W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2015

Re: plain/simple/easy language variant subtag

From: Chaals McCathie Nevile <chaals@yandex-team.ru>
Date: Tue, 15 Sep 2015 01:14:22 +0200
Cc: "IG - WAI Interest Group List list" <w3c-wai-ig@w3.org>
To: "Tobias Bengfort" <tobias.bengfort@posteo.de>, "Gregg Vanderheiden" <gregg@raisingthefloor.org>
Message-ID: <op.x4y6t8dns7agh9@widsith.local>
On Mon, 14 Sep 2015 22:02:19 +0200, Gregg Vanderheiden
<gregg@raisingthefloor.org> wrote:

> I presume the simple web page would be at a different URL than the other  
> — so having the simple >language tag would not help those who landed on  
> the original page.  Or are you thinking of another >arrangement.

Using HTML like

<link rel="alternate" href="plainer.html" hreflang="en-cerfA2">

would enable simple browser customisation to switch to the simpler version.

> MOSTLY I have seen language tags used to identify the language not the  
> language level.   And it is >used for accurate presentation or  
> pronunciation.  What did you see as the purpose of the tag for >language  
> level?

"complex language" and "a different language" are really the same thing.

A simple version of english like globish can be something that is always
also english - but not all english is always globish.

In other cases, the simple language can have some rules that are not true
in the "more complicated" version.

> What I have seen done  with simple language pages - is to either
> 1. provide a link off of the usual landing page  - to the simpler  
> version of the page.

If you have a language tag for the simple language, then you can make the  
link automatically work for people. Or make it easy to find in the  
browser, instead of having to look for it in each, where people often have  
a different place to put it, and a different way to explain what it is.

> It is a tough problem - including because (after you get rid of any  
> unnecessary complexity or complex language on the page) making the page  
> or the language simpler usually involves losing some information.

In my experience, information does not need to be lost. Only literature,  
drama, poetry, usually have information that cannot be separated from the  
words and style used.

Instead, in english simple text is often longer. So having the "not  
simple" version helps some people read it faster. This is not true for all  
languages - often I find making spanish simpler doesn't make it longer.

> So it is desirable to simplify it in stages - so that people of all
> levels can understand it — but those who can understand more lose less  
> information.

That can help. In practice it is very expensive. If people do not know  
*how well* they can read, and pick the wrong level, then it does not help.

> There is no language that is plain enough for all.

True enough. But we can make a lot of text better, really helping people,  
even if we can't be perfect.

> What I really wish for is a translation program that can translate  
> between language levels WITHIN a language - rather than between them -  
> so that everyone can get things explained to them in a form they can  
> individually understand. So a tool that can start high and then be  
> turned down until someone can understand — and then turned back up so  
> they can understand more as they move up (with supports to help them)   
> would give each person maximal access -  and the ability to learn more  
> by getting the basic idea and then having it elaborated - if and as they  
> wish.  But I fear it will be awhile before we have that.

There is no reason we cannot do this with existing technology. But  
automated translation of semantics is still very unreliable. Reading an  
automatically translated document is easy enough for people who have high  
reading skills, and can spot errors made by the machine and fix them in  
their heads.

The best tool for good simple expression is a skilful editor. It is very  
hard to do, but not impossible. Some of those people make a lot of money  
with their skills.

Maybe that is why so often we don't pay much attention to the problem :(



Charles McCathie Nevile - web standards - CTO Office, Yandex
    chaals@yandex-team.ru - - - Find more at http://yandex.com
Received on Monday, 14 September 2015 23:14:53 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:57 UTC