Intent to deprecate and remove: requestAutocomplete

217 views
Skip to first unread message

Zach Koch

unread,
May 5, 2016, 8:34:31 PM5/5/16
to blin...@chromium.org, db...@chromium.org

Primary eng (and PM) emails

db...@chromium.org, zk...@chromium.org


Summary

We want to deprecate and remove requestAutocomplete. We have seen very little adoption from both other browsers as well as developers. Usage stats are found below. We're working on a new payment standard, PaymentRequest, that borrows ideas from rAc.


Motivation

* Blink is the only platform to support it

* Usage is very low

* We're focusing on PaymentRequest, which has cross-browser support


Compatibility Risk

Very little. See usage stats from above. No other platforms implemented requestAutocomplete.


Usage information from UseCounter

Effectively 0%: https://www.chromestatus.com/metrics/feature/timeline/popularity/965


Entry on the feature dashboard:

https://www.chromestatus.com/features/5092662981951488


Requesting approval to remove too?

Yes.


Thanks,


Zach


TAMURA, Kent

unread,
May 5, 2016, 11:37:45 PM5/5/16
to Zach Koch, blink-dev, db...@chromium.org
LGTM1.
The usage looks very very low, and other browsers don't support the feature.  I think it's safe to remove the feature without deprecation

Please propose a feature removal from the specification too.

--
TAMURA Kent
Software Engineer, Google


Jochen Eisinger

unread,
May 6, 2016, 5:58:44 AM5/6/16
to TAMURA, Kent, Zach Koch, blink-dev, db...@chromium.org
Doesn't mozilla work on this feature for a while now? Any feedback from them?

smaug

unread,
May 6, 2016, 10:24:38 AM5/6/16
to Jochen Eisinger, TAMURA, Kent, Zach Koch, blink-dev, db...@chromium.org
On 05/06/2016 12:58 PM, Jochen Eisinger wrote:
> Doesn't mozilla work on this feature for a while now? Any feedback from them?

FWIW, I filed https://bugzilla.mozilla.org/show_bug.cgi?id=1270740 since I haven't seen any work happening recently.
I'll try to get feedback from relevant developers there.


I filed also https://github.com/whatwg/html/issues/1207


-Olli


>
> On Fri, May 6, 2016 at 5:37 AM TAMURA, Kent <tk...@chromium.org <mailto:tk...@chromium.org>> wrote:
>
> LGTM1.
> The usage looks very very low, and other browsers don't support the feature. I think it's safe to remove the feature without deprecation
>
> Please propose a feature removal from the specification too.
>
>
> On Fri, May 6, 2016 at 9:34 AM, 'Zach Koch' via blink-dev <blin...@chromium.org <mailto:blin...@chromium.org>> wrote:
>
> Primary eng (and PM) emails
>
> db...@chromium.org <mailto:db...@chromium.org>, zk...@chromium.org <mailto:zk...@chromium.org>
>
>
> Summary
>
> We want to deprecate and remove requestAutocomplete. We have seen very little adoption from both other browsers as well as developers. Usage
> stats are found below. We're working on a new payment standard, PaymentRequest, that borrows ideas from rAc.
>
>
> Motivation
>
> * Blink is the only platform to support it
>
> * Usage is very low
>
> * We're focusing on PaymentRequest <https://w3c.github.io/browser-payment-api/specs/paymentrequest.html>, which has cross-browser support
>
>
> Compatibility Risk
>
> Very little. See usage stats from above. No other platforms implemented requestAutocomplete.
>
>
> Usage information from UseCounter
>
> Effectively 0%: https://www.chromestatus.com/metrics/feature/timeline/popularity/965
>
>
> Entry on the feature dashboard <https://www.chromestatus.com/>:
>
> https://www.chromestatus.com/features/5092662981951488
>
>
> Requesting approval to remove too?
>
> Yes.
>
>
> Thanks,
>
>
> Zach
>
>
>
>
>
> --
> TAMURA Kent
> Software Engineer, Google
>
>
> --
> You received this message because you are subscribed to the Google Groups "blink-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to blink-dev+...@chromium.org
> <mailto:blink-dev+...@chromium.org>.

Rick Byers

unread,
May 6, 2016, 11:07:32 AM5/6/16
to smaug, Jochen Eisinger, TAMURA, Kent, Zach Koch, blink-dev, Dan Beam
Wow, I'm shocked at how low the usage is given the improvement in checkout user experience (I hate when autocomplete doesn't just work and I need to type this stuff in).  I double checked (including that the UseCounter does indeed trigger on the demo page) and indeed the usage is effectively zero.

LGTM2 to deprecate for at least one release then remove.

Jochen Eisinger

unread,
May 6, 2016, 11:09:37 AM5/6/16
to Rick Byers, smaug, TAMURA, Kent, Zach Koch, blink-dev, Dan Beam

lgtm3

Dan Beam

unread,
May 8, 2016, 11:31:05 PM5/8/16
to Jochen Eisinger, Rick Byers, smaug, TAMURA, Kent, Zach Koch, blink-dev
So, just clarify the current state of this thread:
  • tkent is cool with removal without deprecation
  • rbyers wants deprecation + then removal
  • jochen is ??
Should I be deprecating or just removing?

For what it's worth, rAc never shipped anywhere outside of Chrome (and has been unshipped on iOS for a while now) and we highly encouraged folks to do runtime feature detection; it's unlikely anybody just assumed rAc exists.

-- Dan

Zach Koch

unread,
May 8, 2016, 11:44:10 PM5/8/16
to Dan Beam, Jochen Eisinger, Rick Byers, smaug, TAMURA, Kent, blink-dev
This is also being pulled out of GMSCore soon (which powers the Android implementation), so I would propose going straight to removal.

Brett Wilson

unread,
May 9, 2016, 1:33:54 AM5/9/16
to Zach Koch, Dan Beam, Jochen Eisinger, Rick Byers, smaug, TAMURA, Kent, blink-dev
On Sun, May 8, 2016 at 8:43 PM, 'Zach Koch' via blink-dev <blin...@chromium.org> wrote:
This is also being pulled out of GMSCore soon (which powers the Android implementation), so I would propose going straight to removal.

+1 on removing it right away.

Brett

Jochen Eisinger

unread,
May 9, 2016, 2:46:52 AM5/9/16
to Brett Wilson, Zach Koch, Dan Beam, Rick Byers, smaug, TAMURA, Kent, blink-dev
I'm fine with removing right away

Rick Byers

unread,
May 9, 2016, 2:18:24 PM5/9/16
to Jochen Eisinger, Brett Wilson, Zach Koch, Dan Beam, smaug, TAMURA, Kent, blink-dev
Sorry, I originally missed that tkent@ proposed direct removal.  I was thinking of the odd scenario like some Chrome-specific enterprise app which is invisible to UMA.  But a console warning is unlikely to help that scenario much either.

Given that our metrics say this API is effectively never used, I'm OK with direct removal also.
Reply all
Reply to author
Forward
0 new messages