Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Daisuke Enomoto]

Hi Joe,

This feature is shipping in 104. I updated the chromestatus entry.

Thanks for pinging.

On Tue, Jun 7, 2022 at 12:18 AM Joe Medley  wrote:

> Is this shipping in 105? Please add that to the status entry.
> Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com |
>  816-678-7195
> *If an API's not documented it doesn't exist.*
>
>
> On Wed, Jun 1, 2022 at 2:03 AM Hayato Ito  wrote:
>
>> Thanks all, for your support and the LGTMs.
>>
>> I appreciate the API owners, the Mozilla folks and the others, who
>> reviewed this feature carefully and gave us the advice to make this feature
>> interoperable and be an important step on future improvements on the Web
>> Platform.
>>
>> I'll continue to invest in standards. Thanks!
>>
>> On Wed, Jun 1, 2022 at 2:11 PM Yoav Weiss  wrote:
>>
>>> LGTM3
>>>
>>> I was similarly convinced by the team's efforts and by various folks
>>> chiming in that this has a good chance of reaching eventual interop on its
>>> own. The use cases this tackles span origin isolation, delivery and
>>> performance (despite the lack of cache awareness in the current version),
>>> and I'm hopeful that this would be an important stepping stone on which
>>> future improvements will be built. Thanks!!
>>>
>>> On Wed, Jun 1, 2022 at 4:15 AM Chris Harrelson 
>>> wrote:
>>>
 LGTM2

 Web bundles team: thank you for your thoroughness in responding to
 concerns raised in the TAG review, Mozilla standards position, and various
 offline conversations. I also appreciate those who weighed in on the thread
 with additional sites that would use this feature, and the multiple use
 cases mentioned.

 At first, I was concerned about the risk of not achieving eventual
 interoperability, because of the limited data from experimenting sites and
 mixed or possibly negative signals from other browser vendors. The
 additional data mentioned above, on top of the existing ads-related use
 case from the original intent, and coupled with my belief that the team
 will continue to invest in standards in this area, leads me to conclude
 that this feature is worth shipping now.


 On Wed, May 25, 2022 at 8:37 AM Alex Russell 
 wrote:

> There are high-scale operational challenges to distributing content
> optimally in many scenarios (e.g., Store packaged first-run or
> out-of-the-box-experiences) that need more than what Service Workers alone
> can provide. We are positive about this technology because it can help
> address those problems and open up new product opportunities (that we
> cannot say more about).
>
> Best Regards,
>
> Alex
>
> On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com
> wrote:
>
>> >>  You could have used a service worker and cache things in order to
>> get the same thing, combined with progressive web applications for
>> installability.
>>
>> This is true, but I think ergonomics of an encapsulated
>> page/application make it easier to reason about for very large
>> applications. Maintaining the service worker and the cache at scale can
>> become a barrier to entry for some developers and applications.
>>
>> On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:
>>
>>> Not that I am against, but does this unlock previously locked
>>> opportunities in the specific examples you just mentioned?
>>> You could have used a service worker and cache things in order to
>>> get the same thing, combined with progressive web applications for
>>> installability.
>>>
>>> ☆*PhistucK*
>>>
>>>
>>> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
>>> blin...@chromium.org> wrote:
>>>
 Microsoft has a strong interest in seeing this feature ship. We
 believe that sub-resource bundling is opening the door to a new way of
 shipping and delivering offline web applications, changing the 
 traditional
 definition of “web application”.

 Here are some ways Microsoft products can take advantage of this:

 *PowerApps*

 Microsoft PowerApps allows a developer to author an application and
 deploy to iOS, Android, and the web. The first two platforms allow
 applications to be deployed and used when the device is offline, but 
 the
 latter is currently not “installable” on the device. Web bundling could
 unlock the capability for a web application to be “installed” on a 
 device
 to operate offline.

 *Office Online*

 Office productivity web applications are a perfect example of
 applications that could benefit from a packaged bundle of application
 resources. Combined with local storage APIs, this could help developers
 reach communities that have little to no 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

['Joe Medley' via blink-dev]

Is this shipping in 105? Please add that to the status entry.
Joe Medley | Technical Writer, Chrome DevRel | jmed...@google.com |
 816-678-7195
*If an API's not documented it doesn't exist.*


On Wed, Jun 1, 2022 at 2:03 AM Hayato Ito  wrote:

> Thanks all, for your support and the LGTMs.
>
> I appreciate the API owners, the Mozilla folks and the others, who
> reviewed this feature carefully and gave us the advice to make this feature
> interoperable and be an important step on future improvements on the Web
> Platform.
>
> I'll continue to invest in standards. Thanks!
>
> On Wed, Jun 1, 2022 at 2:11 PM Yoav Weiss  wrote:
>
>> LGTM3
>>
>> I was similarly convinced by the team's efforts and by various folks
>> chiming in that this has a good chance of reaching eventual interop on its
>> own. The use cases this tackles span origin isolation, delivery and
>> performance (despite the lack of cache awareness in the current version),
>> and I'm hopeful that this would be an important stepping stone on which
>> future improvements will be built. Thanks!!
>>
>> On Wed, Jun 1, 2022 at 4:15 AM Chris Harrelson 
>> wrote:
>>
>>> LGTM2
>>>
>>> Web bundles team: thank you for your thoroughness in responding to
>>> concerns raised in the TAG review, Mozilla standards position, and various
>>> offline conversations. I also appreciate those who weighed in on the thread
>>> with additional sites that would use this feature, and the multiple use
>>> cases mentioned.
>>>
>>> At first, I was concerned about the risk of not achieving eventual
>>> interoperability, because of the limited data from experimenting sites and
>>> mixed or possibly negative signals from other browser vendors. The
>>> additional data mentioned above, on top of the existing ads-related use
>>> case from the original intent, and coupled with my belief that the team
>>> will continue to invest in standards in this area, leads me to conclude
>>> that this feature is worth shipping now.
>>>
>>>
>>> On Wed, May 25, 2022 at 8:37 AM Alex Russell 
>>> wrote:
>>>
 There are high-scale operational challenges to distributing content
 optimally in many scenarios (e.g., Store packaged first-run or
 out-of-the-box-experiences) that need more than what Service Workers alone
 can provide. We are positive about this technology because it can help
 address those problems and open up new product opportunities (that we
 cannot say more about).

