Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-23 Thread Alex Russell 
LGTM3 to extend to 95

On Tuesday, September 21, 2021 at 7:26:16 AM UTC-7 Chris Harrelson wrote:

> LGTM2
>
> On Mon, Sep 20, 2021, 10:38 PM Yoav Weiss  wrote:
>
>> O(10) sounds manageable in case we want to change course, so that reduces 
>> the burn-in risk.
>> Since this is an atypical OT extension, approving it would require 3 
>> LGTMs.
>>
>> *LGTM1* to extend experimentation till M95.
>>
>> On Mon, Sep 20, 2021 at 7:09 PM John Delaney  
>> wrote:
>>
>>> Hi Alex,
>>>
>>> Given that this is an x-origin OT, relatively high use might not be the 
 whole picture, so it would be great to know in addition how many 
 individual 
 teams have signed up to use the OT. Is dozens? Hundreds? Thousands?
>>>
>>> There are O(10) teams experimenting with the API.
>>>
>>> On Thu, Sep 16, 2021 at 3:25 PM Alex Russell  
>>> wrote:
>>>
 Discussed at today's API OWNERS, the relatively high use plus the 
 length of the ongoing trial adds a level of concern. Given that this is an 
 x-origin OT, relatively high use might not be the whole picture, so it 
 would be great to know in addition how many individual teams have signed 
 up 
 to use the OT. Is dozens? Hundreds? Thousands?

 Thanks.

 Best Regards,

 Alex

 On Wednesday, September 15, 2021 at 9:10:55 AM UTC-7 John Delaney wrote:

> Can you link to previous extension requests?
>>
> First extension: 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ
>   
> 
> Second extension: 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ
>
> M86 went to stable almost a year ago. What's the plan beyond M94? 
>> Assuming that the bug is fixed, will the experiment end there? Would you 
>> be 
>> able to share learnings from the trial? Are you planning to ship the API 
>> in 
>> its current form?
>
>
> We intend to end the experiment in 95. We are currently working on a 
> number of significant extensions/changes to the API including: aggregate 
> reports [1], support for views [2], and others. These support use-cases 
> which are not covered in the currently implemented "Event-level Click" 
> API. 
> We plan to start a "V2" experiment when this new functionality is 
> available 
> which we believe will produce new, different insights from developers. We 
> are not currently planning to ship Attribution Reporting without first 
> experimenting with those extensions.
>
> We can certainly share learnings with respect to the "Event-level 
> Click" API when this initial experiment concludes. 
>
> [1] 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
> [2] 
> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md
>
> On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss  
> wrote:
>
>>
>>
>> On Sat, Sep 11, 2021 at 12:01 AM John Delaney  
>> wrote:
>>
>>> *Contact emails*
>>> johni...@chromium.org, csharri...@chromium.org
>>>
>>> *Explainer*
>>>
>>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>>>
>>> *Design docs/spec*
>>> https://wicg.github.io/conversion-measurement-api/
>>>
>>> *Summary*
>>> This is a new API for measuring conversions (e.g. purchases) and 
>>> attributing them to clicked ads, without using cross-site persistent 
>>> identifiers like third party cookies.
>>>
>>> *Link to “Intent to Prototype” blink-dev discussion*
>>>
>>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>>>  
>>> *Link to “Intent to Experiment” blink-dev discussion*
>>>
>>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>>>
>>
>> Can you link to previous extension requests?
>>  
>>
>>>
>>> *Goals for experimentation*
>>> For a continuation of the trial, we still hope to see that the 
>>> measurement data made available through the API produces comparable 
>>> results 
>>> to existing click through measurement solutions on the web.
>>>
>>> We would like to see that Chrome 94 addresses the reporting issue 
>>> documented here:
>>>
>>> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>>>
>>
>> That seems reasonable.
>>  
>>
>>>
>>> *Experimental timeline*
>>> We'd like to extend the origin trial for one additional milestone, 
>>> continuing through M94. The experiment has been running since M86.
>>>
>>  
>> M86 went to stable almost a year ago. What's the plan beyond M94? 
>> Ass

