- From: Chaals Nevile <chaals@yandex.ru>
- Date: Thu, 06 Sep 2018 23:56:25 +0200
- To: w3c-wai-ig@w3.org
On Thu, 06 Sep 2018 22:47:42 +0200, David Woolley <forums@david-woolley.me.uk> wrote: > On 06/09/18 14:26, Katie Haritos-Shea wrote: >> I thought <b> and <i> originally came out of SGML.....before HTML >> > SGML has no pre-defined tags. It is the <tag>....</tag> construct, not > any particular value of tag, that comes from SGML (although the original > versions of HTML complied with SGML more fully than that). I believe that a number of very old HTML tags (I know this is the case for the h1-h6 set) were essentially re-using an SGML vocabulary that happened to be familiar to many people who might have been the "first users" of HTML, in CERN. The reason for adopting them was to ease transition to HTML (which was SGML-like, but at the very beginning didn't actually have the formal structure SGML did - that came later on, along with the development of XML as a formal replacement for SGML). 'strong' and 'em' - and lots of other things, some good ideas and some probably not - were introduced in HTML 4 - which was SGML based, attempted to have robust semantics that allowed for things like device independence and decent accessibility support, and so on. Early work on HTML5 also tried to assign more semantics to things that hadn't had them before as well as add another new set of semantic elements. Some of that took hold, some of it didn't. For what it's worth, "semantics" - understanding meaning and how people use it - is really really hard. So it isn't surprising that quite often efforts to make it work more like someone thinks it should don't get global traction and success. cheers -- Chaals: Charles (McCathie) Nevile find more at https://yandex.com Using Opera's long-abandoned mail client: http://www.opera.com/mail/ Is there really still nothing better?
Received on Thursday, 6 September 2018 21:56:49 UTC