Re: [w3c/permissions] Introduce "Automation" section (#151)

Thanks for all the feedback, @jyasskin. I've been a bit slow to implement it
this week because I've been concerned that this patch contains some fundamental
flaw from an implementer's or editor's perspective. After revisiting the TPAC
meeting minutes, reviewing @burg's design document, and checking in with
@AutomatedTester, I don't believe this is the case. I've incorporated most of
your advice, but there are a few things I could use clarification on:

> They may be better ways to word that, but I'm trying to hook into the "If
> there was a previous invocation of this algorithm ..." wording.

I'm trying this for now: "Interpret parameters.state as if it were the result
of an invocation of permission state for typedDescriptor with the argument
target made at this moment." But I think that we should probably formalize this
state.

> If allRealms is true, you'll post a task to all other realms with the same
> origin to do the same thing.

As I understand it, WebDriver algorithms execute outside of any particular
event loop. If that's correct, then I think task queuing is unnecessary. Also
note that in accordance with your other suggestion, this is now worded in terms
of environment settings objects (not realms). I've been a bit more formal about
how that set is defined.

> We'll have some trouble implementing this for the always-granted APIs like
> "midi", so you may want to let the UA explicitly reject the update.

I don't really understand this condition, but it seems as though it can be
enforced at the onset of this algorithm. I'd prefer to do it there if possible.
What do you think?

> Also, could you fix the IPR issue described in
> https://labs.w3.org/hatchery/repo-manager/pr/id/w3c/permissions/151? Thanks.

All set!

(By the way, your suggestion to use WebIDL for type checking was awesome! Thank
you for pointing that out.)

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/permissions/pull/151#issuecomment-350098779