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-21 Thread Chris Harrelson 
LGTM2

On Mon, Sep 20, 2021, 10:38 PM Yoav Weiss  wrote:

> O(10) sounds manageable in case we want to change course, so that reduces
> the burn-in risk.
> Since this is an atypical OT extension, approving it would require 3 LGTMs.
>
> *LGTM1* to extend experimentation till M95.
>
> On Mon, Sep 20, 2021 at 7:09 PM John Delaney 
> wrote:
>
>> Hi Alex,
>>
>> Given that this is an x-origin OT, relatively high use might not be the
>>> whole picture, so it would be great to know in addition how many individual
>>> teams have signed up to use the OT. Is dozens? Hundreds? Thousands?
>>
>> There are O(10) teams experimenting with the API.
>>
>> On Thu, Sep 16, 2021 at 3:25 PM Alex Russell 
>> wrote:
>>
>>> Discussed at today's API OWNERS, the relatively high use plus the length
>>> of the ongoing trial adds a level of concern. Given that this is an
>>> x-origin OT, relatively high use might not be the whole picture, so it
>>> would be great to know in addition how many individual teams have signed up
>>> to use the OT. Is dozens? Hundreds? Thousands?
>>>
>>> Thanks.
>>>
>>> Best Regards,
>>>
>>> Alex
>>>
>>> On Wednesday, September 15, 2021 at 9:10:55 AM UTC-7 John Delaney wrote:
>>>
 Can you link to previous extension requests?
>
 First extension:
 https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ
 
 Second extension:
 https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ

 M86 went to stable almost a year ago. What's the plan beyond M94?
> Assuming that the bug is fixed, will the experiment end there? Would you 
> be
> able to share learnings from the trial? Are you planning to ship the API 
> in
> its current form?


 We intend to end the experiment in 95. We are currently working on a
 number of significant extensions/changes to the API including: aggregate
 reports [1], support for views [2], and others. These support use-cases
 which are not covered in the currently implemented "Event-level Click" API.
 We plan to start a "V2" experiment when this new functionality is available
 which we believe will produce new, different insights from developers. We
 are not currently planning to ship Attribution Reporting without first
 experimenting with those extensions.

 We can certainly share learnings with respect to the "Event-level
 Click" API when this initial experiment concludes.

 [1]
 https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
 [2]
 https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md

 On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss 
 wrote:

>
>
> On Sat, Sep 11, 2021 at 12:01 AM John Delaney 
> wrote:
>
>> *Contact emails*
>> johni...@chromium.org, csharri...@chromium.org
>>
>> *Explainer*
>>
>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>>
>> *Design docs/spec*
>> https://wicg.github.io/conversion-measurement-api/
>>
>> *Summary*
>> This is a new API for measuring conversions (e.g. purchases) and
>> attributing them to clicked ads, without using cross-site persistent
>> identifiers like third party cookies.
>>
>> *Link to “Intent to Prototype” blink-dev discussion*
>>
>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>>
>> *Link to “Intent to Experiment” blink-dev discussion*
>>
>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>>
>
> Can you link to previous extension requests?
>
>
>>
>> *Goals for experimentation*
>> For a continuation of the trial, we still hope to see that the
>> measurement data made available through the API produces comparable 
>> results
>> to existing click through measurement solutions on the web.
>>
>> We would like to see that Chrome 94 addresses the reporting issue
>> documented here:
>>
>> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>>
>
> That seems reasonable.
>
>
>>
>> *Experimental timeline*
>> We'd like to extend the origin trial for one additional milestone,
>> continuing through M94. The experiment has been running since M86.
>>
>
> M86 went to stable almost a year ago. What's the plan beyond M94?
> Assuming that the bug is fixed, will the experiment end there? Would you 
> be
> able to share learnings from the trial? Are you planning to ship the API 
> in
> its current form?
>
>
>> *Any risks when the experiment finishes?*
>> This API is only additive and does not affect any

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-20 Thread Yoav Weiss 
O(10) sounds manageable in case we want to change course, so that reduces
the burn-in risk.
Since this is an atypical OT extension, approving it would require 3 LGTMs.

