesn't need to worry, but
the author of the documents may assume that all applications are
'inheritance-aware'. This will certainly not be true.
- the word 'inheritance' is understood differently by different SGML/XML
experts and therefore may lead to differences in implementation.
My feeling is that many experienced SGML people 'know what 'inheritance' or
'applies to' means - and there may indeed be a consensus in the community.
This is *not* shared with the vast and increasing number of those coming to
XML without an SGML background.
The most constructive thing - for me - is if the XML/XLink drafters could
give us a precise definition of the words 'applies to' and 'inherited'.
This definition have to be precise enough that implementers could decide
what and when applied to them. [However, do not underestimate the amount of
work in this - we still are not very clear what a 'processor' is, when
'validation' takes place, etc. , although we get by at present.]
My hope - and it's probably too optimistic - is that we can produce an API
which describes the implementation of these difficult concepts.
P.
Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg