- From: Thomas Broyer <t.broyer@gmail.com>
- Date: Wed, 30 May 2007 15:51:39 +0200
- To: "Dave Child" <dave@ilovejackdaniels.com>, public-html@w3.org
2007/5/30, Dave Child: > > Presumably, user agents will render future versions of HTML depending upon > the page's DTD. Unless UAs started assuming documents were written in a > new version of HTML despite a missing DTD indicating this, I don't see why > existing form content would be broken. Is the issue the authors themselves? > > Would somebody please be so kind as to explain what I'm missing here (or > to point me to relevant emails in the archive)? The point is that, as actually designed, HTML5 define *one* way of processing HTML that is compatible with what browsers do today. That way, any new comer could write a browser, with only *one* parsing/processing implementation, that will be "compatible" with most (if not all) olde web content. So if you change how form elements should be presented by default then, as soon as you do not use a browser with background and history and quirks modes et al., older web content will appear "broken" in that new browser (and you'd probably accuse the browser itself to be broken, while actually it would be the HTML5 spec) Also, HTML5 has some support yet in many browsers (mainly Safari and Opera, also Firefox), based on their current "standards mode". Changing how form elements should be presented by default would be asking those browser vendors to add yet another "standards mode" dedicated at HTML5. Finally, HTML5 defines how to work with horribly broken documents. People making those documents probably are not aware of HTML5. The chances that they start their document with <!DOCTYPE html> to "turn on" the "HTML5 conformant mode" in browsers are very small. And if they add it but "test" their pages in non HTML5-compliant browsers, those pages will probably appear as "broken" in HTML5-compliant browsers. HTML5, by design, tries to reduce browsers incompatibilities by standardizing their current behaviors. The goal is not to have "yet another standards mode". My 2 c€nts -- Thomas Broyer
Received on Wednesday, 30 May 2007 13:51:44 UTC