*LGTM1* to extend experimentation till M95.

On Mon, Sep 20, 2021 at 7:09 PM John Delaney  wrote:

> Hi Alex,
>
> Given that this is an x-origin OT, relatively high use might not be the
>> whole picture, so it would be great to know in addition how many individual
>> teams have signed up to use the OT. Is dozens? Hundreds? Thousands?
>
> There are O(10) teams experimenting with the API.
>
> On Thu, Sep 16, 2021 at 3:25 PM Alex Russell 
> wrote:
>
>> Discussed at today's API OWNERS, the relatively high use plus the length
>> of the ongoing trial adds a level of concern. Given that this is an
>> x-origin OT, relatively high use might not be the whole picture, so it
>> would be great to know in addition how many individual teams have signed up
>> to use the OT. Is dozens? Hundreds? Thousands?
>>
>> Thanks.
>>
>> Best Regards,
>>
>> Alex
>>
>> On Wednesday, September 15, 2021 at 9:10:55 AM UTC-7 John Delaney wrote:
>>
>>> Can you link to previous extension requests?

>>> First extension:
>>> https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ
>>> 
>>> Second extension:
>>> https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ
>>>
>>> M86 went to stable almost a year ago. What's the plan beyond M94?
 Assuming that the bug is fixed, will the experiment end there? Would you be
 able to share learnings from the trial? Are you planning to ship the API in
 its current form?
>>>
>>>
>>> We intend to end the experiment in 95. We are currently working on a
>>> number of significant extensions/changes to the API including: aggregate
>>> reports [1], support for views [2], and others. These support use-cases
>>> which are not covered in the currently implemented "Event-level Click" API.
>>> We plan to start a "V2" experiment when this new functionality is available
>>> which we believe will produce new, different insights from developers. We
>>> are not currently planning to ship Attribution Reporting without first
>>> experimenting with those extensions.
>>>
>>> We can certainly share learnings with respect to the "Event-level Click"
>>> API when this initial experiment concludes.
>>>
>>> [1]
>>> https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
>>> [2]
>>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md
>>>
>>> On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss 
>>> wrote:
>>>


 On Sat, Sep 11, 2021 at 12:01 AM John Delaney 
 wrote:

> *Contact emails*
> johni...@chromium.org, csharri...@chromium.org
>
> *Explainer*
>
> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>
> *Design docs/spec*
> https://wicg.github.io/conversion-measurement-api/
>
> *Summary*
> This is a new API for measuring conversions (e.g. purchases) and
> attributing them to clicked ads, without using cross-site persistent
> identifiers like third party cookies.
>
> *Link to “Intent to Prototype” blink-dev discussion*
>
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>
> *Link to “Intent to Experiment” blink-dev discussion*
>
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>

 Can you link to previous extension requests?


>
> *Goals for experimentation*
> For a continuation of the trial, we still hope to see that the
> measurement data made available through the API produces comparable 
> results
> to existing click through measurement solutions on the web.
>
> We would like to see that Chrome 94 addresses the reporting issue
> documented here:
>
> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>

 That seems reasonable.


>
> *Experimental timeline*
> We'd like to extend the origin trial for one additional milestone,
> continuing through M94. The experiment has been running since M86.
>

 M86 went to stable almost a year ago. What's the plan beyond M94?
 Assuming that the bug is fixed, will the experiment end there? Would you be
 able to share learnings from the trial? Are you planning to ship the API in
 its current form?


