[tei-council] dating attrs

Lou Burnard lou.burnard at computing-services.oxford.ac.uk
Fri Apr 13 08:42:13 EDT 2007


Syd Bauman wrote:
>  Since we all voted on
> it, I don't think there is much controversial there, except as listed
> below. On the other hand, paying close attention for class errors,
> typos, etc., would be helpful.
> 

The summary of outstanding issues here is useful but I must admit that I 
for one have long since forgotten the rationale for making these 
changes, and there doesn't seem to be any prose added to the Guidelines 
to explain them

Could you perhaps remind us why we needed to make this change and how 
exactly it is meant to be used?

> 
> * Rather than create a new module for ISO dating attributes, I stuck
>   'em directly into the namesdates module.

I don't think see them there, unless I'm missing something. There's a 
specList which mentions the attributes' names, but no text discussing 
them, and the actual classSpec is in ST, where it probably belongs.

However, since I've now handed the ND chapter over to Arianna for 
review, I expect she will comment on this too.

> 
> * We never decided on the names of the ISO attributes. I called them
>     value-iso=
>     dur-iso=
>     notBefore-iso=
>   etc. My thinking (what little there was) was that it would be nice
>   

sounds plausible to me

if these attributes sorted near their "simple" W3C counterparts.
> 
> * We should perhaps discuss how strong the warnings against using
>   "24:00" and basic formats (i.e., no colons or hyphens in some
>   cases) should be. Currently the remarks say
> 
>     <p>For all representations for which ISO 8601 describes both a
>     <term>basic</term> and an <term>extended</term> format, these
>     Guidelines recommend use of the extended format.</p>
>     <p>While ISO 8601 permits the use of both <code>00:00</code> and
>     <code>24:00</code> to represent midnight, these Guidelines
>     strongly recommend against the use of <code>24:00</code>. 


I think "recommend" is enough: it's not so big a deal surely?

The notion of "recommended practice" is something we explicitly talk 
about in the context of conformance, so having a quick checklist of 
occasions, like this, where we *do* make a recommendation would be very 
useful.



More information about the tei-council mailing list