#633: JPEG XL decoding

Visit on Github.

Opened May 12, 2021

Hi TAG!

I'm requesting a TAG review of JPEG XL decoding, a new image format.

Further details:

  • I have reviewed the TAG's Web Platform Design Principles
  • Relevant time constraints or deadlines: N/A
  • The group where the work on this specification is currently being done: JPEG
  • The group where standardization of this work is intended to be done (if current group is a community group or other incubation venue): JPEG

You should also know that it was noted in https://github.com/w3ctag/design-reviews/issues/495 that "[...] a Tag Review wouldn't be applicable for an image decoder implementation.". As there seems to be some disagreement on this point, we're opening this issue for further discussion.

We'd prefer the TAG provide feedback as (please delete all but the desired option):

💬 leave review feedback as a comment in this issue and @-notify @veluca93 and @jonsneyers

Discussions

2021-08-16

Minutes

Ken: Looking at this, it seems that we have given advice (no more mime sniffing) and that it is being discussed by the right people in the mime sniffing repo

Ken: propose close

2021-08-30

Minutes

@cynthia

Yves: I thought we decided a long time ago to close it

Sangwhan: we need a position on whether or not mime sniffing. So far we've been against mime sniffing. This is one case where some people in the wg pushed back and said they want mime sniffing for this case because it makes it convenient... disagreement with people in whatwg

Ken: the right people are discussing this so i don' tknow what we need to say

Yves: with orb work they want to limit things that are mime sniffed and go back to a stricter checking of media types in the future

Sangwhan: i've been following that guideance and suddenly the wg people disagreed and I'm not sure how to react. Technically it seems fine to go.

Yves: as the original issue was not really about that, we can send that comment saying mime sniffing in the ifrst case was not something we should rely on we are against adding new things to mime sniffing, but we give that to whatwg to decide. Then close the original issue which was about content encoding.

Sangwhan: I'm fine with that

Ken: me too

Sangwhan: there's a principles issue here, we do say something about mime sniffing

Dan: we should reference that in the comment

Sangwhan: we did, they asked what security implications

Dan: we should close saying we've provided feedback, they know what we've said in design principles, if there's an issue there they can raise it there if they think there needs to be more explanation

Status: Yves to do closing comment.