> *Any risks when the experiment finishes?*
> This API is only additive and does not affect any existing state.
> Current conversion measurement solutions will work as expected with and
> without the API, so we don’t believe there are any risks.
>
> All web-visible names of the API were modified in Ch

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-20 Thread John Delaney 
Hi Alex,

Given that this is an x-origin OT, relatively high use might not be the
> whole picture, so it would be great to know in addition how many individual
> teams have signed up to use the OT. Is dozens? Hundreds? Thousands?

There are O(10) teams experimenting with the API.

On Thu, Sep 16, 2021 at 3:25 PM Alex Russell 
wrote:

> Discussed at today's API OWNERS, the relatively high use plus the length
> of the ongoing trial adds a level of concern. Given that this is an
> x-origin OT, relatively high use might not be the whole picture, so it
> would be great to know in addition how many individual teams have signed up
> to use the OT. Is dozens? Hundreds? Thousands?
>
> Thanks.
>
> Best Regards,
>
> Alex
>
> On Wednesday, September 15, 2021 at 9:10:55 AM UTC-7 John Delaney wrote:
>
>> Can you link to previous extension requests?
>>>
>> First extension:
>> https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ
>> 
>> Second extension:
>> https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ
>>
>> M86 went to stable almost a year ago. What's the plan beyond M94?
>>> Assuming that the bug is fixed, will the experiment end there? Would you be
>>> able to share learnings from the trial? Are you planning to ship the API in
>>> its current form?
>>
>>
>> We intend to end the experiment in 95. We are currently working on a
>> number of significant extensions/changes to the API including: aggregate
>> reports [1], support for views [2], and others. These support use-cases
>> which are not covered in the currently implemented "Event-level Click" API.
>> We plan to start a "V2" experiment when this new functionality is available
>> which we believe will produce new, different insights from developers. We
>> are not currently planning to ship Attribution Reporting without first
>> experimenting with those extensions.
>>
>> We can certainly share learnings with respect to the "Event-level Click"
>> API when this initial experiment concludes.
>>
>> [1]
>> https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
>> [2]
>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md
>>
>> On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss 
>> wrote:
>>
>>>
>>>
>>> On Sat, Sep 11, 2021 at 12:01 AM John Delaney 
>>> wrote:
>>>
 *Contact emails*
 johni...@chromium.org, csharri...@chromium.org

 *Explainer*

 https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md

 *Design docs/spec*
 https://wicg.github.io/conversion-measurement-api/

 *Summary*
 This is a new API for measuring conversions (e.g. purchases) and
 attributing them to clicked ads, without using cross-site persistent
 identifiers like third party cookies.

 *Link to “Intent to Prototype” blink-dev discussion*

 https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ

 *Link to “Intent to Experiment” blink-dev discussion*

 https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ

>>>
>>> Can you link to previous extension requests?
>>>
>>>

 *Goals for experimentation*
 For a continuation of the trial, we still hope to see that the
 measurement data made available through the API produces comparable results
 to existing click through measurement solutions on the web.

 We would like to see that Chrome 94 addresses the reporting issue
 documented here:

 https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ

>>>
>>> That seems reasonable.
>>>
>>>

 *Experimental timeline*
 We'd like to extend the origin trial for one additional milestone,
 continuing through M94. The experiment has been running since M86.

>>>
>>> M86 went to stable almost a year ago. What's the plan beyond M94?
>>> Assuming that the bug is fixed, will the experiment end there? Would you be
>>> able to share learnings from the trial? Are you planning to ship the API in
>>> its current form?
>>>
>>>
 *Any risks when the experiment finishes?*
 This API is only additive and does not affect any existing state.
 Current conversion measurement solutions will work as expected with and
 without the API, so we don’t believe there are any risks.

 All web-visible names of the API were modified in Chrome 92 which
 mitigated burn in risk as developers needed to update their implementation
 to continue experimenting.

