Hi Syd, On 15-06-01 07:16 AM, Syd Bauman wrote:
Martin, is this some sort of trick? That page only has 2 instances of <defaultVal> listed, but there are roughly 65 of them.
I thought I was tasked with analyzing each one and suggesting an action for it. If not, then I'm inclined to let you proceed with the ticket the way you want to, since you're the owner.
I most certainly do *not* want to move forward with these one at a time. I want to see and analyze a list of all of them. My bet is that, upon seeing said list, I am going to push for leaving <defaultVal> in, but changing its semantics so that the default value is not actually tucked into DTDs and schemas, but rather <defaultVal> is a hint for your TEI processor.
That seems like a completely new idea to me, not related to my original ticket. But it seems to me that if you're wanting to provide hints to a processor, the place to do that is in the PM, not in the schema specification.
After we've changed those semantics (which will take a year or whatever), I would then be happy to see <defaultVal> all over the place.
I wouldn't. I think it's magic however you spin it, and it should go. But if you develop your proposal, Council can look at it in detail. Meanwhile, I'll stop work on individual cases.
And whether we should split the meanings of part=N into two separate values or not is a somewhat different question. (One which I would support wholeheartedly. That said, the current situation is not nearly so crazy as you make it out to be -- part=N indicates to your hunter/gatherer software which you use to reconstitute partial elements into aggregate elements that it does not have to pay attention to this particular element instance. It does not assert *why* your software can ignore it, which is a pity. But it's not useless in a default attr value environment.)
Once again we're embedding processing hints into encoding, then. I don't like that at all, especially when we're not saying "by the way, this attribute is like this because people writing processors might need it."
P.S. And of course the default value of org= is "uniform" ... how many times have you even considered specifying org=composite?
My point is that "uniform" is also a claim, and we shouldn't be having people make claims that they don't know about. Maybe I don't care whether my divs are composite or uniform -- why is a claim that they're uniform foisted on me without my knowledge? As we keep saying, it's magic and it's misleading. Cheers, Martin
OK, page is up on the wiki:
http://wiki.tei-c.org/index.php/DefaultVal_candidates_for_removal
I'd like to move forward with these steadily, though, otherwise there'll just be a long page left on the wiki when I leave Council and nothing will be done, so I'd like ask for any objections to the first two actually to be recorded on the wiki.