LGTM3 On Wed, Jun 1, 2022 at 8:58 PM Mike Taylor <miketa...@chromium.org> wrote:
> On 6/1/22 1:34 PM, Chris Harrelson wrote: > > > On Tue, May 31, 2022 at 8:26 PM Ayu Ishii <a...@chromium.org> wrote: > >> Hi Mike! >> >> With the current usage measurements we see, we hadn't considered any >> enterprise policy for opt-out. >> But certainly can follow the process to do so if you feel that there may >> be risk of undercounting. >> Deprecation of WebSQL in third-party contexts added a policy that lasted >> 3 milestones after deprecation before full removal as an example. >> Although the usages were quite different from that deprecation, we can >> follow the same process if this sounds reasonable. >> > > I think this plan sounds good. LGTM1 once you have an enterprise opt-out > in place that will remain for 3 milestones. Also please make sure to > communicate this change in the enterprise notes and other communication > channels. > > A couple of notes I took last Friday and forgot to post: > > I dumped the list of sites from HTTPArchive (query below) and after > de-duping them, ended up with 835 sites. > > I then ran a script which naively looks at response codes, and got the > following results: > > 2XX count: 685/835 > 3XX to HTTP endpoint count: 74/835 > 4XX count: 3/835 > 5XX count: 2/835 > > So, from this dataset, roughly 9% of those redirect to an HTTP endpoint. > > That said, I think reducing risk of breakage for enterprise environments > is a useful and friendly thing to do. LGTM2 w/ that done. > > SELECT > rank, > url, > FROM > `httparchive.blink_features.features` > WHERE feature = 'OpenWebDatabaseInsecureContext' > ORDER BY rank ASC > > > >> >> - Ayu >> >> On Monday, May 30, 2022 at 10:57:01 PM UTC-7 Mike West wrote: >> >>> I'm happy to see this moving forward, thanks for pushing it ahead! >>> >>> That said, this seems like the kind of thing that's likely-enough to >>> impact enterprise that we should include a temporary opt-out to give >>> ourselves some wiggle room if it turns out that we're undercounting usage. >>> Have y'all already put something like that together? >>> >>> -mike >>> >>> >>> On Fri, May 27, 2022 at 2:18 AM Ayu Ishii <a...@chromium.org> wrote: >>> >>>> >>>> *Contact emails *a...@chromium.org, jsb...@chromium.org, >>>> ajayrahate...@google.com >>>> >>>> >>>> *Specification *https://www.w3.org/TR/webdatabase/ >>>> >>>> >>>> *Summary *We intend to deprecate and remove usage of WebSQL in >>>> non-secure contexts. Deprecation is targeted for M105 and removal is >>>> targeted for M107. >>>> >>>> >>>> *Blink component *Blink>Storage>WebSQL >>>> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EStorage%3EWebSQL> >>>> >>>> >>>> *Motivation *The Web SQL Database standard was first proposed in April >>>> 2009 and abandoned in November 2010. Gecko never implemented this feature >>>> and WebKit deprecated this feature in 2019 >>>> <https://github.com/WebKit/WebKit/commit/761bce943c0696a6bb93116eb0576ed07dbfdc65>. >>>> The W3C encouraged those needing web databases to adopt Web Storage >>>> <https://www.w3.org/TR/webstorage/> or Indexed Database >>>> <https://www.w3.org/TR/IndexedDB/>. >>>> >>>> WebSQL has been deprecated and removed >>>> <https://groups.google.com/a/chromium.org/g/blink-dev/c/TM6YDx1Hh08> >>>> for third-party contexts in M97. >>>> >>>> We hope to entirely deprecate and remove WebSQL at some future point >>>> when usage is low enough. >>>> >>>> >>>> *TAG review *N/A >>>> >>>> Risks >>>> Based on usage measurements >>>> <https://chromestatus.com/metrics/feature/timeline/popularity/4073> >>>> rolled out in M97, 0.005% of page loads use WebSQL in a non-secure >>>> context. Less than 0.01% of top sites have adopted this feature. >>>> >>>> Out of the 20 top sites listed for the month of April 2022, 11 of the >>>> sites use a feature detection library Modernizr 1.5 >>>> <https://modernizr.com/news/modernizr-15>, on a version released in >>>> 2010. This would create a test database to check feature availability. I >>>> was unable to reproduce the creation of other WebSQL databases outside of >>>> the one created by Modernizr on these sites. 4 sites that seem to use an >>>> outdated >>>> private mode detection script <https://stackoverflow.com/q/48169810> >>>> which was intended for older versions of iOS. All the scripts I have found >>>> checked if window.openDatabase existed before usage, likely due to the lack >>>> of support in Gecko and WebKit. >>>> >>>> Interoperability and Compatibility >>>> >>>> Gecko: Never implemented >>>> >>>> WebKit: Deprecation shipped in iOS 13 and Safari 13 >>>> <https://trac.webkit.org/changeset/277564/webkit> >>>> >>>> Web developers: No signals >>>> >>>> >>>> *Debuggability *N/A >>>> >>>> Is this feature fully tested by web-platform-tests >>>> <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md> >>>> ? >>>> Not fully, one test >>>> <https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/web_tests/external/wpt/clear-site-data/support/test_utils.sub.js;l=136> >>>> checks the availability of the feature. >>>> >>>> >>>> *Tracking bug *https://crbug.com/1212492 >>>> >>>> Link to related intents >>>> Intent to Deprecate and Remove: WebSQL in third-party contexts >>>> <https://groups.google.com/a/chromium.org/g/blink-dev/c/TM6YDx1Hh08> >>>> >>>> >>>> *Link to entry on the Chrome Platform Status * >>>> https://chromestatus.com/feature/5175124599767040 >>>> >>>> >>>> -- >>>> 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+unsubscr...@chromium.org. >>>> To view this discussion on the web visit >>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/bc5f407d-e6fe-4743-ae46-84541d271a92n%40chromium.org >>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/bc5f407d-e6fe-4743-ae46-84541d271a92n%40chromium.org?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >> 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+unsubscr...@chromium.org. >> To view this discussion on the web visit >> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/aa706101-184b-4a25-a446-6643a96e062fn%40chromium.org >> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/aa706101-184b-4a25-a446-6643a96e062fn%40chromium.org?utm_medium=email&utm_source=footer> >> . >> > -- > 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+unsubscr...@chromium.org. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw9xqv76X-fAQjvb4iiT%2Bx-GxUffO0H16AYMWDk91ma-hw%40mail.gmail.com > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw9xqv76X-fAQjvb4iiT%2Bx-GxUffO0H16AYMWDk91ma-hw%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > -- > 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+unsubscr...@chromium.org. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/5a30e8f7-49cf-ab6e-9e36-d63fb0f139fa%40chromium.org > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/5a30e8f7-49cf-ab6e-9e36-d63fb0f139fa%40chromium.org?utm_medium=email&utm_source=footer> > . > -- 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+unsubscr...@chromium.org. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAL5BFfWp6Q0kTRU3%2B9%3DOrSUsCkBYq%2BF2j-YxwLebsPr4ba36Nw%40mail.gmail.com.