 Best Regards,

 Alex

 On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com
 wrote:

> >>  You could have used a service worker and cache things in order to
> get the same thing, combined with progressive web applications for
> installability.
>
> This is true, but I think ergonomics of an encapsulated
> page/application make it easier to reason about for very large
> applications. Maintaining the service worker and the cache at scale can
> become a barrier to entry for some developers and applications.
>
> On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:
>
>> Not that I am against, but does this unlock previously locked
>> opportunities in the specific examples you just mentioned?
>> You could have used a service worker and cache things in order to get
>> the same thing, combined with progressive web applications for
>> installability.
>>
>> ☆*PhistucK*
>>
>>
>> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
>> blin...@chromium.org> wrote:
>>
>>> Microsoft has a strong interest in seeing this feature ship. We
>>> believe that sub-resource bundling is opening the door to a new way of
>>> shipping and delivering offline web applications, changing the 
>>> traditional
>>> definition of “web application”.
>>>
>>> Here are some ways Microsoft products can take advantage of this:
>>>
>>> *PowerApps*
>>>
>>> Microsoft PowerApps allows a developer to author an application and
>>> deploy to iOS, Android, and the web. The first two platforms allow
>>> applications to be deployed and used when the device is offline, but the
>>> latter is currently not “installable” on the device. Web bundling could
>>> unlock the capability for a web application to be “installed” on a 
>>> device
>>> to operate offline.
>>>
>>> *Office Online*
>>>
>>> Office productivity web applications are a perfect example of
>>> applications that could benefit from a packaged bundle of application
>>> resources. Combined with local storage APIs, this could help developers
>>> reach communities that have little to no network connectivity.
>>>
>>>
>>>
>>> While there have been concerns brought up by the community, we
>>> welcome the opportunity to collaborate on addressing these issues in the
>>> next iteration of this project. We feel confident we can resolve them 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Hayato Ito]

Thanks all, for your support and the LGTMs.

I appreciate the API owners, the Mozilla folks and the others, who reviewed
this feature carefully and gave us the advice to make this feature
interoperable and be an important step on future improvements on the Web
Platform.

I'll continue to invest in standards. Thanks!

On Wed, Jun 1, 2022 at 2:11 PM Yoav Weiss  wrote:

> LGTM3
>
> I was similarly convinced by the team's efforts and by various folks
> chiming in that this has a good chance of reaching eventual interop on its
> own. The use cases this tackles span origin isolation, delivery and
> performance (despite the lack of cache awareness in the current version),
> and I'm hopeful that this would be an important stepping stone on which
> future improvements will be built. Thanks!!
>
> On Wed, Jun 1, 2022 at 4:15 AM Chris Harrelson 
> wrote:
>
>> LGTM2
>>
>> Web bundles team: thank you for your thoroughness in responding to
>> concerns raised in the TAG review, Mozilla standards position, and various
>> offline conversations. I also appreciate those who weighed in on the thread
>> with additional sites that would use this feature, and the multiple use
>> cases mentioned.
>>
>> At first, I was concerned about the risk of not achieving eventual
>> interoperability, because of the limited data from experimenting sites and
>> mixed or possibly negative signals from other browser vendors. The
>> additional data mentioned above, on top of the existing ads-related use
>> case from the original intent, and coupled with my belief that the team
>> will continue to invest in standards in this area, leads me to conclude
>> that this feature is worth shipping now.
>>
>>
>> On Wed, May 25, 2022 at 8:37 AM Alex Russell 
>> wrote:
>>
>>> There are high-scale operational challenges to distributing content
>>> optimally in many scenarios (e.g., Store packaged first-run or
>>> out-of-the-box-experiences) that need more than what Service Workers alone
>>> can provide. We are positive about this technology because it can help
>>> address those problems and open up new product opportunities (that we
>>> cannot say more about).
>>>
>>> Best Regards,
>>>
>>> Alex
>>>
>>> On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com
>>> wrote:
>>>
 >>  You could have used a service worker and cache things in order to
 get the same thing, combined with progressive web applications for
 installability.

 This is true, but I think ergonomics of an encapsulated
 page/application make it easier to reason about for very large
 applications. Maintaining the service worker and the cache at scale can
 become a barrier to entry for some developers and applications.

 On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:

> Not that I am against, but does this unlock previously locked
> opportunities in the specific examples you just mentioned?
> You could have used a service worker and cache things in order to get
> the same thing, combined with progressive web applications for
> installability.
>
> ☆*PhistucK*
>
>
> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
> blin...@chromium.org> wrote:
>
>> Microsoft has a strong interest in seeing this feature ship. We
>> believe that sub-resource bundling is opening the door to a new way of
>> shipping and delivering offline web applications, changing the 
>> traditional
>> definition of “web application”.
>>
>> Here are some ways Microsoft products can take advantage of this:
>>
>> *PowerApps*
>>
>> Microsoft PowerApps allows a developer to author an application and
>> deploy to iOS, Android, and the web. The first two platforms allow
>> applications to be deployed and used when the device is offline, but the
>> latter is currently not “installable” on the device. Web bundling could
>> unlock the capability for a web application to be “installed” on a device
>> to operate offline.
>>
>> *Office Online*
>>
>> Office productivity web applications are a perfect example of
>> applications that could benefit from a packaged bundle of application
>> resources. Combined with local storage APIs, this could help developers
>> reach communities that have little to no network connectivity.
>>
>>
>>
>> While there have been concerns brought up by the community, we
>> welcome the opportunity to collaborate on addressing these issues in the
>> next iteration of this project. We feel confident we can resolve them in 
>> a
>> way that preserves the integrity of the open web.
>>
>>
>> Ben Mathwig
>> Senior Product Manager
>> Microsoft Edge Web Platform
>>
>>
>> On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org
>> wrote:
>>
>>> Hello
>>>
>>> I am sharing the feedback from the Origin Trial with 12 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Yoav Weiss]

LGTM3

I was similarly convinced by the team's efforts and by various folks
chiming in that this has a good chance of reaching eventual interop on its
own. The use cases this tackles span origin isolation, delivery and
performance (despite the lack of cache awareness in the current version),
and I'm hopeful that this would be an important stepping stone on which
future improvements will be built. Thanks!!

On Wed, Jun 1, 2022 at 4:15 AM Chris Harrelson 
wrote:

> LGTM2
>
> Web bundles team: thank you for your thoroughness in responding to
> concerns raised in the TAG review, Mozilla standards position, and various
> offline conversations. I also appreciate those who weighed in on the thread
> with additional sites that would use this feature, and the multiple use
> cases mentioned.
>
> At first, I was concerned about the risk of not achieving eventual
> interoperability, because of the limited data from experimenting sites and
> mixed or possibly negative signals from other browser vendors. The
> additional data mentioned above, on top of the existing ads-related use
> case from the original intent, and coupled with my belief that the team
> will continue to invest in standards in this area, leads me to conclude
> that this feature is worth shipping now.
>
>
> On Wed, May 25, 2022 at 8:37 AM Alex Russell 
> wrote:
>
>> There are high-scale operational challenges to distributing content
>> optimally in many scenarios (e.g., Store packaged first-run or
>> out-of-the-box-experiences) that need more than what Service Workers alone
>> can provide. We are positive about this technology because it can help
>> address those problems and open up new product opportunities (that we
>> cannot say more about).
>>
>> Best Regards,
>>
>> Alex
>>
>> On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com
>> wrote:
>>
>>> >>  You could have used a service worker and cache things in order to
>>> get the same thing, combined with progressive web applications for
>>> installability.
>>>
>>> This is true, but I think ergonomics of an encapsulated page/application
>>> make it easier to reason about for very large applications. Maintaining the
>>> service worker and the cache at scale can become a barrier to entry for
>>> some developers and applications.
>>>
>>> On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:
>>>
 Not that I am against, but does this unlock previously locked
 opportunities in the specific examples you just mentioned?
 You could have used a service worker and cache things in order to get
 the same thing, combined with progressive web applications for
 installability.

