[Prev][Next][Index][Thread]

Re: URLs/MIME only?



On Tue, 31 Dec 1996 12:13:43 -0500 Gavin Nicol said, quoting Eliot
Kimber:
>>The only problem with this approach is the same one HyTime and DSSSL faced:
>>what objects make up the nodes of the tree you're addressing.  We currently
>>have three different tree views of SGML documents in established specs:
>>
>>1. DSSSL default grove plan: characters and elements are children of elements
>>2. HyTime default grove plan: pelements and elements are children of elements
>>3. TEI implicit grove plan: elements only are children of elements

I'm confused here.  It has been said several times in the SGML WG
discussion (by whom, I cannot currently recall, but in this instance by
Eliot Kimber) that in TEI extended pointers, 'pseudo-elements' are not
children of elements, only subelements are children.  But the CHILD
keyword of TEI extended pointers refers to elements or contiguous spans
of untagged character data, and the same is true of DESCENDANT,
PREVIOUS, NEXT, PRECEDING, and FOLLOWING.  In the data

  <p ID='P1'>Aaa.<q>Qqq.</q> Bbb.<note>Nnn.</note> Ccc. <q>Q2q2.</q></p>

the TEI location latter ID (P1) CHILD (3) refers to the string ' Bbb.'

If left to my own devices, I would have assumed that 'pseudoelements'
was a term intended to refer to things like the strings 'Aaa. ', '
Bbb.', and ' Ccc. ' in the example.  Is that not what the word means?

If it is, then let's correct this repeated error before it gives rise to
a bad misconception (and if there is some passage in the Guidelines
which suggests another interpretation, please give me a reference, so we
can fix the inconsistency).  If it's not, then please, what *are*
pseudo-elements?

In any case, my reading of chapter 14 (esp. sec. 14.2.2.7) makes me
think it's misleading to say without qualification "elements only are
children of elements" in the TEI's extended pointers.


-Michael Sperberg-McQueen