On Wed, Sep 10, 2014 at 2:09 AM, Henri Sivonen <hsivo...@hsivonen.fi> wrote:
> On Tue, Sep 9, 2014 at 8:13 PM, Eric Rescorla <e...@rtfm.com> wrote: > > Sure, I think there are some reasonable cases. Say that a site asks to > > take your picture for the purpose of displaying an avatar. So you give it > > temporary access, it takes the picture, and then it relinquishes access. > > Because there are UI indicators that show when the camera is being > > accessed, and I control what's in front of the camera, and the site > > is planning to publish the picture I don't really need to trust the site > > much. > > Oh. I had thought that that use case was outside the scope of gUM and > in scope for http://www.w3.org/TR/html-media-capture/ (possibly to be > implemented on top of gUM + https://w3c.github.io/mediacapture-image/ > by the Firefox OS browser chrome.) > My understanding is that html-media-capture is basically being abandoned in favor of gUM. > (Of course, avatars imply login and login *should* imply https...) I don't disagree with this in principle, but I think we have to admit that a huge number of sites which do some form of persistent identity/login and don't use HTTPS. In any case, there are plenty of other legit applications such as photobooths, voice control APIs, etc. > >> > Don't look at me. My assessment is that this isn't superb, but it's > not > >> > a > >> > hill worth dying on. > >> > >> Is gUM already in a "hill worth dying on" stage? With Presto out of > >> the picture, the implementations are just Gecko and Blink, right? And > >> both Gecko and Blink still have gUM vendor prefixed. (Which, if you > >> believe how prefixing is supposed to work, which I don't believe but > >> am willing to pretend in this case, should mean that we can still > >> change stuff.) And at least now if not at the time of first shipping > >> gUM, there's will among Chrome folks to restrict stuff to > >> authenticated origins. > > > > > > I'm not sure I understand the question. There certainly is fairly wide > gUM > > usage. > > I mean: Does gUM have high non-demo, non-test usage on http origins? I don't know. To some extent people are being driven to use HTTPS because they want persistent permissions. > > That obviously doesn't mean things can't be > > changed, but I think you'd need WG consensus to do so, and I don't have > > the sense we're going to get that. > > Per the reply Anne got, > http://lists.w3.org/Archives/Public/public-media-capture/2014Sep/0030.html > , it looks like the WG is deferring the decision to implementors as a > "MAY", so it seems to be more an issue of Gecko+Blink consensus than > WG consensus. Sure. However, my sense is that the Chrome people aren't in favor of this change. > >> seems that Chrome doesn't offer have persistent grants in either case. > >> https://hsivonen.fi/gum-test/ > >> http://hsivonen.com/test/moz/gum-test.html > > > > Chrome auto-decides whether the grant is persistent based on whether > > the URL is http or https. > > Whoa. That's non-obvious and creepy. As a user, I find it creepy for > an UI that looks like a one-time grant to actually do a persistent > grant. (Consider the non-http use case about taking a picture for a > random site you gave above, that use case happening on an https origin > and ending up leaving the origin with a persistent permission to turn > the camera on later.) As a site operator, I find that this makes me > appear creepier than I want to appear if I host (I currently don't but > I want to eventually) SimpleWebRTC app on my site for ad hoc video > conferencing without having to trust a third party to do the call > setup in a way that doesn't involve extra parties and I don't want to > ask the other party of the conversation to have to trust my site not > to turn on the camera later after the conversation. I sympathise with this position, which is why Firefox doesn't do it this way. -Ekr _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform