[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] Proposed change to WebGL Event definition

On Sat, Aug 21, 2010 at 6:25 PM, Gregg Tavares (wrk) <gman@google.com> wrote:

On Fri, Aug 20, 2010 at 11:22 AM, Vladimir Vukicevic <vladimir@mozilla.com> wrote:

----- Original Message -----
> On Aug 20, 2010, at 10:02 AM, Vladimir Vukicevic wrote:
> > Looks good to me, and nicely solves the "give more info if we have
> > to return null". It might be nice to add a status code that says
> > something like "isn't available, might be soon" and then another for
> > "became available" -- I'm thinking of the case where the browser
> > wants to query the user if they want to allow WebGL.
> My small concern there is that such a message might be a security
> and/or privacy violation. It would give the author information that
> the user is being asked a question and has responded negatively. Seems
> like you want to tell the author nothing when the dialog comes up. But
> I'm no security expert...

Hmm, so maybe a better way to do it would be to just add a context_available message; initially, you would get a not_available without necessarily knowing why.  Only if the user accepts would you get the context_available event, and can try again.  That lets the app put up some kind of UI on not available, but get notified if it does become available.

Do we really need messages for WebGL? Is WebGL more special than 2d canvas or Flash? 

Can't this happen behind the scenes? If the browser wants to give the user an option for WebGL it should freeze the _javascript_ on that page, ask the question and then unfreeze the _javascript_ letting context creation succeed or fail. 

I understand there are a few issues with WebGL but users are unlikely to care about those issues enough to want a question on every page that uses WebGL. I'd suspect they are more likely to want a permission message for canvas in general to stop annoying ads. Since this solution won't work for canvas in general it seems like the wrong solution.

To be clear, I don't have a problem with an event for failure.

I have a problem, if i understand the suggestion above, that the correct way to use WebGL will require asynchronous initialization.


   - Vlad
You are currently subscribed to public_webgl@khronos.org.
To unsubscribe, send an email to majordomo@khronos.org with
the following command in the body of your email: