- From: Sam Ruby <rubys@intertwingly.net>
- Date: Mon, 18 Feb 2013 12:53:59 -0500
- To: public-html@w3.org
On 02/18/2013 12:34 PM, David Carlisle wrote: > On 13/02/2013 22:36, Eric J. Bowman wrote: >> Apparently I need to make this point, again: If there was no >> interest in polyglot, there would be no HTML parser in libxml2; its >> presence, and widespread use if xsl-list is any indication, >> indicates otherwise. > > I actually think pologlot spec is worth having (and probably I made as > many bugzilla comments on it as anyone). I still have some issues with > the wording but as a general idea I think it's fine... > > But I don't understand your comment there at all. The HTML parser in > libxml2 (or tagsoup or Henri's validator.nu parsers in java) are exactly > the reason that some people (not entirely unreasonably) say that it > isn't needed. If you can put an HTML parser in front of an XML > tool-chain then you can pull in unrestricted HTML syntax and you have no > need to produce HTML documents following the polyglot guidelines which > are designed to allow an HTML document to be fed to the tool-chain via > an XML parser. I disagree. Boy I wish all "HTML parsers" supported unrestricted HTML syntax. Henri's parser is better than tagsoup which is much better than libxml2. Heck, even Google's parser is buggy: http://lists.w3.org/Archives/Public/www-archive/2013Feb/0059.html At least in that case, the HTML spec declares such a page as invalid, and thereby attempts to encourage interoperability even in the face of imperfect implementations. But that isn't the case for other interop problems. Until that is addressed: http://intertwingly.net/blog/2012/11/09/In-defence-of-Polyglot > David - Sam Ruby
Received on Monday, 18 February 2013 17:54:29 UTC