LGTM3 for the first CL On Thu, May 30, 2024 at 6:22 PM Domenic Denicola <[email protected]> wrote:
> LGTM2, to land the first CL. > > Thanks for explaining. It sounds like the first CL, which this Intent is > covering, is purely additive and should not have significant compatibility > risk. It also brings us in line with the behavior Firefox has had for a > long time, and the behavior that WebKit has recently adopted. > > I look forward to your second Intent to Ship for the second CL, which will > require more caution. > > On Thursday, May 30, 2024 at 4:52:23 PM UTC+9 [email protected] wrote: > >> Hi Yoav&Alex, thanks for the feedback. I'll make it more clear. >> >> Generally speaking, the final goal is that, we want to avoid queuing a >> task to dispatch a selectionchange event when there is already a task >> scheduled to do that. This is to catch up the latest spec: >> https://w3c.github.io/selection-api/#scheduling-selectionhange-event. >> Currently we divide this into 2 steps with one CL for each step to >> impelment this goal. >> >> In this issue, we'd like to land *the first CL* that updates the >> implementation of selectionchange event to be fired on input element and >> textarea element instead of document. So the difference in the first CL is: >> >> 1. The old behaviour: When the element(input/textarea) provides a >> text selectionchange or its selection changes, we dispatch a >> selectionchange event on the document directly. >> 2. The new behaviour: The new codes goes to that, in the same >> situations, if the selection changes, we fire this event first on the >> element(input/textarea) itself, and then it would be bubbled to the >> doument. >> >> >> So, the final results are the same that, we finally get a selectionchange >> event on the document. *The feature in this issue* is mainly to do this >> change. After the first CL is landed, then we come to the second CL. The >> main change of the behaviour would be in the second CL. >> >> In *the second CL*, we will change the behaviour: >> >> 1. The old behaviour: When the element(input/textarea) provides a >> text selectionchange or its selection changes, we dispatch a >> selectionchange event on the element. We don't care whether this event is >> executed, and we just dipatch it. >> 2. The new behaviour: Now based on the new spec( >> https://w3c.github.io/selection-api/#scheduling-selectionhange-event), >> we dispatch this event only when there is no scheduled one in the queue. >> If >> we have already scheduled one but not get callbacked, we would skip to >> dispatch this event. >> >> To sum up, in this issue or mail thread, we only focus on what we did in >> the first CL. So previouly we think this is low-risk because we don't >> change anything at all. The real impact would happen in the second CL. So >> for the compat risk or maybe the potential for site breakage, maybe we >> could collect these data in the second CL. >> >> These are the main difference between the first CL and the second CL. If >> you have any other concerns, feel free to ping me! >> >> Thanks! >> >> On Wednesday, May 29, 2024 at 11:52:57 PM UTC+8 [email protected] >> wrote: >> >>> hey folks, >>> >>> We spent a lot of time in API OWNERs today trying to understand this >>> change, and we couldn't crisply describe: >>> >>> >>> - What the old behaviour was >>> - What the new behaviour is going to be >>> - If we've analysed the potential for site breakage, and if not, if >>> we should be adding usecounters ahead of the change >>> >>> Most of this is stuff we'd expect to see in an Explainer. Is it possible >>> to produce one? Seeing example code for whatw this improves would be >>> helpful. >>> >>> Best, >>> >>> Alex >>> >>> On Wednesday, May 29, 2024 at 1:43:00 AM UTC-7 Yoav Weiss wrote: >>> >>>> I'm sorry, but re-reading this, I realized that I don't fully >>>> understand this change and the plan here. >>>> Can you elaborate on: >>>> * What is the behavior change that we want to drive here? >>>> * Which part