W3C home > Mailing lists > Public > whatwg@whatwg.org > May 2014

Re: [whatwg] [Notifications] Notification.hasPermission() instead of Notification.permission

From: Anne van Kesteren <annevk@annevk.nl>
Date: Tue, 13 May 2014 14:07:21 +0200
Message-ID: <CADnb78jXHQLDLnQTxzdr7e63HOxJtLU73u1xaa7a99bt30bJ0A@mail.gmail.com>
To: Andrew Wilson <atwilson@google.com>
Cc: WHATWG <whatwg@whatwg.org>, Peter Beverloo <beverloo@google.com>
On Tue, May 13, 2014 at 1:59 PM, Andrew Wilson <atwilson@google.com> wrote:
> Just out of curiosity, what are we hoping that the use counter will show?
> I'm presuming that every single app that uses the notification API will make
> at least some use of Notification.permission, so it mostly seems like it's
> going to show us how often the Notification APIs *in general* are used in
> the wild. Is that going to be useful in making an API decision? If so, how
> (what numbers would allow us to feel comfortable removing
> Notification.permission)?
>
> I ask this, because the paradigm I'm most used to is to define a new
> replacement API, mark the old API as deprecated, then use UseCounters to
> determine when enough sites have migrated to the new API that you feel
> comfortable removing the deprecated API. Using UseCounters to measure raw
> usage when there really isn't an alternative API available yet is not an
> approach I'm familiar with.

I'd like to know whether the numbers are low enough to even make an
effort here. We should also figure out what different approaches we've
taken so far I suppose and what we want to do going forward.


-- 
http://annevankesteren.nl/
Received on Tuesday, 13 May 2014 12:07:45 UTC

This archive was generated by hypermail 2.4.0 : Wednesday, 22 January 2020 17:00:20 UTC