Consider switching from xhtml to html5
Issue #207
open
Originally reported on Google Code with ID 207
Are there any advantages in using xhtml for PLA's output?
Switching to simple html (keeping the same xml serialization) would allow us to use
new features that already enjoy wide browser support.
A good example would be forms: we could use 'autofocus' and 'required' for input field
to improve* user interaction, as "progressive enhancement" in browsers that support
them. Also, remove default action attributes (Issue #172).
Our current output is already 99% valid html, except for a few 'javascript:void' and
table styling attributes (border, cellpadding, cellspacing, valign, width) that can
be replaced by css.
Reported by dreadnaut on 2013-04-04 09:42:26
Comments (6)
-
-
reporter
| I guess we are talking about html5? html5 is the buzzword, 'html' is the language. We would not be switching to "html5" because we cannot introduce tags which are not compatible with older browsers* [without js hacks.] What we can do is switch doctype to html, which gives us standard rendering in all browsers (IE6 includes). Standard mode means using the latest html and css implementation supported by the browser, be it 4,5,6,n. Again, the version number is actually not important. If the browser supports a feature, it will use it; if it does not, it will ignore it —contrary to xhtml, where stuff can break (e.g., IE vs application/xml+xhtml) [*] http://html5doctor.com/how-to-get-html5-working-in-ie-and-firefox-2/ | Although I don't see a huge advantage currently. Agreed, but we'll switch "someday" anyway, and unless there's any good reason to stay on xhtml, I could finally add autofocus to the password field in the login screen! I'll try in the next days maybe, and check whether it breaks every single theme :-pReported by
dreadnauton 2013-04-04 11:09:16 - Status changed:Accepted -
| html5 is the buzzword, 'html' is the language. Well, yes. But switching to html < 5 would simply be a step backwards. It would allow us to use <font>, omit <body> and stuff like this. HTML4 days are gone. Using HTML5 without the new features is not the same as using html4: not everything that was valid html4 is valid html5 (although it "works" and does not break). | We would not be switching to "html5" because we cannot introduce tags which are not | compatible with older browsers* [without js hacks.] Switching to html5 doesn't necessarily mean we have to use all (or even any) of the new features. | Again, the version number is actually not important. Which would mean we could also use html3!? The difference that xhtml is more strict does not mean it is worse. In fact I consider this one of the things that made markup used nowadays a lot cleaner - even in html5. | we'll switch "someday" anyway Not necessarily, but probably yes.Reported by
crazy4chrission 2013-04-04 12:31:49 -
reporter
| Well, yes. But switching to html < 5 would simply be a step backwards. I see, we are just approaching from two different sides. I don't want to switch to either html4 or html5, but to `html`, the current version available :) Some of it is part of html5, some of it is not yet part of html5, most of it is compatible with html4. In the end, the doctype is `html`, not `html5` :) [ok, to be fair it's `html` just to trigger standard mode in IE6] XHTML is two separate things: is a different language (that happens to overlap a lot with html) and is the xml serialization of html. The first is becoming obsolete, the second (the stricter rules you mention) are in any case good practice. I'd like to keep the latter of course (still served as text/html, because application/xml+xhtml means looking for trouble). More here: http://wiki.whatwg.org/wiki/HTML_vs._XHTML Anyway, the above is philosophy. I think we both understood the overall idea: - change doctype to <!doctype html> - fix what does not validate - when possible and useful, introduce new html[5] features - ensure backward compatibilityReported by
dreadnauton 2013-04-04 12:56:34 -
| Anyway, the above is philosophy. I think we both understood the overall idea Yes. Just go ahead and prepare a patch if you feel it is already worth it.Reported by
crazy4chrission 2013-04-04 14:59:44 -
reporter
| Yes. Just go ahead and prepare a patch if you feel it is already worth it. Actually, I think I'll wait: we should "batch" css-breaking changes together and go through the alternative themes only once. (the table/th stuff is still pending, etc.)Reported by
dreadnauton 2013-04-04 15:04:52 - Labels added: Priority-Low - Labels removed: Priority-Medium - Log in to comment
Reported by
crazy4chrission 2013-04-04 10:22:13