of that would be covered by the first CL? which by the >>>> second? >>>> * Where do we predict compat risk here? How can we estimate it? >>>> >>>> Thanks! :) >>>> >>>> >>>> On Tue, May 28, 2024 at 1:54 AM Shuangshuang Zhou < >>>> [email protected]> wrote: >>>> >>>>> Hi Yoav, thanks for the question! From chromium's side, currently I >>>>> don't have any usecounters to trace how many sites would be impacted. I >>>>> checked WebKit' status, seems they don't have any data. Maybe Olli could >>>>> comment this from Gecko's side. >>>>> >>>>> Since the first CL doesn't change the original logic, we might collect >>>>> some data in the second CL(avoid to dispatch events that are already >>>>> scheduled but not executed). >>>>> >>>>> Thanks! >>>>> >>>>> On Monday, May 27, 2024 at 3:08:29 PM UTC+8 [email protected] >>>>> wrote: >>>>> >>>>>> Given Olli's comments on this change being backwards-incompatible, do >>>>>> we have any data on how many sites would be impacted? Any usecounters to >>>>>> that effect? >>>>>> >>>>>> On Mon, May 27, 2024 at 2:19 AM Shuangshuang Zhou < >>>>>> [email protected]> wrote: >>>>>> >>>>>>> Thanks Olli for the implement in Gecko side, and also Dan's comment! >>>>>>> >>>>>>> Hi Mike, would you please take a look and give some suggestion? >>>>>>> Also, seems we need 3 LGTMs, Kent or Mike, would you add another >>>>>>> reviewer >>>>>>> to to take a review? >>>>>>> >>>>>>> Thanks! >>>>>>> On Saturday, May 25, 2024 at 1:05:59 AM UTC+8 Olli Pettay wrote: >>>>>>> >>>>>>>> I am trying to land the change, so it should be in FF 128 Nightly >>>>>>>> soon, next week at latest. So hopefully we'll get feedback if it breaks >>>>>>>> something. >>>>>>>> I don't think there is need for a standards-position for this >>>>>>>> particular small tweak. >>>>>>>> (But the process of making this kind of breaking spec change was >>>>>>>> rather unusual.) >>>>>>>> >>>>>>>> >>>>>>>> -Olli >>>>>>>> >>>>>>>> On Friday, May 24, 2024 at 7:55:44 PM UTC+3 Daniel Clark wrote: >>>>>>>> >>>>>>>>> Thanks Shuangshuang for clarifying what’s going on with these two >>>>>>>>> separate but related changes, and Olli for the update on >>>>>>>>> https://bugzilla.mozilla.org/show_bug.cgi?id=1898343. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Since Gecko already implements the behavior of firing >>>>>>>>> selectionchange on input/textarea (confirmed here >>>>>>>>> <https://glow-curvy-cerise.glitch.me/>), sounds good to me to go >>>>>>>>> ahead with shipping this change in Chromium (I’m not an API owner >>>>>>>>> though). >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Separately, for the behavior of not firing duplicate >>>>>>>>> selectionchange, it might not hurt to go ahead and request an official >>>>>>>>> Mozilla position on that since it sounds like there’s still a bit of >>>>>>>>> uncertainty about landing that change given back compat concerns, >>>>>>>>> unless >>>>>>>>> Olli confirms here that the plan is to go ahead and ship it. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -- Dan >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *From:* Olli Pettay <[email protected]> >>>>>>>>> *Sent:* Friday, May 24, 2024 12:26 AM >>>>>>>>> *To:* blink-dev <[email protected]> >>>>>>>>> *Cc:* Shuangshuang Zhou <[email protected]>; >>>>>>>>> [email protected] <[email protected]>; [email protected] < >>>>>>>>> [email protected]>; Daniel Clark <[email protected]>; Olli >>>>>>>>> Pettay <[email protected]> >>>>>>>>> *Subject:* Re: [blink-dev] Intent to Ship: Dispatch >>>>>>>>> selectionchange event per element >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> You don't often get email from [email protected]. Learn why this >>>>>>>>> is important <https://aka.ms/LearnAboutSenderIdentification> >>>>>>>>> >>>>>>>>> FWIW, I'm implementing the change in >>>>>>>>> https://bugzilla.mozilla.org/show_bug.cgi?id=1898343 >>>>>>>>> >>>>>>>>> Need to still fix some tests, which do reveal how the spec change >>>>>>>>> indeed isn't fully backwards compatible. >>>>>>>>> >>>>>>>>> Hopefully it doesn't break any real web sites. