- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Tue, 26 Jan 2010 18:52:18 +0100
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- CC: Shelley Powers <shelley.just@gmail.com>, HTMLWG WG <public-html@w3.org>
Tab Atkins Jr. wrote: > On Tue, Jan 26, 2010 at 11:41 AM, Julian Reschke <julian.reschke@gmx.de> wrote: >> Tab Atkins Jr. wrote: >>> If you omit the charset, I think it will attempt to default to >>> windows-1252. At least, it apparently will with text/html. Not sure >>> ... >> I don't think this is true. As far as I can tell, the default is impl/local >> specific. > > True, but having the US locale and possibly others default to > windows-1252 is still a bad thing. It's an annoying legacy constraint > that causes problems every time I forget to write the charset meta in > my pages. > >> Anyway, adding a BOM should be sufficient for triggering UTF-8. > > Adding a BOM to the data url? Why would I do that? That's less bytes > than the charset declaration, but more difficult to remember and > enormously more arcane. Why would you care when it's automatically generated? >>> what happens with text/html-sandboxed. >>> >>> The DOCTYPE is required or else the page will be in quirks mode. >> Well, that's something we could change for text/html-sandboxed. Also, we >> could allow fragments of HTML. (Or even require them?) > > Indeed, that seems like it could be a good solution. Indeed. Fine-tuning the behaviour for the new MIME type for the use cases it should cover sounds like a good plan. Best regards, Julian
Received on Tuesday, 26 January 2010 17:52:58 UTC