Early wide review of IFT about ift HOT 23 OPEN

@vlevantovsky @garretrieger my feeling is that we should wrap up any outstanding edits and then republish the spec before asking for TAG review. I also need to write an explainer before we ask for their review.

Security has had plenty of time to respond, we don't need to wait longer; marking their review as complete due to timeout.

from ift.

I18n issues raised:

• usage of terminology 'codepoints' in patch subset spec.
• Supporting fonts with composite glyphs via range-request

from ift.

Privacy issues raised:

• Proposal would allow pages to learn how the user is interacting with the site
• Repeat fingerprinting analysis assuming a malicious font serve
• Including network state is a privacy harm (and maybe not needed)
• Define functionality in terms of fetch() (or otherwise make sure that caching / storage policy matches the rest of the platform)

from ift.

Security should be given a second change to review, once we publish the merged spec (they never got a chance to look at the FPWD of the separate Range Request spec)

from ift.

The two open privacy issues need further discussion, because I think we are talking at cross purposes, but I am optimistic that they can be closed relatively quickly

from ift.

I'd like to request TAG review once we publish the merged spec, though I think it would be preferable to do the network perf review first, to save them telling us to go do that :)

from ift.

Adding network performance review as suggested in #74 by @mnot

from ift.

Further review progress pending on /TR publication of merged spec

from ift.

The merged spec has been published and the history indicates that the separate Range Request spec is superseded by the merged IFT spec

from ift.

I18n review is closed

from ift.

Web Perf WG review requested 28 June 2022

from ift.

IETF HTTP WG review requested 28 June 2022

from ift.

On July 7 Perf WG Agenda

from ift.

Response from Martin Thomson raises several issues, which we should track as individual GH issues.

Apparently the rationale from the PFE report was not obvious, and I responded on that already and will add an issue.

from ift.

Minutes of 7 July Perf WG meeting

from ift.

Response from Martin Thomson raises several issues, which we should track as individual GH issues.

This is now done:

• #115
• #116
• #117
• #118
• #119

from ift.

Minutes of 7 July Perf WG meeting called for any new issues to be raised here and none have.

from ift.

There are zero open privacy issues so marking privacy review as completed.

from ift.

All the issues from Martin Thomson are now closed, so marking the IETF HTTP WG Review as closed

from ift.

Sounds good, we can discuss what's left to do at the next working group call. Should we also split out the range request specification before republishing and asking for a TAG review?

from ift.

Should we also split out the range request specification before republishing and asking for a TAG review?

I think so, there are a bunch of issues tagged Range Request and it is advancing more slowly than Patch Subset.

from ift.

OK I took a pass at splitting here: #138

I kept the split simple, just moved the range request specific section out while keeping the range request bits in the "common" section around in the main document.

from ift.

I think that was a good call.

from ift.