(response to) comments on the draft...

[Peter Constable]

> From: Addison Phillips [wM] [mailto:aphillips at webmethods.com]


> Mark and I have crafted a response on the extensions issue, which
awaits
> final blessing from one of us. You will have that shortly.
> 
> I will note that your question here suggests that we envision a new
> protocol that uses language tags, as opposed to extending language
tags as
> used in their current range of protocols so that they can contain
> additional language-related information. Neither Mark nor I
necessarily
> envision a new protocol I think.

Perhaps all will be clear when your general response is ready. It just
seems that no extension can be used until a new protocol, by which I
mean a new RFC, is defined. Either it is a replacement of the RFC you
are now preparing, or it is a distinct protocol that derives its tags
using this RFC. In either case, the syntax for extensions *could* have
been defined in that specification. Yet for some reason you are putting
them here, with the result that processes or consuming protocols that
need to interpret language tags will also need to allow for extensions
that may have no relevance for them whatsoever. This strikes me as
unusual, so I can only imagine you and Mark have some particular purpose
in mind for doing things this way. Yet that has never been explained
that I recall.


Kind regards,
Peter Constable