[tei-council] sluice your glosses

[Lou Burnard]

On 28/02/13 10:23, James Cummings wrote:
> On 28/02/13 01:24, Kevin Hawkins wrote:
>> Since we go around telling people that you shouldn't read
>> element names literally -- for example, that a <monogr> can
>> include things that you don't think of as a "monograph" and
>> that <msDesc> can be used to describe early printed books, not
>> just handwritten manuscripts

I think these are two rather exceptional cases : for the most part tag 
names (once you know what they are) *should* be used for the things 
their names suggest. Even in these cases, the fact that they can also be 
used for other things doesn't invalidate that.

-- it seems to me that doing away
>> with <gloss> in element specs, as Lou and Sebastian suggest,
>> would be in line with what we say.
>
> Yes, I'd agree with this. Retain the element (and continue to
> allow it at that point for those existing ODDs that use it) but
> remove it from all official TEI use as an example. Move any
> existing glosses that would be helpful (e.g. 'sic' but with more
> explanation) to a <remark> or <desc> with more explanation.
>
> If no one objects strongly to this in the next couple days, can I
> have a volunteer to implement it?

I have no idea what "all official TEI use" means as distinct from the 
"existing ODDs" (are our existing ODDs unofficial in some way?), but I 
do think this is a somewhat draconian and misguided reaction to the 
problem I was raising.

The intended use for the things currently tagged as <gloss> is to 
indicate an expansion of a tagname which is otherwise meaningless or 
ambiguous to a native speaker. Examples include <lb> and <p>. This is 
very useful! Lots of people like using short tags (cf the dictionary 
chapter) even though we try hard to make tag names complete words and it 
doesn't cramp their style too much to require them also to add an 
expansion. It would be a highly retrograde step to remove them, though I 
don't think that's what James is proposing, but it would also be a shame 
  to preclude their use in future short tags <imho/> (which I think he 
is proposing).

These expansions are also useful (if properly recognised for what they 
are) to the non-native speaker, though sometimes it is necessary to add 
further explanation of why that (expanded) term is used or to translate 
it. As we're all non-native speakers of Latin (I assume) this is why 
<sic> is an interesting case. <eg> might be another, if it wasn't for 
the fact that Molesworth has made this effectively an English word -- no 
one is proposing that we add a gloss of "exemplia gratia" rather than 
the current "example" (I hope); the problem with <sic> is that people 
apparently don't understand what it means, even though it's not an 
abbreviation.

We should recognise that <gloss> is actually being used as a synonym for 
<expan> when it appears in an <elemntSpec> or similar; we shouldn't use 
it to provide discursive or etymological explanations of why the thing 
has that name. If that's necessary, provide it in the <remarks>. See 
<fw> for example -- NOBODY knows why it's called "forme work", even 
native speakers, so we should add an explanation there.

Alternatively, we provide a different tag for the "etymological 
discussion" of a tag rethink <gloss> with more structure within it, e.g. 
<tagEtym><expan>forme work</expan><gloss>from the printer's term for 
material common to all formes (i.e the block in which type was 
set)</gloss></tagEtym>

This has the advantage of addressing a frequent reproach made by 
non-native speakers, viz that even after the tagname has been expanded 
they still don't know what the deuce it means. And the disadvantage of 
giving everyone's favourite ODD editor even more white hairs.

>
>
>> Lou said in his message "Otherwise we'll never finish" [if we
>> don't do this], though I'm not sure what he means.
>
> I think he means that we'd talk about it forever and not make a
> decision.
>

I might well have meant that, but actually I meant that providing 
reasons why any tag has the name it has is a rather open ended task.