Request for Feedback: "Subresource Loading with Web Bundles"

6 views
Skip to first unread message

Hayato Ito

unread,
Apr 21, 2022, 3:36:10 AM4/21/22
to m...@mozilla.com, avanke...@mozilla.com, Daisuke Enomoto, Kunihiko Sakamoto, Kenji Baheux, Chris Harrelson, webpack...@chromium.org
Hi Martin and Anne,

I'm Hayato, working on the WebBundle team in Google.

We've sent "Intent to Ship: Subresource Loading with Web Bundles" [1] in blink-dev.
Blink API owners recommended us to ask for feedback from other browser vendors.
It'd be great for us to get more signals.

We'd appreciate it if we could hear feedback here [2] from Mozilla.
If you have any questions, please feel free to ask. We are happy to answer!

- [1]: https://groups.google.com/a/chromium.org/g/blink-dev/c/VS9UgOC7Wqc/m/P0vG6qfEEwAJ
- [2]: https://github.com/mozilla/standards-positions/issues/590

--
Hayato

Hayato Ito

unread,
Apr 21, 2022, 3:36:11 AM4/21/22
to wila...@apple.com, Daisuke Enomoto, Kunihiko Sakamoto, Kenji Baheux, Chris Harrelson, webpack...@chromium.org
Hi John,


I'm Hayato, working on the WebBundle team in Google.

We've sent "Intent to Ship: Subresource Loading with Web Bundles" [1] in blink-dev.
Blink API owners recommended us to ask for feedback from other browser vendors.
It'd be great for us to get more signals.

We'd appreciate it if we could hear feedback here [2] from Apple.

If you have any questions, please feel free to ask. We are happy to answer!

- [1]: https://groups.google.com/a/chromium.org/g/blink-dev/c/VS9UgOC7Wqc/m/P0vG6qfEEwAJ

Martin Thomson

unread,
Apr 26, 2022, 5:26:12 PM4/26/22
to Hayato Ito, Anne van Kesteren, Daisuke Enomoto, Kunihiko Sakamoto, Kenji Baheux, Chris Harrelson, webpack...@chromium.org
Hi Hayato,

You caught me on vacation and I'm only recently returned now to a massive pile of work.  This might take a little while to put together, especially given the size of this feature (I'm four links deep now and I still can't answer the first question I had).

I do want to make it clear that there is a distinction between what we think of the current shape of the proposal (our response in [2]) and whether we agree with you that shipping this feature is a good idea.  This is a massive chunk of work that - despite you working in public on it for some years now - is still essentially proprietary.  Shipping something in that state is something we might object to, even if we end up liking the specification[*].

Cheers,
Martin

[*] Note that I think that us liking the specification in its current form is unlikely.  I found an issue with the CSP section just skimming through the explainer.



Reply all
Reply to author
Forward
0 new messages