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -Olli >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Friday, May 24, 2024 at 10:05:47 AM UTC+3 Shuangshuang Zhou >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> Hi Dan&Mike, I think we could submit a standard-position issue to >>>>>>>>> Mozilla for reducing duplicated events in our second CL(mentioned in >>>>>>>>> my >>>>>>>>> last comment) for this feature. Because in the second CL, we might >>>>>>>>> change >>>>>>>>> the current logic of dispatching selectionchange event. For this >>>>>>>>> time, we >>>>>>>>> might just do it in chromium. WDYT? >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Thanks! >>>>>>>>> >>>>>>>>> On Friday, May 24, 2024 at 9:20:33 AM UTC+8 [email protected] >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> LGTM1. >>>>>>>>> >>>>>>>>> Firefox and Safari statuses are strong signals, and the >>>>>>>>> compatibility risk looks very low. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Thu, May 23, 2024 at 11:07 AM Shuangshuang Zhou < >>>>>>>>> [email protected]> wrote: >>>>>>>>> >>>>>>>>> Hi Dan&Mike, for your concern that Firefox is still failing the >>>>>>>>> WPTs at >>>>>>>>> https://wpt.fyi/results/selection/onselectionchange-on-distinct-text-controls.html, >>>>>>>>> that might be another issue. Let me make more explanation. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Actually, the final goal of w3c modifys the spec of >>>>>>>>> selectionchange event is to avoid firing duplicated selectionchange >>>>>>>>> event >>>>>>>>> which are not executed yet on the same element(as Rniwa commented in >>>>>>>>> his >>>>>>>>> comment >>>>>>>>> <https://github.com/w3c/selection-api/issues/170#issuecomment-1956096554>). >>>>>>>>> Then in WebKit/WebCore, Rniwa submitted 2 CLs to achieve this ieda. >>>>>>>>> The >>>>>>>>> first CL in WebKit is 276238@main >>>>>>>>> <https://commits.webkit.org/276238@main>, and in this CL they >>>>>>>>> just modified the logic to fire selectionchange event on >>>>>>>>> input/textarea for >>>>>>>>> the first step. Then the second CL in WebKit is 276388@main >>>>>>>>> <https://commits.webkit.org/276388@main>, which finally avoided >>>>>>>>> to fire duplicated selectionchange events according to the latest >>>>>>>>> spec on firing >>>>>>>>> selectionchange event >>>>>>>>> <https://w3c.github.io/selection-api/#firing-selectionhange-event> >>>>>>>>> . >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> So for this feature in Blink here , we also want to take 2 steps >>>>>>>>> like WebKit does that in the first CL we change to fire on >>>>>>>>> input/textarea >>>>>>>>> and then sovle the duplicated events in the second CL. And those >>>>>>>>> tests are >>>>>>>>> already modified to what both CLs are merged. So in the above test of >>>>>>>>> onselectionchange-on-distinct-text-controls.html, chrome and firefox >>>>>>>>> should >>>>>>>>> be failed because only Safari has landed 2 CLs to avoid duplicated >>>>>>>>> selectionchange events. Or in other words, firefox might already >>>>>>>>> dispatch >>>>>>>>> selectionchange event on elements but doesn't have the logic to avoid >>>>>>>>> to >>>>>>>>> fire scheduled-but-not-executed events. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> For WebKit, currently the two CLs are both landed in Safari >>>>>>>>> Technology Preview 192(link >>>>>>>>> <https://developer.apple.com/documentation/safari-technology-preview-release-notes/stp-release-192>), >>>>>>>>> but it might need some time to ship in Safari because Safari is >>>>>>>>> updated >>>>>>>>> within OS updates. >>>>>>>>> >>>>>>>>> For the status in FireFox, I'll double-check and confirm with >>>>>>>>> Tkent(one reviewer of my first CL). >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> >>>>>>>>> Shuangshuang >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Thursday, May 23, 2024 at 9:29:59 AM UTC+8 [email protected] >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> Yes, agree. Can we please request a position from Mozilla at >>>>>>>>> https://github.com/mozilla/standards-positions/issues/new? >>>>>>>>> >>>>>>>>> On 5/22/24 6:43 PM, 'Daniel Clark' via blink-dev wrote: >>>>>>>>> >>>>>>>>> Yeah, that’s what I was trying to get at – the Intent implies that >>>>>>>>> Gecko has also shipped the breaking change but it seems that might >>>>>>>>> not be >>>>>>>>> the case. