From NATCORP@vax.oxford.ac.uk Tue Apr 2 18:20:45 1991 Received: from [129.12.1.5] by dkuug.dk via EUnet with SMTP (5.64+/8+bit/IDA-1.2.8) id AA03377; Tue, 2 Apr 91 18:20:45 +0200 Message-Id: <9104021620.AA03377@dkuug.dk> Received: from vax.oxford.ac.uk by kestrel.Ukc.AC.UK via Janet (UKC CAMEL FTP) id aa23999; 2 Apr 91 17:07 BST Date: Tue, 2 Apr 91 17:00 GMT From: Dominic Dunlop To: KELD , I18N Subject: ISO 8879 -- SGML -- and its relavence to internationalization X-Charset: ASCII X-Char-Esc: 29 Keld and others, Firstly, here I am at a new address in a Real Job (as opposed to being a self-employed person). But don't use this new address unless and until I tell you to. Continue to use domo@tsa.co.uk. Secondly, this job involved the amassing of 100 million words of English- language text, ``marked up'' so that computer-aided linguists can chew it over. (Previously, too little has been readily available on-line for them to get their teeth into.) The marking-up used SGML (Standard Generalized Mark-up Language), as defined in ISO 8879:1986, as amended in 1988. This document makes interesting reading, even if it is of the type that requires that you have drunk no beer before picking it up, and which demands that you need a beer after putting it down. The standard, although basically American in origin (the work started inside IBM rather more than ten years ogo) addresses a considerable number of internationalization issues, among them coded character sets, repertoires and character names. (Yes, it does name characters; yes, the names are in lower-case.) Before the Netherlands meeting of WG15RIN, I shall summarize the internationalization-related content of ISO 8879, as I consider that the information will be useful to the group from two points of view: 1. So that we can see the problems encountered and solutions arrived at by another group working in the field of internationalization; and 2. So that we can avoid gratuitously contradicting the content of an existing ISO standard.