- From: Peter Murray-Rust <Peter@ursus.demon.co.uk>
- Date: Tue, 03 Jun 1997 23:43:18 GMT
- To: w3c-sgml-wg@w3.org
In message <v03020901afb8a02c4781@DialupEudora> "B. Tommie Usdin" writes: > It seems to me that much of this discussion can be summarized as a debate > between people saying (in essence): > "XYZ (in this case PEs) can be used badly, so we shouldn't allow them" My understanding was different - that parser implementers were having considerable difficulty in implementing PEs, possibly in part because their implementation was no completely defined from the spec. My concern was that (a) PEs could be seen as sufficiently complex to deter parser-writers from implementing them and (b) that if not precisely defined there could be different opinions in how they were implemented. [Subsequent discussion has suggested that their implementation is moderately straightforward but still requires significant effort.] The claim from the disestablishmentarians (which includes me) is that every little bit added to XML-LANG makes it that bit more difficult to: (a) implement (b) sell to potential customers. For example, I was evangelising XML to non-SGML people today and pointing out that the main features of the language are balanced tags and quoted attributes. This is probably heresy on this list :-), but it made them feel quite relaxed about the language and positive towards it. Later they can be introduced to the more complex constructs. P. -- Peter Murray-Rust, domestic net connection Virtual School of Molecular Sciences http://www.vsms.nottingham.ac.uk/
Received on Tuesday, 3 June 1997 19:24:25 UTC