W3C home > Mailing lists > Public > public-media-capture@w3.org > October 2015

RE: Gutting the screen sharing constraints

From: Keith Griffin (kegriffi) <kegriffi@cisco.com>
Date: Fri, 30 Oct 2015 10:51:01 +0000
To: Martin Thomson <martin.thomson@gmail.com>
CC: "public-media-capture@w3.org" <public-media-capture@w3.org>
Message-ID: <2474bf256acb4401857091f3653e42c1@XCH-ALN-010.cisco.com>
I like that the user selects what to share and the site does not directly ask. This would still allow for common site driven requests e.g. a meeting host/chair asking a participant to share their screen can be more safely achieved by a request from the site resulting in a user driven action.
While removing all the scree-sharing constraints, would the share capture types not still have to be defined in a similar way or have some sort of capabilities interface that allows the user to select what share options are currently available to them.

-----Original Message-----
From: Martin Thomson [mailto:martin.thomson@gmail.com] 
Sent: 29 October 2015 14:13
To: Keith Griffin (kegriffi) <kegriffi@cisco.com>
Cc: public-media-capture@w3.org
Subject: Re: Gutting the screen sharing constraints

No, the idea here is to remove all the screen-sharing constraints.
You would have a single call that caused the user to select what to share.  This would bundle all those various types of thing into the same prompt always.  The browser might probably want to structure the question somehow to aid comprehension.

This would prevent a site from directly requesting "monitor" or "browser" sharing (i.e., the least safe options).  We might still retain those terms in the document, but they would not be API-visible constructs.

On 29 October 2015 at 20:35, Keith Griffin (kegriffi) <kegriffi@cisco.com> wrote:
> Just to be clear, in your proposed change we would keep the enum for each capture type so that the sharing application can present different share options to the user and to capture user consent.
>
> enum OutputCaptureSurfaceType {
>     "monitor",
>     "window",
>     "application",
>     "browser"
> };
>
> Authorising the display capture would not require application constraints to be pushed to the user. The user would choose what they want to share from the list of defined capture types?
>
>
> -----Original Message-----
> From: Martin Thomson [mailto:martin.thomson@gmail.com]
> Sent: 29 October 2015 07:55
> To: public-media-capture@w3.org
> Subject: Gutting the screen sharing constraints
>
> At the last meeting, we more or less agreed to remove the distinction between "application" sharing and "window" sharing.
>
> After a lot more thought about this, I want to suggest a more dramatic
> change: removing the constraints that allow an application to restrict which display surfaces are offered to the user to select from.
>
> I originally added these constraints because that is how the Firefox implementation works.  But the more that I think about this, the less sense it makes in the general case.  Having the application push particular options to the fore is fraught from a security perspective and it's not clear that narrowing the selection has been of much use in applications.  Letting the user choose seems like the best option.
>
> If we think that narrowing the selection is necessary for UX reasons, I think that we can easily find better UX options.
>
Received on Friday, 30 October 2015 10:51:57 UTC

This archive was generated by hypermail 2.3.1 : Friday, 30 October 2015 10:51:58 UTC