 ☆*PhistucK*


 On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
 blin...@chromium.org> wrote:

> Microsoft has a strong interest in seeing this feature ship. We
> believe that sub-resource bundling is opening the door to a new way of
> shipping and delivering offline web applications, changing the traditional
> definition of “web application”.
>
> Here are some ways Microsoft products can take advantage of this:
>
> *PowerApps*
>
> Microsoft PowerApps allows a developer to author an application and
> deploy to iOS, Android, and the web. The first two platforms allow
> applications to be deployed and used when the device is offline, but the
> latter is currently not “installable” on the device. Web bundling could
> unlock the capability for a web application to be “installed” on a device
> to operate offline.
>
> *Office Online*
>
> Office productivity web applications are a perfect example of
> applications that could benefit from a packaged bundle of application
> resources. Combined with local storage APIs, this could help developers
> reach communities that have little to no network connectivity.
>
>
>
> While there have been concerns brought up by the community, we welcome
> the opportunity to collaborate on addressing these issues in the next
> iteration of this project. We feel confident we can resolve them in a way
> that preserves the integrity of the open web.
>
>
> Ben Mathwig
> Senior Product Manager
> Microsoft Edge Web Platform
>
>
> On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org
> wrote:
>
>> Hello
>>
>> I am sharing the feedback from the Origin Trial with 12 participants:
>>
>>-
>>
>>10 of them responded "Extremely likely" to "How likely are you to
>>keep using this feature?"
>>-
>>
>>Qualitative feedback:
>>
>>
>>-
>>
>>"I'm very excited about the CSP interpretation change rolling out
>>in M92"
>>-
>>
>>"looking forward to the CSP fix!"
>>-
>>
>>"I'm very glad you're working on this!"
>>-
>>
>>"This feature is great! 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Chris Harrelson]

LGTM2

Web bundles team: thank you for your thoroughness in responding to concerns
raised in the TAG review, Mozilla standards position, and various offline
conversations. I also appreciate those who weighed in on the thread with
additional sites that would use this feature, and the multiple use cases
mentioned.

At first, I was concerned about the risk of not achieving eventual
interoperability, because of the limited data from experimenting sites and
mixed or possibly negative signals from other browser vendors. The
additional data mentioned above, on top of the existing ads-related use
case from the original intent, and coupled with my belief that the team
will continue to invest in standards in this area, leads me to conclude
that this feature is worth shipping now.


On Wed, May 25, 2022 at 8:37 AM Alex Russell 
wrote:

> There are high-scale operational challenges to distributing content
> optimally in many scenarios (e.g., Store packaged first-run or
> out-of-the-box-experiences) that need more than what Service Workers alone
> can provide. We are positive about this technology because it can help
> address those problems and open up new product opportunities (that we
> cannot say more about).
>
> Best Regards,
>
> Alex
>
> On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com wrote:
>
>> >>  You could have used a service worker and cache things in order to get
>> the same thing, combined with progressive web applications for
>> installability.
>>
>> This is true, but I think ergonomics of an encapsulated page/application
>> make it easier to reason about for very large applications. Maintaining the
>> service worker and the cache at scale can become a barrier to entry for
>> some developers and applications.
>>
>> On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:
>>
>>> Not that I am against, but does this unlock previously locked
>>> opportunities in the specific examples you just mentioned?
>>> You could have used a service worker and cache things in order to get
>>> the same thing, combined with progressive web applications for
>>> installability.
>>>
>>> ☆*PhistucK*
>>>
>>>
>>> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
>>> blin...@chromium.org> wrote:
>>>
 Microsoft has a strong interest in seeing this feature ship. We believe
 that sub-resource bundling is opening the door to a new way of shipping and
 delivering offline web applications, changing the traditional definition of
 “web application”.

 Here are some ways Microsoft products can take advantage of this:

 *PowerApps*

 Microsoft PowerApps allows a developer to author an application and
 deploy to iOS, Android, and the web. The first two platforms allow
 applications to be deployed and used when the device is offline, but the
 latter is currently not “installable” on the device. Web bundling could
 unlock the capability for a web application to be “installed” on a device
 to operate offline.

 *Office Online*

 Office productivity web applications are a perfect example of
 applications that could benefit from a packaged bundle of application
 resources. Combined with local storage APIs, this could help developers
 reach communities that have little to no network connectivity.



 While there have been concerns brought up by the community, we welcome
 the opportunity to collaborate on addressing these issues in the next
 iteration of this project. We feel confident we can resolve them in a way
 that preserves the integrity of the open web.


 Ben Mathwig
 Senior Product Manager
 Microsoft Edge Web Platform


 On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org wrote:

> Hello
>
> I am sharing the feedback from the Origin Trial with 12 participants:
>
>-
>
>10 of them responded "Extremely likely" to "How likely are you to
>keep using this feature?"
>-
>
>Qualitative feedback:
>
>
>-
>
>"I'm very excited about the CSP interpretation change rolling out
>in M92"
>-
>
>"looking forward to the CSP fix!"
>-
>
>"I'm very glad you're working on this!"
>-
>
>"This feature is great! I'd love to see it fully launch"
>
> Daisuke
>
>
> On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
> blin...@chromium.org> wrote:
>
>> Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted
>> to give a bit more context on how web bundles improve on what is possible
>> for automatic site optimization tools like mod_pagesped:
>>
>> 1. Combining many small images into a single file otherwise requires
>> spriting (with css to identify which area of the image you want for each
>> usage) and mod_pagespeed's ability to do 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Alex Russell]

There are high-scale operational challenges to distributing content 
optimally in many scenarios (e.g., Store packaged first-run or 
out-of-the-box-experiences) that need more than what Service Workers alone 
can provide. We are positive about this technology because it can help 
address those problems and open up new product opportunities (that we 
cannot say more about).

Best Regards,

Alex

On Tuesday, May 24, 2022 at 10:26:28 AM UTC-7 bema...@microsoft.com wrote:

