This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 17954 - Attribute to indicate that the grammar should be translated (e.g. apply declensions), but that the base of the words should remain in the original language
Summary: Attribute to indicate that the grammar should be translated (e.g. apply decle...
Status: RESOLVED WONTFIX
Alias: None
Product: WHATWG
Classification: Unclassified
Component: HTML (show other bugs)
Version: unspecified
Hardware: Other other
: P3 enhancement
Target Milestone: Needs Impl Interest
Assignee: Ian 'Hixie' Hickson
QA Contact: contributor
URL: http://www.whatwg.org/specs/web-apps/...
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-07-18 07:22 UTC by contributor
Modified: 2016-03-16 18:03 UTC (History)
5 users (show)

See Also:

Attachments

Description contributor 2012-07-18 07:22:56 UTC 
This was was cloned from bug 17456 as part of operation convergence.
Originally filed: 2012-06-09 23:44:00 +0000

================================================================================
 #0   contributor@whatwg.org                          2012-06-09 23:44:07 +0000 
--------------------------------------------------------------------------------
Specification: http://dev.w3.org/html5/spec/spec.html
Multipage: http://www.whatwg.org/C#top
Complete: http://www.whatwg.org/c#top

Comment:
Regarding Section "The translate attribute":

If I am writing prose of fiction, and I write a name I don't want the
translation mechanism to translate, then I understand I would set the
attribute to "no" on that particular word.

But. When I write (for example) in Polish, then a declension could be applied
to the word. The problem is, I'd like to instruct the software that when a
translation (for example) to English is applied, the software should not
translate the word, but only change its grammatical form to match that of the
target language.

For example, I am using the word "realm" in the meaning of "space" AND
"kingdom", so if it's translated, then the double meaning will be lost.

Since the specification says that "no" means to "leave it unchanged", what I
understand means to not apply any transformation at all to the text, I cannot
possibly instruct the software to act the way I'd want it to.

This could be solved by a new text-level element, since it only applies to a
text, or a boolean attribute on the span element.

I understand it's an obscure problem, though.

Posted from: 2001:470:71:12d:211:5bff:fe35:9864
User agent: Mozilla/5.0 (X11; Linux i686) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/21.0.1145.0 Safari/537.1
================================================================================
Comment 1 Ian 'Hixie' Hickson 2012-09-25 04:29:36 UTC 
Can machine-translating software even make that distinction right now?

I think the right solution here is to not do anything for now, but at some point introduce a new keyword on this attribute that means "just translate the grammar". So I'm going to mark this LATER for now and keep a note in the spec.
Comment 2 contributor 2012-09-25 04:30:22 UTC 
Checked in as WHATWG revision r7403.
Check-in comment: Note for future features.
http://html5.org/tools/web-apps-tracker?from=7402&to=7403
Comment 3 Ian 'Hixie' Hickson 2013-03-09 00:13:11 UTC 
I'm punting on this until a translation service implementor indicates that they would be capable of using this.
Comment 4 Anne 2016-03-16 18:03:38 UTC 
Resolving WONTFIX given the lack of such interest.