#[tantek]it's not because none of that is very helpful in practice. better to HEAD request the video to get its content-type. any content-type in markup is advisory at best
#[tantek]as in, no better than file extension hacks
#aaronpkSee also: browsers rendering a jpg fine even if the filename and web server header is .png
#aaronpkWhat microformats *can* be helpful for is if a particular kind of media requires different html markup to render
#sknebelafaik browser do use these tags for source selection, so if information is provided passing it through is likely useful
rattroupe[d], kimberlyhirsh[d], jacky and angelo joined the channel
#[tantek]sknebel, yes it's likely implemented somewhere, but in general it's a bad approach as with all metadata separate (and duplicated) from the data, it tends to rot
#[tantek]aaronpk is correct, even if *served* as one content-type on HTTP, browsers sniff the bytes to actually figure out what kind of image etc. it is
#[tantek]so basically, this avenue of technology development is not worth investing in, propagating
#[tantek]it's only in HTML for legacy reasons and it's too much work for too little reward to remove.
#[tantek]that is, the "type" attribute on img etc.
zack[m]1, barnaby and Darius_Dunlap[d] joined the channel
#aaronpki do think sknebel is right that the browser does use the html attributes for video source stuff for handling multiple quality of the same video and such