> >>  You could have used a service worker and cache things in order to get 
> the same thing, combined with progressive web applications for 
> installability.
>
> This is true, but I think ergonomics of an encapsulated page/application 
> make it easier to reason about for very large applications. Maintaining the 
> service worker and the cache at scale can become a barrier to entry for 
> some developers and applications.
>
> On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:
>
>> Not that I am against, but does this unlock previously locked 
>> opportunities in the specific examples you just mentioned?
>> You could have used a service worker and cache things in order to get the 
>> same thing, combined with progressive web applications for installability.
>>
>> ☆*PhistucK*
>>
>>
>> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
>> blin...@chromium.org> wrote:
>>
>>> Microsoft has a strong interest in seeing this feature ship. We believe 
>>> that sub-resource bundling is opening the door to a new way of shipping and 
>>> delivering offline web applications, changing the traditional definition of 
>>> “web application”.
>>>
>>> Here are some ways Microsoft products can take advantage of this:
>>>
>>> *PowerApps*
>>>
>>> Microsoft PowerApps allows a developer to author an application and 
>>> deploy to iOS, Android, and the web. The first two platforms allow 
>>> applications to be deployed and used when the device is offline, but the 
>>> latter is currently not “installable” on the device. Web bundling could 
>>> unlock the capability for a web application to be “installed” on a device 
>>> to operate offline.
>>>
>>> *Office Online*
>>>
>>> Office productivity web applications are a perfect example of 
>>> applications that could benefit from a packaged bundle of application 
>>> resources. Combined with local storage APIs, this could help developers 
>>> reach communities that have little to no network connectivity.
>>>
>>>  
>>>
>>> While there have been concerns brought up by the community, we welcome 
>>> the opportunity to collaborate on addressing these issues in the next 
>>> iteration of this project. We feel confident we can resolve them in a way 
>>> that preserves the integrity of the open web.
>>>
>>>
>>> Ben Mathwig
>>> Senior Product Manager
>>> Microsoft Edge Web Platform
>>>
>>>
>>> On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org wrote:
>>>
 Hello

 I am sharing the feedback from the Origin Trial with 12 participants:

- 

10 of them responded "Extremely likely" to "How likely are you to 
keep using this feature?" 
- 

Qualitative feedback:


- 

"I'm very excited about the CSP interpretation change rolling out 
in M92"
- 

"looking forward to the CSP fix!"
- 

"I'm very glad you're working on this!"
- 

"This feature is great! I'd love to see it fully launch"

 Daisuke


 On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
 blin...@chromium.org> wrote:

> Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted 
> to give a bit more context on how web bundles improve on what is possible 
> for automatic site optimization tools like mod_pagesped:
>
> 1. Combining many small images into a single file otherwise requires 
> spriting (with css to identify which area of the image you want for each 
> usage) and mod_pagespeed's ability to do that automatically 
> (sprite_images 
> filter ) is 
> limited.  It needs to understand the site's css and the publisher needs 
> to 
> have already set their css up to minimize the changes required.  With 
> bundles it is much simpler: you put all the tiny image files in the 
> bundle, 
> and you rewrite the URLs to point into the bundle.
>
> 2. Combining many small css or js files into a single file (
> combine_css , 
> combine_js ) 
> requires hacks to prevent invalid css or js from breaking the rest.  It's 
> reasonably common that publishers will have  href=invalid.css> that doesn't parse, and if you blindly concatenate with 
> other css you will change the 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

['Ben Mathwig' via blink-dev]

>>  You could have used a service worker and cache things in order to get 
the same thing, combined with progressive web applications for 
installability.

This is true, but I think ergonomics of an encapsulated page/application 
make it easier to reason about for very large applications. Maintaining the 
service worker and the cache at scale can become a barrier to entry for 
some developers and applications.

On Tuesday, May 24, 2022 at 10:12:33 AM UTC-7 PhistucK wrote:

> Not that I am against, but does this unlock previously locked 
> opportunities in the specific examples you just mentioned?
> You could have used a service worker and cache things in order to get the 
> same thing, combined with progressive web applications for installability.
>
> ☆*PhistucK*
>
>
> On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
> blin...@chromium.org> wrote:
>
>> Microsoft has a strong interest in seeing this feature ship. We believe 
>> that sub-resource bundling is opening the door to a new way of shipping and 
>> delivering offline web applications, changing the traditional definition of 
>> “web application”.
>>
>> Here are some ways Microsoft products can take advantage of this:
>>
>> *PowerApps*
>>
>> Microsoft PowerApps allows a developer to author an application and 
>> deploy to iOS, Android, and the web. The first two platforms allow 
>> applications to be deployed and used when the device is offline, but the 
>> latter is currently not “installable” on the device. Web bundling could 
>> unlock the capability for a web application to be “installed” on a device 
>> to operate offline.
>>
>> *Office Online*
>>
>> Office productivity web applications are a perfect example of 
>> applications that could benefit from a packaged bundle of application 
>> resources. Combined with local storage APIs, this could help developers 
>> reach communities that have little to no network connectivity.
>>
>>  
>>
>> While there have been concerns brought up by the community, we welcome 
>> the opportunity to collaborate on addressing these issues in the next 
>> iteration of this project. We feel confident we can resolve them in a way 
>> that preserves the integrity of the open web.
>>
>>
>> Ben Mathwig
>> Senior Product Manager
>> Microsoft Edge Web Platform
>>
>>
>> On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org wrote:
>>
>>> Hello
>>>
>>> I am sharing the feedback from the Origin Trial with 12 participants:
>>>
>>>- 
>>>
>>>10 of them responded "Extremely likely" to "How likely are you to 
>>>keep using this feature?" 
>>>- 
>>>
>>>Qualitative feedback:
>>>
>>>
>>>- 
>>>
>>>"I'm very excited about the CSP interpretation change rolling out in 
>>>M92"
>>>- 
>>>
>>>"looking forward to the CSP fix!"
>>>- 
>>>
>>>"I'm very glad you're working on this!"
>>>- 
>>>
>>>"This feature is great! I'd love to see it fully launch"
>>>
>>> Daisuke
>>>
>>>
>>> On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
>>> blin...@chromium.org> wrote:
>>>
 Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted to 
 give a bit more context on how web bundles improve on what is possible for 
 automatic site optimization tools like mod_pagesped:

 1. Combining many small images into a single file otherwise requires 
 spriting (with css to identify which area of the image you want for each 
 usage) and mod_pagespeed's ability to do that automatically (sprite_images 
 filter ) is 
 limited.  It needs to understand the site's css and the publisher needs to 
 have already set their css up to minimize the changes required.  With 
 bundles it is much simpler: you put all the tiny image files in the 
 bundle, 
 and you rewrite the URLs to point into the bundle.

 2. Combining many small css or js files into a single file (combine_css 
 , combine_js 
 ) requires hacks 
 to prevent invalid css or js from breaking the rest.  It's reasonably 
 common that publishers will have  
 that doesn't parse, and if you blindly concatenate with other css you will 
 change the layout on the page. Since automatic site optimization tools 
 like 
 mod_pagespeed want to make the site load faster without making any changes 
 to how the site looks, that isn't acceptable.  Same issue with js.

 3. Today you need to have separate files for combined images, css, and 
 js.  With web bundles there can be just one.

 Jeff

 On Friday, May 20, 2022 at 1:11:16 PM UTC-4 Otto van der Schaaf wrote:

> As a maintainer of mod_pagespeed , I 
> would love to see this ship.
>
>
> On Wednesday, May 4, 2022 at 5:42:15 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[PhistucK]

Not that I am against, but does this unlock previously locked
opportunities in the specific examples you just mentioned?
You could have used a service worker and cache things in order to get the
same thing, combined with progressive web applications for installability.