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> If not, we should send a Request for Position to figure out >>>>>>>>> whether there would be cross-browser alignment on shipping this. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -- Dan >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *From:* Olli Pettay <[email protected]> >>>>>>>>> *Sent:* Wednesday, May 22, 2024 9:53 AM >>>>>>>>> *To:* blink-dev <[email protected]> >>>>>>>>> *Cc:* Daniel Clark <[email protected]>; Shuangshuang Zhou < >>>>>>>>> [email protected]> >>>>>>>>> *Subject:* Re: [blink-dev] Intent to Ship: Dispatch >>>>>>>>> selectionchange event per element >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> You don't often get email from [email protected]. Learn why this >>>>>>>>> is important <https://aka.ms/LearnAboutSenderIdentification> >>>>>>>>> >>>>>>>>> I think it is because of this rather surprising non-backwards >>>>>>>>> compatible recent change. >>>>>>>>> >>>>>>>>> On Wednesday, May 22, 2024 at 7:28:45 PM UTC+3 Daniel Clark wrote: >>>>>>>>> >>>>>>>>> The Gecko status is marked as Shipped/Shipping and the >>>>>>>>> Interop/Compat section mentions Firefox having shipped this, but >>>>>>>>> Firefox is >>>>>>>>> still failing the WPTs at >>>>>>>>> https://wpt.fyi/results/selection/onselectionchange-on-distinct-text-controls.html >>>>>>>>> . >>>>>>>>> >>>>>>>>> Can you help me understand the discrepancy there? >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> >>>>>>>>> Dan >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *From:* [email protected] <[email protected]> *On Behalf Of >>>>>>>>> *Shuangshuang Zhou >>>>>>>>> *Sent:* Tuesday, May 21, 2024 10:24 PM >>>>>>>>> *To:* blink-dev <[email protected]> >>>>>>>>> *Subject:* [blink-dev] Intent to Ship: Dispatch selectionchange >>>>>>>>> event per element >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> You don't often get email from [email protected]. Learn why >>>>>>>>> this is important <https://aka.ms/LearnAboutSenderIdentification> >>>>>>>>> >>>>>>>>> *Contact **emails*[email protected] >>>>>>>>> >>>>>>>>> *Explainer*None >>>>>>>>> >>>>>>>>> *Specification* >>>>>>>>> https://w3c.github.io/selection-api/#selectionchange-event >>>>>>>>> >>>>>>>>> *Summary* >>>>>>>>> >>>>>>>>> Dispatches selectionchange event per element when this >>>>>>>>> element(input/textarea) provides a text selection or its selection >>>>>>>>> changes. >>>>>>>>> This is to match the latest specification of selectionchange event. >>>>>>>>> This >>>>>>>>> also matches Safari behavior. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Blink component*Blink>Editing>Selection >>>>>>>>> <https://bugs.chromium.org/p/chromium/issues/list?q=component:Blink%3EEditing%3ESelection> >>>>>>>>> >>>>>>>>> *TAG review*None >>>>>>>>> >>>>>>>>> *TAG review status*Issues addressed >>>>>>>>> >>>>>>>>> *Risks* >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Interoperability and Compatibility* >>>>>>>>> >>>>>>>>> Interoperability risk is low because Firefox and Safari have >>>>>>>>> shipped this according to the specification. Compatibility risk is low >>>>>>>>> because the selectionchange event targeting input/textarea would >>>>>>>>> bubble up, >>>>>>>>> and existing codes listening on Document will work well as ever. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Gecko*: Shipped/Shipping >>>>>>>>> >>>>>>>>> *WebKit*: Shipped/Shipping (https://commits.webkit.org/276238@main >>>>>>>>> ) >>>>>>>>> >>>>>>>>> *Web developers*: No signals >>>>>>>>> >>>>>>>>> *Other signals*: >>>>>>>>> >>>>>>>>> *WebView application risks* >>>>>>>>> >>>>>>>>> *Does this intent deprecate or change behavior of existing APIs, >>>>>>>>> such that it has potentially high risk for Android WebView-based >>>>>>>>> applications?