Intent to Experiment: AppCache

[Adrienne Walker]

Contact emails

en...@chromium.org, pwn...@chromium.org

Spec

https://html.spec.whatwg.org/multipage/offline.html#offline

Summary

As a part of the deprecation and removal process for AppCache, we would like to allow developers to sign up for a reverse origin trial so that, if needed, they can have more time to transition their AppCache usage over to service workers.

The reasoning for AppCache removal is documented in the blink-dev discussion thread links below.

Link to “Intent to Deprecate” blink-dev discussion

https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/FvM-qo7BfkI/0daqyD8kCQAJ

https://groups.google.com/a/chromium.org/g/blink-dev/c/FvM-qo7BfkI/m/AvxoE6JpBgAJ

Goals for experimentation

AppCache manifests on disk must be refreshed with an origin trial token so that in a future release (targeted at M85) we can continue to load the cached resources referenced by the manifest even if users are offline.  We also plan to gate the AppCache api using the origin trial in this release.

Experimental timeline

The experiment will begin in M83.  We will require origin trial tokens in M85.  The experiment is targeted to continue until M90, where we will disable AppCache for all users.

Any risks when the experiment finishes?

We will need to time the expiration of the origin trial to extend through the point that AppCache is removed.

Ongoing technical constraints

None

Will this feature be supported on all five Blink platforms supported by Origin Trials (Windows, Mac, Linux, Chrome OS, and Android)?

Yes.

Link to entry on the feature dashboard

https://www.chromestatus.com/feature/6192449487634432

[Anne van Kesteren]

On Tue, Apr 21, 2020 at 7:10 PM Adrienne Walker <en...@chromium.org> wrote:
> The reasoning for AppCache removal is documented in the blink-dev discussion thread links below.

In Firefox we have some separation between disabling of the cache and
disabling of the accompanying API (which no-ops if only the former is
disabled) and thus far we've only really looked at disabling the
cache. Does Chrome have something similar or is this really about
removing all?


> https://groups.google.com/a/chromium.org/g/blink-dev/c/FvM-qo7BfkI/m/AvxoE6JpBgAJ

This is a 404 for me.


> https://www.chromestatus.com/feature/6192449487634432

This lists Firefox as having shipped this, but it's actually still in progress.

[Jeremy Roman]

On Tue, Apr 21, 2020 at 1:34 PM Anne van Kesteren <ann...@annevk.nl> wrote:

On Tue, Apr 21, 2020 at 7:10 PM Adrienne Walker <en...@chromium.org> wrote:
> The reasoning for AppCache removal is documented in the blink-dev discussion thread links below.

In Firefox we have some separation between disabling of the cache and
disabling of the accompanying API (which no-ops if only the former is
disabled) and thus far we've only really looked at disabling the
cache. Does Chrome have something similar or is this really about
removing all?


> https://groups.google.com/a/chromium.org/g/blink-dev/c/FvM-qo7BfkI/m/AvxoE6JpBgAJ

This is a 404 for me.

https://groups.google.com/a/chromium.org/d/topic/blink-dev/FvM-qo7BfkI/discussion

 

> https://www.chromestatus.com/feature/6192449487634432

This lists Firefox as having shipped this, but it's actually still in progress.

--
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.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CADnb78i10iW4hp3HFEOz9gdnjXxMwHxtJwEE2Hiqd7RkkguwWA%40mail.gmail.com.

[Jeremy Roman]

On Tue, Apr 21, 2020 at 1:39 PM Jeremy Roman <jbr...@chromium.org> wrote:

On Tue, Apr 21, 2020 at 1:34 PM Anne van Kesteren <ann...@annevk.nl> wrote:

On Tue, Apr 21, 2020 at 7:10 PM Adrienne Walker <en...@chromium.org> wrote:
> The reasoning for AppCache removal is documented in the blink-dev discussion thread links below.

In Firefox we have some separation between disabling of the cache and
disabling of the accompanying API (which no-ops if only the former is
disabled) and thus far we've only really looked at disabling the
cache. Does Chrome have something similar or is this really about
removing all?