☆*PhistucK*


On Tue, May 24, 2022 at 4:35 PM 'Ben Mathwig' via blink-dev <
blink-dev@chromium.org> wrote:

> Microsoft has a strong interest in seeing this feature ship. We believe
> that sub-resource bundling is opening the door to a new way of shipping and
> delivering offline web applications, changing the traditional definition of
> “web application”.
>
> Here are some ways Microsoft products can take advantage of this:
>
> *PowerApps*
>
> Microsoft PowerApps allows a developer to author an application and deploy
> to iOS, Android, and the web. The first two platforms allow applications to
> be deployed and used when the device is offline, but the latter is
> currently not “installable” on the device. Web bundling could unlock the
> capability for a web application to be “installed” on a device to operate
> offline.
>
> *Office Online*
>
> Office productivity web applications are a perfect example of applications
> that could benefit from a packaged bundle of application resources.
> Combined with local storage APIs, this could help developers reach
> communities that have little to no network connectivity.
>
>
>
> While there have been concerns brought up by the community, we welcome the
> opportunity to collaborate on addressing these issues in the next iteration
> of this project. We feel confident we can resolve them in a way that
> preserves the integrity of the open web.
>
>
> Ben Mathwig
> Senior Product Manager
> Microsoft Edge Web Platform
>
>
> On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org wrote:
>
>> Hello
>>
>> I am sharing the feedback from the Origin Trial with 12 participants:
>>
>>-
>>
>>10 of them responded "Extremely likely" to "How likely are you to
>>keep using this feature?"
>>-
>>
>>Qualitative feedback:
>>
>>
>>-
>>
>>"I'm very excited about the CSP interpretation change rolling out in
>>M92"
>>-
>>
>>"looking forward to the CSP fix!"
>>-
>>
>>"I'm very glad you're working on this!"
>>-
>>
>>"This feature is great! I'd love to see it fully launch"
>>
>> Daisuke
>>
>>
>> On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
>> blin...@chromium.org> wrote:
>>
>>> Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted to
>>> give a bit more context on how web bundles improve on what is possible for
>>> automatic site optimization tools like mod_pagesped:
>>>
>>> 1. Combining many small images into a single file otherwise requires
>>> spriting (with css to identify which area of the image you want for each
>>> usage) and mod_pagespeed's ability to do that automatically (sprite_images
>>> filter ) is
>>> limited.  It needs to understand the site's css and the publisher needs to
>>> have already set their css up to minimize the changes required.  With
>>> bundles it is much simpler: you put all the tiny image files in the bundle,
>>> and you rewrite the URLs to point into the bundle.
>>>
>>> 2. Combining many small css or js files into a single file (combine_css
>>> , combine_js
>>> ) requires hacks to
>>> prevent invalid css or js from breaking the rest.  It's reasonably common
>>> that publishers will have  that
>>> doesn't parse, and if you blindly concatenate with other css you will
>>> change the layout on the page. Since automatic site optimization tools like
>>> mod_pagespeed want to make the site load faster without making any changes
>>> to how the site looks, that isn't acceptable.  Same issue with js.
>>>
>>> 3. Today you need to have separate files for combined images, css, and
>>> js.  With web bundles there can be just one.
>>>
>>> Jeff
>>>
>>> On Friday, May 20, 2022 at 1:11:16 PM UTC-4 Otto van der Schaaf wrote:
>>>
 As a maintainer of mod_pagespeed , I
 would love to see this ship.


 On Wednesday, May 4, 2022 at 5:42:15 PM UTC+2 slightlyoff via
 Chromestatus wrote:

> Microsoft would like to see this ship ASAP. LGTM1
>
 --
>>>
>> 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/45f97f97-7db0-42ee-bea8-d01ddb344ba6n%40chromium.org
>>> 
>>> .
>>>
>> --
> You received 

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

['Ben Mathwig' via blink-dev]

Microsoft has a strong interest in seeing this feature ship. We believe 
that sub-resource bundling is opening the door to a new way of shipping and 
delivering offline web applications, changing the traditional definition of 
“web application”.

Here are some ways Microsoft products can take advantage of this:

*PowerApps*

Microsoft PowerApps allows a developer to author an application and deploy 
to iOS, Android, and the web. The first two platforms allow applications to 
be deployed and used when the device is offline, but the latter is 
currently not “installable” on the device. Web bundling could unlock the 
capability for a web application to be “installed” on a device to operate 
offline.

*Office Online*

Office productivity web applications are a perfect example of applications 
that could benefit from a packaged bundle of application resources. 
Combined with local storage APIs, this could help developers reach 
communities that have little to no network connectivity.

 

While there have been concerns brought up by the community, we welcome the 
opportunity to collaborate on addressing these issues in the next iteration 
of this project. We feel confident we can resolve them in a way that 
preserves the integrity of the open web.


Ben Mathwig
Senior Product Manager
Microsoft Edge Web Platform


On Sunday, May 22, 2022 at 6:28:50 PM UTC-7 deno...@chromium.org wrote:

> Hello
>
> I am sharing the feedback from the Origin Trial with 12 participants:
>
>- 
>
>10 of them responded "Extremely likely" to "How likely are you to keep 
>using this feature?" 
>- 
>
>Qualitative feedback:
>
>
>- 
>
>"I'm very excited about the CSP interpretation change rolling out in 
>M92"
>- 
>
>"looking forward to the CSP fix!"
>- 
>
>"I'm very glad you're working on this!"
>- 
>
>"This feature is great! I'd love to see it fully launch"
>
> Daisuke
>
>
> On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
> blin...@chromium.org> wrote:
>
>> Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted to 
>> give a bit more context on how web bundles improve on what is possible for 
>> automatic site optimization tools like mod_pagesped:
>>
>> 1. Combining many small images into a single file otherwise requires 
>> spriting (with css to identify which area of the image you want for each 
>> usage) and mod_pagespeed's ability to do that automatically (sprite_images 
>> filter ) is 
>> limited.  It needs to understand the site's css and the publisher needs to 
>> have already set their css up to minimize the changes required.  With 
>> bundles it is much simpler: you put all the tiny image files in the bundle, 
>> and you rewrite the URLs to point into the bundle.
>>
>> 2. Combining many small css or js files into a single file (combine_css 
>> , combine_js 
>> ) requires hacks to 
>> prevent invalid css or js from breaking the rest.  It's reasonably common 
>> that publishers will have  that 
>> doesn't parse, and if you blindly concatenate with other css you will 
>> change the layout on the page. Since automatic site optimization tools like 
>> mod_pagespeed want to make the site load faster without making any changes 
>> to how the site looks, that isn't acceptable.  Same issue with js.
>>
>> 3. Today you need to have separate files for combined images, css, and 
>> js.  With web bundles there can be just one.
>>
>> Jeff
>>
>> On Friday, May 20, 2022 at 1:11:16 PM UTC-4 Otto van der Schaaf wrote:
>>
>>> As a maintainer of mod_pagespeed , I 
>>> would love to see this ship.
>>>
>>>
>>> On Wednesday, May 4, 2022 at 5:42:15 PM UTC+2 slightlyoff via 
>>> Chromestatus wrote:
>>>
 Microsoft would like to see this ship ASAP. LGTM1

>>> -- 
>>
> 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/45f97f97-7db0-42ee-bea8-d01ddb344ba6n%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/81ea3aac-58c3-4a77-941a-b647c03b2edfn%40chromium.org.

