- From: Richard Ishida <ishida@w3.org>
- Date: Tue, 4 Dec 2007 12:00:25 -0000
- To: "'Lieske, Christian'" <christian.lieske@sap.com>
- Cc: <public-i18n-its@w3.org>
Commenting on BP 19, Christian said: > -----Original Message----- > From: Lieske, Christian [mailto:christian.lieske@sap.com] > Sent: 04 December 2007 07:19 > To: Richard Ishida > Cc: public-i18n-its@w3.org > Subject: RE: Accepting changes to BP doc - BP 19 > ... > Quote > Assign a unique identifier to elements with > translatable content. > > >From my understanding, we might have the following special use case: > inlines/elements within text/placeables. It is usally > recommended to assign identifiers to them in order to allow > them to be handled properly during translation. > > Quote > [Ed. note: How are they to know where this will be > useful for localization? ] > > I guess one consideration is related to inlines/elements > within text/placeables (see above): message-like expressions > with placeholders are very likely to undergo a kind > reordering (cf. "There are <x> file(s) on <x>.") during translation. I think this is something we ought to discuss a little. My inclination is to say that anything that counts for segmentation (see BP6) should have an id. This will include embedded quotations etc. Also, variables and embedded text in composite messages should have ids, as Christian says. I don't think inline/phrasal elements like <em> need ids, if used properly. RI
Received on Tuesday, 4 December 2007 11:57:38 UTC