- From: Julian Reschke <julian.reschke@gmx.de>
- Date: Mon, 19 Jul 2010 19:57:10 +0200
- To: Boris Zbarsky <bzbarsky@MIT.EDU>
- CC: public-html@w3.org
On 19.07.2010 19:50, Boris Zbarsky wrote: > On 7/19/10 1:42 PM, Julian Reschke wrote: >> So there's a foreseeable race-to-the-bottom here. Please, this time, do >> not blame the server implementers. > > Er.... there's a commonly-deployed server that sends > application/octet-stream when it should send no type at all. How is this > different from the text/plain situation we used to have? It's different that this time we're introducing a new language feature, that we are aware of the problem, and have all UA developers in the same forum. Let's try to do this right this time: - do not sniff - do not require to sniff - make sure that the media types are actually registered > I agree that if all UAs were to hold the line on this we could manage. > But the server behavior (and the fact that some UAs have already crossed > the line) makes that seem pretty unlikely, no matter what I may happen > to think about sniffing. :( OK, so who crossed the line? Do they sniff for application/octet-stream, or for other types as well? (text/plain comes to mind). Best regards, Julian
Received on Monday, 19 July 2010 17:57:52 UTC