Re: [blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Daisuke Enomoto]

Hello

I am sharing the feedback from the Origin Trial with 12 participants:

   -

   10 of them responded "Extremely likely" to "How likely are you to keep
   using this feature?"
   -

   Qualitative feedback:


   -

   "I'm very excited about the CSP interpretation change rolling out in M92"
   -

   "looking forward to the CSP fix!"
   -

   "I'm very glad you're working on this!"
   -

   "This feature is great! I'd love to see it fully launch"

Daisuke


On Sat, May 21, 2022 at 5:38 AM 'Jeff Kaufman' via blink-dev <
blink-dev@chromium.org> wrote:

> Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted to
> give a bit more context on how web bundles improve on what is possible for
> automatic site optimization tools like mod_pagesped:
>
> 1. Combining many small images into a single file otherwise requires
> spriting (with css to identify which area of the image you want for each
> usage) and mod_pagespeed's ability to do that automatically (sprite_images
> filter ) is
> limited.  It needs to understand the site's css and the publisher needs to
> have already set their css up to minimize the changes required.  With
> bundles it is much simpler: you put all the tiny image files in the bundle,
> and you rewrite the URLs to point into the bundle.
>
> 2. Combining many small css or js files into a single file (combine_css
> , combine_js
> ) requires hacks to
> prevent invalid css or js from breaking the rest.  It's reasonably common
> that publishers will have  that
> doesn't parse, and if you blindly concatenate with other css you will
> change the layout on the page. Since automatic site optimization tools like
> mod_pagespeed want to make the site load faster without making any changes
> to how the site looks, that isn't acceptable.  Same issue with js.
>
> 3. Today you need to have separate files for combined images, css, and
> js.  With web bundles there can be just one.
>
> Jeff
>
> On Friday, May 20, 2022 at 1:11:16 PM UTC-4 Otto van der Schaaf wrote:
>
>> As a maintainer of mod_pagespeed , I
>> would love to see this ship.
>>
>>
>> On Wednesday, May 4, 2022 at 5:42:15 PM UTC+2 slightlyoff via
>> Chromestatus wrote:
>>
>>> Microsoft would like to see this ship ASAP. LGTM1
>>>
>> --
> 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/45f97f97-7db0-42ee-bea8-d01ddb344ba6n%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/CAA5e698aaK%3DnmW7_8ni14m2qn1-51f1uLWNPPsp1FozYNN0p7Q%40mail.gmail.com.

[blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

['Jeff Kaufman' via blink-dev]

Thanks, Otto!  As someone who used to work on mod_pagespeed I wanted to 
give a bit more context on how web bundles improve on what is possible for 
automatic site optimization tools like mod_pagesped:

1. Combining many small images into a single file otherwise requires 
spriting (with css to identify which area of the image you want for each 
usage) and mod_pagespeed's ability to do that automatically (sprite_images 
filter ) is limited.  
It needs to understand the site's css and the publisher needs to have 
already set their css up to minimize the changes required.  With bundles it 
is much simpler: you put all the tiny image files in the bundle, and you 
rewrite the URLs to point into the bundle.

2. Combining many small css or js files into a single file (combine_css 
, combine_js 
) requires hacks to 
prevent invalid css or js from breaking the rest.  It's reasonably common 
that publishers will have  that 
doesn't parse, and if you blindly concatenate with other css you will 
change the layout on the page. Since automatic site optimization tools like 
mod_pagespeed want to make the site load faster without making any changes 
to how the site looks, that isn't acceptable.  Same issue with js.

3. Today you need to have separate files for combined images, css, and js.  
With web bundles there can be just one.

Jeff

On Friday, May 20, 2022 at 1:11:16 PM UTC-4 Otto van der Schaaf wrote:

> As a maintainer of mod_pagespeed , I would 
> love to see this ship.
>
>
> On Wednesday, May 4, 2022 at 5:42:15 PM UTC+2 slightlyoff via Chromestatus 
> wrote:
>
>> Microsoft would like to see this ship ASAP. LGTM1
>>
>

-- 
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/45f97f97-7db0-42ee-bea8-d01ddb344ba6n%40chromium.org.

[blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[Otto van der Schaaf]

As a maintainer of mod_pagespeed , I would 
love to see this ship.


On Wednesday, May 4, 2022 at 5:42:15 PM UTC+2 slightlyoff via Chromestatus 
wrote:

> Microsoft would like to see this ship ASAP. LGTM1
>

-- 
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/efadf7e5-dba9-4d9a-abfc-6506955a03f1n%40chromium.org.

[blink-dev] Re: Intent to Ship: Subresource loading with Web Bundles

[slightlyoff via Chromestatus]

Microsoft would like to see this ship ASAP. LGTM1

-- 
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/faf64d05de317773%40google.com.

Re: [blink-dev] Re: Intent to Ship: Subresource Loading with Web Bundles

[Hayato Ito]

Thanks for the question. Responses inline below.

On Tue, Apr 19, 2022 at 6:29 PM Titouan Rigoudy 
wrote:

> Hi there,
>
> Has anything changed between the OT and this launch?
>
>
A good question.  No behavior change between the OT and this launch.

Thanks!



> Cheers,
> Titouan
>
> On Wed, Apr 13, 2022 at 10:39 AM Hayato Ito  wrote:
>
>> Thanks for the clarification.
>>
>> Yeah, I think it's up-to each browser, however, it is worth clarifying
>> that extensions in chrome are able to see and operate on subresources in
>> Web Bundles.
>>
>> See https://bugs.chromium.org/p/chromium/issues/detail?id=1155402 for
>> details.
>>
>> On Wed, Apr 13, 2022 at 5:18 PM Kenji Baheux 
>> wrote:
>>
>>> A quick clarification ✔️ to avoid confusion  or needless agitation 掠:
>>>
>>> On Wednesday, April 13, 2022 at 12:43:24 PM UTC+9 Hayato Ito wrote:
>>>
 Contact emails:hay...@chromium.org, denom...@chromium.org


 Explainer:


 https://github.com/WICG/webpackage/blob/main/explainers/subresource-loading.md


 Specification:

 https://wicg.github.io/webpackage/subresource-loading.html


 Summary:

 Provides a new approach to load a large number of resources efficiently
 using a format that allows multiple resources to be bundled, Web
 Bundles
 
 .


 Note that the scope of this "Intent to Ship" is only v1, which is
 explained in our long term plan
 . We have been
 conducting the Origin Trial for v1 features
 .
 Given that the outcome is promising, we’d like to ship v1 as MVP in order
 to mitigate the risk of "develop everything and ship them all at once".


 After shipping v1, we'll explore *v2*, designing and implementing
 missing key features, eg. cache-aware WebBundles. There are several
 proposals, such as Bundle Preloading
  (discussion
 ), Bundle Dependencies
 ,
 and others. We continue to explore this area to identify a missing Web
 Platform primitive in order to serve large Web Apps in a cache-efficient
 way with Web Bundles.