> https://groups.google.com/a/chromium.org/g/blink-dev/c/FvM-qo7BfkI/m/AvxoE6JpBgAJ

This is a 404 for me.

https://groups.google.com/a/chromium.org/d/topic/blink-dev/FvM-qo7BfkI/discussion

erm, more specifically https://groups.google.com/a/chromium.org/d/msg/blink-dev/FvM-qo7BfkI/AvxoE6JpBgAJ

apologies; some of us are clearly opted into something with a different URL path format

[Mike West]

LGTM1.

As discussed on https://groups.google.com/a/chromium.org/forum/#!topic/blink-api-owners-discuss/ivh_NDBpjEs (which I've checked in Incognito, and seems to actually work :) ), starting a reverse origin trial so that folks who need to can begin grabbing keys is a good idea.

Anne raises good questions about the details, though, and I'd appreciate it if y'all could collaborate on a PR against HTML that detailed the approach in a more formal way so that we have a shared understanding of what developers ought to be expecting. I don't think that needs to block the origin trial, but I do think it would reduce the interop risk when y'all pull the trigger in ~M85.

-mike

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CACuR13fh083w0jH7zoTi0ae0qVDVsXMtWUj4EF4m1WWd2NrwNA%40mail.gmail.com.

[Chris Harrelson]

On Tue, Apr 21, 2020 at 12:15 PM Mike West <mk...@chromium.org> wrote:

LGTM1.

BTW, since this is an I2E, only that one LGTM is necessary.

 

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAKXHy%3Dd4hH0TeUC6oXcD257_9vnGti%3DLw5Y56%2Bj4bTopoWgSVA%40mail.gmail.com.

[Adrienne Walker]

On Tue, Apr 21, 2020 at 10:34 AM Anne van Kesteren <ann...@annevk.nl> wrote:
>
> On Tue, Apr 21, 2020 at 7:10 PM Adrienne Walker <en...@chromium.org> wrote:
> > The reasoning for AppCache removal is documented in the blink-dev
> > discussion thread links below.
>
> In Firefox we have some separation between disabling of the cache and
> disabling of the accompanying API (which no-ops if only the former is
> disabled) and thus far we've only really looked at disabling the
> cache. Does Chrome have something similar or is this really about
> removing all?

These two are separable in Chrome, but our plan is to disable both of
them at once.

I see a couple of negatives to disabling only the backend. The large
one is that having the API stop working earlier is a last-ditch
communication effort to reach developers who need to transition off of
AppCache. The second minor negative is that having the AppCache API
partially be working may be confusing to developers. Because there's an
escape valve of an origin trial, I think this is the best path for
Chrome.

(That said, because of how Chrome's AppCache origin trial is
implemented, developers could still get into this state of having an API
but no backend and so I'm happy to spec out the details of what that
would look like.)

> > https://groups.google.com/a/chromium.org/g/blink-dev/c/FvM-qo7BfkI/m/AvxoE6JpBgAJ
>
> This is a 404 for me.

Sorry about that!

> > https://www.chromestatus.com/feature/6192449487634432
>
> This lists Firefox as having shipped this, but it's actually still in
> progress.

Fixed.

On Tue, Apr 21, 2020 at 12:15 PM Mike West <mk...@chromium.org> wrote:
>
> Anne raises good questions about the details, though, and I'd
> appreciate it if y'all could collaborate on a PR against HTML that
> detailed the approach in a more formal way so that we have a shared
> understanding of what developers ought to be expecting. I don't think
> that needs to block the origin trial, but I do think it would reduce
> the interop risk when y'all pull the trigger in ~M85.

Yes, agreed. I will put a PR together to start a discussion.

[Adrienne Walker]

On Mon, Apr 27, 2020 at 11:14 AM Adrienne Walker <en...@chromium.org> wrote:
>
> On Tue, Apr 21, 2020 at 12:15 PM Mike West <mk...@chromium.org> wrote:
> >
> > Anne raises good questions about the details, though, and I'd
> > appreciate it if y'all could collaborate on a PR against HTML that
> > detailed the approach in a more formal way so that we have a shared
> > understanding of what developers ought to be expecting. I don't think
> > that needs to block the origin trial, but I do think it would reduce
> > the interop risk when y'all pull the trigger in ~M85.
>
> Yes, agreed.  I will put a PR together to start a discussion.

To follow up, this discussion is here:
https://github.com/whatwg/html/pull/5505

[Mike West]

Thanks for following up on this. It looks like there's reasonable agreement about the short-term direction in that PR against HTML; thank you for starting that conversation!

-mike