[tei-council] Fwd: bug report for Council, if you like

[James Cummings]

On 04/10/10 20:53, Martin Holmes wrote:
> Right. So the only remaining question is one Sebastian raised, and which
> I couldn't find the answer to in our previous correspondence: do we:
>
> a) Require that the last component of a polygon be identical to the
> first, thus closing the polygon, and deem the attribute value invalid if
> this is not the case; or
>
> b) Deem any attribute value whose final coordinate is not identical to
> its first to be closed anyway, thus assuming the existence of the final
> point where it is not explicit; or
>
> c) Allow that some shapes may not be closed (in other words, they may be
> paths), and therefore assume that any shape not explicitly closed is open.
>
> Thoughts? My instinct is to be permissive and choose c), but then we
> have the issue of possible conflict between the definition of a<zone>
> and the idea of a path. That's somewhat moot anyway, because the current
> definition of a<zone>  is inconsistent with @points:

I would also say c).  I think the TEI should be providing the general 
solution here for both paths and areas.  I think having a) or b) is a 
matter for local encoding guidelines and/or processing.  I recognise 
that b) is probably the most common way it would be used, but don't see 
how it really benefits us to make this a requirement rather than just a 
common practice.

-James

-- 
Dr James Cummings
Research Technologies Service, University of Oxford