[tei-council] xml-colon-thing

[David G. Durand]

I'm assuming that I'm still a council member until the end of the 
calendar year...

I don't like this idea at all, nor the one (apparently) discussed at 
the TEI members' meeting. As roughly described to me second-hand, it 
seems to have been a proposal that we should rename elements and 
attributes changed as to function in P5 so that it will be easier to 
extend P4 by the re-addition of p4 elements that have been replaced by 
newer functions/notations in P5.

We shouldn't (and can't under our own rules) prevent people from using 
any tagging practices we want, but I don't think we should "ease 
migration" in a way that allows existing P4 documents to become P5 
documents without updating the markup. There's no shame at staying with 
P4 if that suits one's needs. The decision to allow non-compatible 
changes is a good one in that it enables things to be cleaned up 
without constraint as to compatibility, and that's a good thing, I 
think. Renaming things just to make using the old names easier is bad 
because it makes the new names less useful, and the whole issue of 
support _more_ complicated.

In the XML committee we had a slogan of not being "gratuitously 
incompatible" with SGML, and in that situation the need for backward 
compatibility seemed much stronger. I'd also note that XML itself could 
have been more incompatible without serious problems, as things turned 
out.

I don't really see how it helps in moving legacy data: for xml:id, you 
have to add a '#'. For xml:lang you have to make up a table converting 
whatever strings you used in your TEI "lang" to the standard strings as 
defined by the ISO and W3C.

P5 is going to have to go through a process of acceptance and tool 
building (as previous versions have) and the center of gravity will 
slowly shift to the newer version, as people so the work of conversion.

People will have to compare the effort of touching old data and 
shifting their tool environment, versus staying with the old markup and 
foregoing the improvements of P5, or the duplicated effort if they 
maintain data in both formats and thus have to deal with two separate 
tool chains. Granted this can be a hard decision to make, and will 
certainly involve some extra work if one wants to use new P5 features 
or tools, but it seems to me that that's the reality of this kind of 
work.

I'd be more sympathetic if I thought that people _had_ to move to the 
new world, but they can linger on the margins of P4 as long as they 
want. They can even add stuff from P5 into P4 if they want to.

If we're to make a break (and we've decided that long ago), I say that 
it should be a clean break, with no guilty half-measures or apologies.

   -- David

On Nov 11, 2004, at 12:15 PM, Lou Burnard wrote:

> The Character set workgroup has proposed and the Council has (I think) 
> accepted that  the global lang attribute should be replaced by a 
> global xml:lang attribute; likewise the Standoff WG has proposed and 
> the Council is (I think) still ruminating about the notion that the 
> global "id" attribute should be replaced by a global xml:id attribute.
>
> Pondering how to actually implement these proposals in P5 without 
> causing too much havoc, I have reached the conclusion that it would be 
> both feasible and advisable to make the transition in a rather more 
> cautious way. I am proposing to add xml:lang as an *alternative* to 
> lang in the definition for  tei.global.attributes rather than as a 
> replacement for it.  (And similarly xml:lang). By "alternative" here I 
> mean that a given element can specify one or the other but not both 
> (supplying both would in any case be impossible since both have ID 
> declared values). The current ODD syntax already supports this -- we 
> added it art Christian's suggestion during the Paris Meta meeting -- 
> and indeed uses it for similar backwards- compatibility-reasons in the 
> definition for <figure> I think. The schemas generated do exactly the 
> same thing: the DTD generated may need some further hand adjustment.
>
> I think it is a good idea to hedge our bets about the take up of W3C 
> recommendations in this way, and it will definitely make the job of 
> moving legacy data to P5 a whole lot easier. Not least for us!
>
> Comments?
>
> Lou
>
> _______________________________________________
> tei-council mailing list
> tei-council at lists.village.Virginia.EDU
> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council
>
>

David G. Durand
Director, Electronic Publishing Services
Ingenta Inc.
111R Chestnut St.
Providence, RI  02903 USA

T: +1 401-331-2014 x111
T: +1 401-935-5317 Mobile
E: david.durand at ingenta.com