 We have previously requested an exception to the .5% page load usage
 limit for Origin Trials. Page load usage is currently around .92% for the
 feature.
 https://chromestatus.com/metrics/feature/timeline/popularity/3365

 *Reason this experiment i

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-16 Thread Alex Russell 
Discussed at today's API OWNERS, the relatively high use plus the length of 
the ongoing trial adds a level of concern. Given that this is an x-origin 
OT, relatively high use might not be the whole picture, so it would be 
great to know in addition how many individual teams have signed up to use 
the OT. Is dozens? Hundreds? Thousands?

Thanks.

Best Regards,

Alex

On Wednesday, September 15, 2021 at 9:10:55 AM UTC-7 John Delaney wrote:

> Can you link to previous extension requests?
>>
> First extension: 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ
>   
> 
> Second extension: 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ
>
> M86 went to stable almost a year ago. What's the plan beyond M94? Assuming 
>> that the bug is fixed, will the experiment end there? Would you be able to 
>> share learnings from the trial? Are you planning to ship the API in its 
>> current form?
>
>
> We intend to end the experiment in 95. We are currently working on a 
> number of significant extensions/changes to the API including: aggregate 
> reports [1], support for views [2], and others. These support use-cases 
> which are not covered in the currently implemented "Event-level Click" API. 
> We plan to start a "V2" experiment when this new functionality is available 
> which we believe will produce new, different insights from developers. We 
> are not currently planning to ship Attribution Reporting without first 
> experimenting with those extensions.
>
> We can certainly share learnings with respect to the "Event-level Click" 
> API when this initial experiment concludes. 
>
> [1] 
> https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
> [2] 
> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md
>
> On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss  
> wrote:
>
>>
>>
>> On Sat, Sep 11, 2021 at 12:01 AM John Delaney  
>> wrote:
>>
>>> *Contact emails*
>>> johni...@chromium.org, csharri...@chromium.org
>>>
>>> *Explainer*
>>>
>>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>>>
>>> *Design docs/spec*
>>> https://wicg.github.io/conversion-measurement-api/
>>>
>>> *Summary*
>>> This is a new API for measuring conversions (e.g. purchases) and 
>>> attributing them to clicked ads, without using cross-site persistent 
>>> identifiers like third party cookies.
>>>
>>> *Link to “Intent to Prototype” blink-dev discussion*
>>>
>>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>>>  
>>> *Link to “Intent to Experiment” blink-dev discussion*
>>>
>>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>>>
>>
>> Can you link to previous extension requests?
>>  
>>
>>>
>>> *Goals for experimentation*
>>> For a continuation of the trial, we still hope to see that the 
>>> measurement data made available through the API produces comparable results 
>>> to existing click through measurement solutions on the web.
>>>
>>> We would like to see that Chrome 94 addresses the reporting issue 
>>> documented here:
>>>
>>> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>>>
>>
>> That seems reasonable.
>>  
>>
>>>
>>> *Experimental timeline*
>>> We'd like to extend the origin trial for one additional milestone, 
>>> continuing through M94. The experiment has been running since M86.
>>>
>>  
>> M86 went to stable almost a year ago. What's the plan beyond M94? 
>> Assuming that the bug is fixed, will the experiment end there? Would you be 
>> able to share learnings from the trial? Are you planning to ship the API in 
>> its current form?
>>
>>
>>> *Any risks when the experiment finishes?*
>>> This API is only additive and does not affect any existing state. 
>>> Current conversion measurement solutions will work as expected with and 
>>> without the API, so we don’t believe there are any risks.
>>>
>>> All web-visible names of the API were modified in Chrome 92 which 
>>> mitigated burn in risk as developers needed to update their implementation 
>>> to continue experimenting.
>>>
>>> We have previously requested an exception to the .5% page load usage 
>>> limit for Origin Trials. Page load usage is currently around .92% for the 
>>> feature.
>>> https://chromestatus.com/metrics/feature/timeline/popularity/3365
>>>
>>> *Reason this experiment is being extended*
>>> There are a number of new internal partners experimenting with the API. 
>>> Chrome 93 has a shorter than normal Stable lifetime, with Stable starting 
>>> on Aug 31st, and 93 OT tokens expiring on September 14th, only giving two 
>>> weeks to experiment. We would like to allow a bit longer for 
>>> experimentation.
>>>
>>> We recently received some great feedback from a partner touching on d