 Blink component

 Blink>Loader>WebPackaging


 TAG review: https://github.com/w3ctag/design-reviews/issues/616

 TAG review status: While the status of the review is open, we have
 addressed the concerns raised in the discussion. In particular, we plan to
 explore the designs in v2 as explained above. Additionally, if a concern
 with the extension part
 
 continues to be an issue, we aim to address it in v2 as well.


 Risks:

 Interoperability and Compatibility:

 No interoperability and compatibility risk has been identified. This is
 purely a feature addition. A browser which doesn’t support this feature
 should load subresources from the network, as usual.


 Gecko: No signal (issue
 )

 WebKit: No signal (ML
 
 )


 Web Developers:

 Google Ads (use case )
 (origin trial participant)

 > Web bundle serving is a major overhaul of how GPT requests and
 renders ads, built on top of a new browser API which we have been designing
 with the Chrome team.  It offers large loading performance improvements and
 security and privacy relative to safeframe rendering:

-

Performance improvements by fetching multiple Ads creatives in a
single request.
-

Enhance privacy: Creative contents can no longer be read or
modified by the publisher or others in the publisher's JS context.
Creatives can no longer read or modify each other.


>>> This hardening of read/write access is only meant for code originally
>>> present on the page, not code coming from user extensions.
>>> In other words, *extensions can see and operate* on ads served in this
>>> manner.
>>> In particular, *extensions can still act upstream *of these ads
>>> requests as is typically the case (e.g. directly acting on the scripts that
>>> drive these requests).
>>>
>>> The goal for this particular use case is to improve users’ privacy and
>>> security by preventing non-user controlled/added code from peeking into, or
>>> 

Re: [blink-dev] Re: Intent to Ship: Subresource Loading with Web Bundles

[Titouan Rigoudy]

Hi there,

Has anything changed between the OT and this launch?

Cheers,
Titouan

On Wed, Apr 13, 2022 at 10:39 AM Hayato Ito  wrote:

> Thanks for the clarification.
>
> Yeah, I think it's up-to each browser, however, it is worth clarifying
> that extensions in chrome are able to see and operate on subresources in
> Web Bundles.
>
> See https://bugs.chromium.org/p/chromium/issues/detail?id=1155402 for
> details.
>
> On Wed, Apr 13, 2022 at 5:18 PM Kenji Baheux 
> wrote:
>
>> A quick clarification ✔️ to avoid confusion  or needless agitation 掠:
>>
>> On Wednesday, April 13, 2022 at 12:43:24 PM UTC+9 Hayato Ito wrote:
>>
>>> Contact emails:hay...@chromium.org, denom...@chromium.org
>>>
>>>
>>> Explainer:
>>>
>>>
>>> https://github.com/WICG/webpackage/blob/main/explainers/subresource-loading.md
>>>
>>>
>>> Specification:
>>>
>>> https://wicg.github.io/webpackage/subresource-loading.html
>>>
>>>
>>> Summary:
>>>
>>> Provides a new approach to load a large number of resources efficiently
>>> using a format that allows multiple resources to be bundled, Web Bundles
>>> 
>>> .
>>>
>>>
>>> Note that the scope of this "Intent to Ship" is only v1, which is
>>> explained in our long term plan
>>> . We have been
>>> conducting the Origin Trial for v1 features
>>> .
>>> Given that the outcome is promising, we’d like to ship v1 as MVP in order
>>> to mitigate the risk of "develop everything and ship them all at once".
>>>
>>>
>>> After shipping v1, we'll explore *v2*, designing and implementing
>>> missing key features, eg. cache-aware WebBundles. There are several
>>> proposals, such as Bundle Preloading
>>>  (discussion
>>> ), Bundle Dependencies
>>> ,
>>> and others. We continue to explore this area to identify a missing Web
>>> Platform primitive in order to serve large Web Apps in a cache-efficient
>>> way with Web Bundles.
>>>
>>>
>>> Blink component
>>>
>>> Blink>Loader>WebPackaging
>>>
>>>
>>> TAG review: https://github.com/w3ctag/design-reviews/issues/616
>>>
>>> TAG review status: While the status of the review is open, we have
>>> addressed the concerns raised in the discussion. In particular, we plan to
>>> explore the designs in v2 as explained above. Additionally, if a concern
>>> with the extension part
>>> 
>>> continues to be an issue, we aim to address it in v2 as well.
>>>
>>>
>>> Risks:
>>>
>>> Interoperability and Compatibility:
>>>
>>> No interoperability and compatibility risk has been identified. This is
>>> purely a feature addition. A browser which doesn’t support this feature
>>> should load subresources from the network, as usual.
>>>
>>>
>>> Gecko: No signal (issue
>>> )
>>>
>>> WebKit: No signal (ML
>>> 
>>> )
>>>
>>>
>>> Web Developers:
>>>
>>> Google Ads (use case )
>>> (origin trial participant)
>>>
>>> > Web bundle serving is a major overhaul of how GPT requests and renders
>>> ads, built on top of a new browser API which we have been designing with
>>> the Chrome team.  It offers large loading performance improvements and
>>> security and privacy relative to safeframe rendering:
>>>
>>>-
>>>
>>>Performance improvements by fetching multiple Ads creatives in a
>>>single request.
>>>-
>>>
>>>Enhance privacy: Creative contents can no longer be read or modified
>>>by the publisher or others in the publisher's JS context. Creatives can 
>>> no
>>>longer read or modify each other.
>>>
>>>
>> This hardening of read/write access is only meant for code originally
>> present on the page, not code coming from user extensions.
>> In other words, *extensions can see and operate* on ads served in this
>> manner.
>> In particular, *extensions can still act upstream *of these ads requests
>> as is typically the case (e.g. directly acting on the scripts that drive
>> these requests).
>>
>> The goal for this particular use case is to improve users’ privacy and
>> security by preventing non-user controlled/added code from peeking into, or
>> making malicious changes to, ads that were served by another party, with
>> better or neutral performance (see the TAG review for more details and
>> analysis).
>>
>> I hope this is clear and useful  .
>>
>>
>>
>>>
>>> Ergonomics
>>>
>>> This feature can be used to improve loading performance by fetching
>>> multiple 

[blink-dev] Re: Intent to Ship: Subresource Loading with Web Bundles

[Hayato Ito]

Thanks for the clarification.

Yeah, I think it's up-to each browser, however, it is worth clarifying that
extensions in chrome are able to see and operate on subresources in Web
Bundles.

See https://bugs.chromium.org/p/chromium/issues/detail?id=1155402 for
details.