* >>>>>>>>> >>>>>>>>> Low WebView application risks. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Debuggability* >>>>>>>>> >>>>>>>>> None >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Will this feature be supported on all six Blink platforms >>>>>>>>> (Windows, Mac, Linux, ChromeOS, Android, and Android WebView)?*Yes >>>>>>>>> >>>>>>>>> *Is this feature fully tested by **web-platform-tests* >>>>>>>>> <https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md> >>>>>>>>> *?*Yes >>>>>>>>> >>>>>>>>> >>>>>>>>> https://wpt.fyi/results/selection/onselectionchange-on-distinct-text-controls.html >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> *Flag name on chrome://flags*None >>>>>>>>> >>>>>>>>> *Finch feature name*DispatchSelectionchangeEventPerElement >>>>>>>>> >>>>>>>>> *Requires code in //chrome?*False >>>>>>>>> >>>>>>>>> *Tracking bug*https://issues.chromium.org/issues/330766600 >>>>>>>>> >>>>>>>>> *Estimated milestones* >>>>>>>>> Shipping on desktop >>>>>>>>> 127 >>>>>>>>> >>>>>>>>> *Anticipated spec changes* >>>>>>>>> >>>>>>>>> *Open questions about a feature may be a source of future web >>>>>>>>> compat or interop issues. Please list open issues (e.g. links to known >>>>>>>>> github issues in the project for the feature specification) whose >>>>>>>>> resolution may introduce web compat/interop risk (e.g., changing to >>>>>>>>> naming >>>>>>>>> or structure of the API in a non-backward-compatible way).* >>>>>>>>> >>>>>>>>> None >>>>>>>>> >>>>>>>>> *Link to entry on the Chrome Platform Status* >>>>>>>>> https://chromestatus.com/feature/5255454895898624?gate=6043023317401600 >>>>>>>>> >>>>>>>>> This intent message was generated by Chrome Platform Status >>>>>>>>> <https://chromestatus.com/>. >>>>>>>>> >>>>>>>>> -- >>>>>>>>> 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 [email protected]. >>>>>>>>> To view this discussion on the web visit >>>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/de18f3c4-df5a-4423-80ec-e505e0c9fb2bn%40chromium.org >>>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/de18f3c4-df5a-4423-80ec-e505e0c9fb2bn%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 [email protected]. >>>>>>>>> >>>>>>>>> To view this discussion on the web visit >>>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/PH8PR00MB1445785C11C8FCA346EAE963C5EB2%40PH8PR00MB1445.namprd00.prod.outlook.com >>>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/PH8PR00MB1445785C11C8FCA346EAE963C5EB2%40PH8PR00MB1445.namprd00.prod.outlook.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 [email protected]. >>>>>>>>> >>>>>>>>> To view this discussion on the web visit >>>>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/2d3b76b1-890f-4a07-961a-ef813c14eef3n%40chromium.org >>>>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/2d3b76b1-890f-4a07-961a-ef813c14eef3n%40chromium.org?utm_medium=email&utm_source=footer> >>>>>>>>> . >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -- >>>>>>>>> >>>>>>>>> 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 [email protected]. >>>>>>> >>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/b17e9c59-e826-44e3-9a52-e24067f1cbc9n%40chromium.org >>>>>>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/b17e9c59-e826-44e3-9a52-e24067f1cbc9n%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 [email protected]. > To view this discussion on the web visit > https://groups.google.com/a/chromium.org/d/msgid/blink-dev/e2d678cb-a4ae-4856-8518-161b8a3ffc38n%40chromium.org > <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/e2d678cb-a4ae-4856-8518-161b8a3ffc38n%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 [email protected]. To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw_U252oGp4tDKO91GiKHv7V27BpS7_XA9jsREi4v3QWPA%40mail.gmail.com.