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-15 Thread John Delaney 
>
> Can you link to previous extension requests?
>
First extension:
https://groups.google.com/a/chromium.org/g/blink-dev/c/C0P7ePjITJQ/m/ZYFmn30SDQAJ

Second extension:
https://groups.google.com/a/chromium.org/g/blink-dev/c/xCWP1ltlAgw/m/acwN0GIRBQAJ

M86 went to stable almost a year ago. What's the plan beyond M94? Assuming
> that the bug is fixed, will the experiment end there? Would you be able to
> share learnings from the trial? Are you planning to ship the API in its
> current form?


We intend to end the experiment in 95. We are currently working on a number
of significant extensions/changes to the API including: aggregate reports
[1], support for views [2], and others. These support use-cases which are
not covered in the currently implemented "Event-level Click" API. We plan
to start a "V2" experiment when this new functionality is available which
we believe will produce new, different insights from developers. We are not
currently planning to ship Attribution Reporting without first
experimenting with those extensions.

We can certainly share learnings with respect to the "Event-level Click"
API when this initial experiment concludes.

[1]
https://groups.google.com/a/chromium.org/g/blink-dev/c/2zA5-TuVSkA/m/GVqUUkhfAwAJ
[2]
https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_views.md

On Tue, Sep 14, 2021 at 12:06 AM Yoav Weiss  wrote:

>
>
> On Sat, Sep 11, 2021 at 12:01 AM John Delaney 
> wrote:
>
>> *Contact emails*
>> johni...@chromium.org, csharri...@chromium.org
>>
>> *Explainer*
>>
>> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>>
>> *Design docs/spec*
>> https://wicg.github.io/conversion-measurement-api/
>>
>> *Summary*
>> This is a new API for measuring conversions (e.g. purchases) and
>> attributing them to clicked ads, without using cross-site persistent
>> identifiers like third party cookies.
>>
>> *Link to “Intent to Prototype” blink-dev discussion*
>>
>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>>
>> *Link to “Intent to Experiment” blink-dev discussion*
>>
>> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>>
>
> Can you link to previous extension requests?
>
>
>>
>> *Goals for experimentation*
>> For a continuation of the trial, we still hope to see that the
>> measurement data made available through the API produces comparable results
>> to existing click through measurement solutions on the web.
>>
>> We would like to see that Chrome 94 addresses the reporting issue
>> documented here:
>>
>> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>>
>
> That seems reasonable.
>
>
>>
>> *Experimental timeline*
>> We'd like to extend the origin trial for one additional milestone,
>> continuing through M94. The experiment has been running since M86.
>>
>
> M86 went to stable almost a year ago. What's the plan beyond M94? Assuming
> that the bug is fixed, will the experiment end there? Would you be able to
> share learnings from the trial? Are you planning to ship the API in its
> current form?
>
>
>> *Any risks when the experiment finishes?*
>> This API is only additive and does not affect any existing state. Current
>> conversion measurement solutions will work as expected with and without the
>> API, so we don’t believe there are any risks.
>>
>> All web-visible names of the API were modified in Chrome 92 which
>> mitigated burn in risk as developers needed to update their implementation
>> to continue experimenting.
>>
>> We have previously requested an exception to the .5% page load usage
>> limit for Origin Trials. Page load usage is currently around .92% for the
>> feature.
>> https://chromestatus.com/metrics/feature/timeline/popularity/3365
>>
>> *Reason this experiment is being extended*
>> There are a number of new internal partners experimenting with the API.
>> Chrome 93 has a shorter than normal Stable lifetime, with Stable starting
>> on Aug 31st, and 93 OT tokens expiring on September 14th, only giving two
>> weeks to experiment. We would like to allow a bit longer for
>> experimentation.
>>
>> We recently received some great feedback from a partner touching on data
>> loss due to network errors:
>> https://github.com/WICG/conversion-measurement-api/issues/201. Chrome 94
>> includes a fix for this issue, which we would like to verify works.
>>
>> *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://chromestatus.com/feature/6412002824028160
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "blink-dev" group.
>> To unsubscribe from this group and s