On Wed, Apr 13, 2022 at 5:18 PM Kenji Baheux 
wrote:

> A quick clarification ✔️ to avoid confusion  or needless agitation 掠:
>
> On Wednesday, April 13, 2022 at 12:43:24 PM UTC+9 Hayato Ito wrote:
>
>> Contact emails:hay...@chromium.org, denom...@chromium.org
>>
>>
>> Explainer:
>>
>>
>> https://github.com/WICG/webpackage/blob/main/explainers/subresource-loading.md
>>
>>
>> Specification:
>>
>> https://wicg.github.io/webpackage/subresource-loading.html
>>
>>
>> Summary:
>>
>> Provides a new approach to load a large number of resources efficiently
>> using a format that allows multiple resources to be bundled, Web Bundles
>> 
>> .
>>
>>
>> Note that the scope of this "Intent to Ship" is only v1, which is
>> explained in our long term plan
>> . We have been conducting
>> the Origin Trial for v1 features
>> .
>> Given that the outcome is promising, we’d like to ship v1 as MVP in order
>> to mitigate the risk of "develop everything and ship them all at once".
>>
>>
>> After shipping v1, we'll explore *v2*, designing and implementing
>> missing key features, eg. cache-aware WebBundles. There are several
>> proposals, such as Bundle Preloading
>>  (discussion
>> ), Bundle Dependencies
>> ,
>> and others. We continue to explore this area to identify a missing Web
>> Platform primitive in order to serve large Web Apps in a cache-efficient
>> way with Web Bundles.
>>
>>
>> Blink component
>>
>> Blink>Loader>WebPackaging
>>
>>
>> TAG review: https://github.com/w3ctag/design-reviews/issues/616
>>
>> TAG review status: While the status of the review is open, we have
>> addressed the concerns raised in the discussion. In particular, we plan to
>> explore the designs in v2 as explained above. Additionally, if a concern
>> with the extension part
>> 
>> continues to be an issue, we aim to address it in v2 as well.
>>
>>
>> Risks:
>>
>> Interoperability and Compatibility:
>>
>> No interoperability and compatibility risk has been identified. This is
>> purely a feature addition. A browser which doesn’t support this feature
>> should load subresources from the network, as usual.
>>
>>
>> Gecko: No signal (issue
>> )
>>
>> WebKit: No signal (ML
>> 
>> )
>>
>>
>> Web Developers:
>>
>> Google Ads (use case )
>> (origin trial participant)
>>
>> > Web bundle serving is a major overhaul of how GPT requests and renders
>> ads, built on top of a new browser API which we have been designing with
>> the Chrome team.  It offers large loading performance improvements and
>> security and privacy relative to safeframe rendering:
>>
>>-
>>
>>Performance improvements by fetching multiple Ads creatives in a
>>single request.
>>-
>>
>>Enhance privacy: Creative contents can no longer be read or modified
>>by the publisher or others in the publisher's JS context. Creatives can no
>>longer read or modify each other.
>>
>>
> This hardening of read/write access is only meant for code originally
> present on the page, not code coming from user extensions.
> In other words, *extensions can see and operate* on ads served in this
> manner.
> In particular, *extensions can still act upstream *of these ads requests
> as is typically the case (e.g. directly acting on the scripts that drive
> these requests).
>
> The goal for this particular use case is to improve users’ privacy and
> security by preventing non-user controlled/added code from peeking into, or
> making malicious changes to, ads that were served by another party, with
> better or neutral performance (see the TAG review for more details and
> analysis).
>
> I hope this is clear and useful  .
>
>
>
>>
>> Ergonomics
>>
>> This feature can be used to improve loading performance by fetching
>> multiple resources in a single request. If a browser doesn’t support this
>> feature, a request should be served from a network as usual.
>>
>> `HTMLScriptElement.supports("webbundle")` can be used for feature
>> detection as well.
>>
>> There are several tools and plugins
>> 

[blink-dev] Re: Intent to Ship: Subresource Loading with Web Bundles

[Kenji Baheux]

A quick clarification ✔️ to avoid confusion  or needless agitation 掠:

On Wednesday, April 13, 2022 at 12:43:24 PM UTC+9 Hayato Ito wrote:

> Contact emails:hay...@chromium.org, denom...@chromium.org
>
>
> Explainer:
>
>
> https://github.com/WICG/webpackage/blob/main/explainers/subresource-loading.md
>
>
> Specification:
>
> https://wicg.github.io/webpackage/subresource-loading.html
>
>
> Summary:
>
> Provides a new approach to load a large number of resources efficiently 
> using a format that allows multiple resources to be bundled, Web Bundles 
> 
> .
>
>
> Note that the scope of this "Intent to Ship" is only v1, which is 
> explained in our long term plan 
> . We have been conducting 
> the Origin Trial for v1 features 
> .
>  
> Given that the outcome is promising, we’d like to ship v1 as MVP in order 
> to mitigate the risk of "develop everything and ship them all at once".
>
>
> After shipping v1, we'll explore *v2*, designing and implementing missing 
> key features, eg. cache-aware WebBundles. There are several proposals, such 
> as Bundle Preloading  (
> discussion ), Bundle 
> Dependencies 
> ,
>  
> and others. We continue to explore this area to identify a missing Web 
> Platform primitive in order to serve large Web Apps in a cache-efficient 
> way with Web Bundles.
>
>
> Blink component
>
> Blink>Loader>WebPackaging
>
>
> TAG review: https://github.com/w3ctag/design-reviews/issues/616
>
> TAG review status: While the status of the review is open, we have 
> addressed the concerns raised in the discussion. In particular, we plan to 
> explore the designs in v2 as explained above. Additionally, if a concern 
> with the extension part 
> 
>  
> continues to be an issue, we aim to address it in v2 as well.
>
>
> Risks:
>
> Interoperability and Compatibility:
>
> No interoperability and compatibility risk has been identified. This is 
> purely a feature addition. A browser which doesn’t support this feature 
> should load subresources from the network, as usual.
>
>
> Gecko: No signal (issue 
> )
>
> WebKit: No signal (ML 
> )
>
>
> Web Developers:
>
> Google Ads (use case ) 
> (origin trial participant)
>
> > Web bundle serving is a major overhaul of how GPT requests and renders 
> ads, built on top of a new browser API which we have been designing with 
> the Chrome team.  It offers large loading performance improvements and 
> security and privacy relative to safeframe rendering:
>
>- 
>
>Performance improvements by fetching multiple Ads creatives in a 
>single request.
>- 
>
>Enhance privacy: Creative contents can no longer be read or modified 
>by the publisher or others in the publisher's JS context. Creatives can no 
>longer read or modify each other.
>
>
This hardening of read/write access is only meant for code originally 
present on the page, not code coming from user extensions.
In other words, *extensions can see and operate* on ads served in this 
manner. 
In particular, *extensions can still act upstream *of these ads requests as 
is typically the case (e.g. directly acting on the scripts that drive these 
requests).

The goal for this particular use case is to improve users’ privacy and 
security by preventing non-user controlled/added code from peeking into, or 
making malicious changes to, ads that were served by another party, with 
better or neutral performance (see the TAG review for more details and 
analysis).

I hope this is clear and useful  .

 

>
> Ergonomics
>
> This feature can be used to improve loading performance by fetching 
> multiple resources in a single request. If a browser doesn’t support this 
> feature, a request should be served from a network as usual.
>
> `HTMLScriptElement.supports("webbundle")` can be used for feature 
> detection as well.
>
> There are several tools and plugins 
>  available for packaging 
> Web Bundles.
>
>
> Security
>
> Received approval for Security and Privacy in our launch bug 
> . We had addressed a security concern for the 
> usage of  elements. The API now uses