- From: Felix Sasaki <fsasaki@w3.org>
- Date: Fri, 05 Jul 2013 16:57:05 +0200
- To: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
- CC: Jirka Kosek <jirka@kosek.cz>, public-multilingualweb-lt-comments@w3.org
Am 05.07.13 16:51, schrieb Daniel Glazman: > On 05/07/13 16:42, Felix Sasaki wrote: > >> Let's assume you raise the objection and we do what you implement with >> CDATA sections / XML comments that you proposed to resolve it. Then test >> files like these >> https://github.com/finnle/ITS-2.0-Testsuite/blob/master/its2.0/inputdata/translate/html/translate7html.html >> >> >> won't work any more. We have several localization service providers in >> the group who work with an XML tool chain and that rely on the current >> representation. Your CDATA approach would put a burden on them, as Jirka >> stated. >> >> So in summary: if we follow your approach we will resolve your formal >> objection and create another one. > > Felix, sorry to play the devil's advocate, but you're telling me > implementors shipping implementations _before_ REC will suffer and > a test must be updated... SHOULD we really care? I am not talking about the burden of updating tests - you got me wrong. I am talking about implementer burden, and Jirka described the burden. The test update is just one part of this. The main part is "burden on HTML editor" vs. "burden on XML tool chain". So please answer my question on a technical level, not in terms of the standardization process. Best, Felix
Received on Friday, 5 July 2013 14:57:35 UTC