- From: Karl Dubost <karl@la-grange.net>
- Date: Sat, 6 Apr 2013 08:56:37 -0400
- To: Larry Masinter <masinter@adobe.com>
- Cc: Anne van Kesteren <annevk@annevk.nl>, "www-tag.w3.org" <www-tag@w3.org>
Le 5 avr. 2013 à 18:40, Larry Masinter a écrit : > that is, all four of the conditions have to hold for sniffing to > be a good idea to propagate. There is a nuance in between "being a good idea" and "being necessary " or "being the most effective for the users". > While that might apply to hyperlinks and transclusion FROM > HTML documents, I don't think it applies to network APIs, > applications other than browsers, (SIP? Instant messaging? > hyperlinks from content other than HTML?). For example in the feed world [1] "In Stupid HTTP Tricks": * 6. Sending back things that are not XML (eg, videos). It can help to check Content-Type and Content-Length headers, but sometimes they misidentify RSS as something else (eg, text/plain). My main issue so far with the document "Authoritative Metadata" [2] is that it is not about "HTTP Content-Type". Check all the "Good Pratice" and "Constraint" lines and there is no reference to Content-Type. "Content-Type" is an example in the document. So it doesn't seem a good idea to focus on this. Suggestion: With regards to all "Good Practice" statements, it might be better (before starting to say what is wrong) * to list the circumstances when this good practice breaks * how, once deployed in a social environment, is it broken * what are the difficulties for fixing them * what/who consumes the data [1]: http://inessential.com/2013/03/18/brians_stupid_feed_tricks [2] http://www.w3.org/2001/tag/doc/mime-respect-20130405.html -- Karl Dubost http://www.la-grange.net/karl/
Received on Saturday, 6 April 2013 12:56:46 UTC