Re: [blink-dev] Intent to Extend Origin Trial: Conversion Measurement API (Attribution Reporting API)

2021-09-13 Thread Yoav Weiss 
On Sat, Sep 11, 2021 at 12:01 AM John Delaney  wrote:

> *Contact emails*
> johni...@chromium.org, csharri...@chromium.org
>
> *Explainer*
>
> https://github.com/WICG/conversion-measurement-api/blob/main/event_attribution_reporting_clicks.md
>
> *Design docs/spec*
> https://wicg.github.io/conversion-measurement-api/
>
> *Summary*
> This is a new API for measuring conversions (e.g. purchases) and
> attributing them to clicked ads, without using cross-site persistent
> identifiers like third party cookies.
>
> *Link to “Intent to Prototype” blink-dev discussion*
>
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/7B0ldtZR_68/GjLBu0n4DgAJ
>
> *Link to “Intent to Experiment” blink-dev discussion*
>
> https://groups.google.com/a/chromium.org/d/msg/blink-dev/Ib9-tDFitns/Sm9RXTNaBwAJ
>

Can you link to previous extension requests?


>
> *Goals for experimentation*
> For a continuation of the trial, we still hope to see that the measurement
> data made available through the API produces comparable results to existing
> click through measurement solutions on the web.
>
> We would like to see that Chrome 94 addresses the reporting issue
> documented here:
>
> https://groups.google.com/a/chromium.org/g/attribution-reporting-api-dev/c/2KKqWG70w7U/m/7M8fWEbmAgAJ
>

That seems reasonable.


>
> *Experimental timeline*
> We'd like to extend the origin trial for one additional milestone,
> continuing through M94. The experiment has been running since M86.
>

M86 went to stable almost a year ago. What's the plan beyond M94? Assuming
that the bug is fixed, will the experiment end there? Would you be able to
share learnings from the trial? Are you planning to ship the API in its
current form?


> *Any risks when the experiment finishes?*
> This API is only additive and does not affect any existing state. Current
> conversion measurement solutions will work as expected with and without the
> API, so we don’t believe there are any risks.
>
> All web-visible names of the API were modified in Chrome 92 which
> mitigated burn in risk as developers needed to update their implementation
> to continue experimenting.
>
> We have previously requested an exception to the .5% page load usage limit
> for Origin Trials. Page load usage is currently around .92% for the feature.
> https://chromestatus.com/metrics/feature/timeline/popularity/3365
>
> *Reason this experiment is being extended*
> There are a number of new internal partners experimenting with the API.
> Chrome 93 has a shorter than normal Stable lifetime, with Stable starting
> on Aug 31st, and 93 OT tokens expiring on September 14th, only giving two
> weeks to experiment. We would like to allow a bit longer for
> experimentation.
>
> We recently received some great feedback from a partner touching on data
> loss due to network errors:
> https://github.com/WICG/conversion-measurement-api/issues/201. Chrome 94
> includes a fix for this issue, which we would like to verify works.
>
> *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://chromestatus.com/feature/6412002824028160
>
> --
> 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/b863ba43-5b38-40d2-95d2-12f64a10fca7n%40chromium.org
> 
> .
>

-- 
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/CAL5BFfVApAbO_V__ZA97oHOXj%2BH7ta369jnWK%3DHod%2B0ca_0UXQ%40mail.gmail.com.