W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > October to December 1998

Re: Techniques suggestions

From: Masayasu Ishikawa <mimasa@w3.org>
Date: Sun, 15 Nov 1998 01:19:39 +0900
To: w3c-wai-ua@w3.org
Message-Id: <19981115011939Q.mimasa@w3.org>
Al Gilman <asgilman@access.digex.net> wrote:

> But I am still concerned on a slightly different, related issue:
> What do the guidelines say a User Agent should _do_ with the LANG
> information?  What is meant by "implement LANG per HTML 4.0" in
> the present context?  It seems to me that LANG is purely
> informative as defined by HTML 4.0.  There is no defined behavior
> associated with it.  Or is there?

Well, the intent of the LANG attribute is to allow user agents to
render content more meaningfully based on accepted cultural practice
for a given language [1].  For example, Q element is designed to be
rendered in a language-sensitive manner [2].  In this case, LANG is
important for proper rendering.

I know several user agents can render CJK unified ideographs with
different glyphs depending on the LANG information.  But in many cases,
you're right, LANG is informative for visual rendering.

However, I believe LANG is more important for non-visual rendering.
For example, visual user agents can render the word "Stern" even if
they don't know its language, but for non-visual user agents, such as
voice browsers, it is important to know whether it's in English or in
German (or whatever) to render it in meaningful way.

BTW, in CSS2, there is a pseudo-class ":lang", which is specially
designed for language information.  With this selector and ACSS,
it is possible to speak different languages with different voice
family, when LANG is provided.  That can be another reason why
LANG should be implemented.

[1] http://www.w3.org/TR/REC-html40/struct/dirlang.html#h-8.1
[2] http://www.w3.org/TR/REC-html40/struct/text.html#h-9.2.2
[3] http://www.w3.org/TR/REC-CSS2/selector.html#lang

Masayasu Ishikawa / mimasa@w3.org
W3C - World Wide Web Consortium
Received on Saturday, 14 November 1998 11:19:43 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:49:21 UTC