http-extensions icon indicating copy to clipboard operation
http-extensions copied to clipboard

RESUMABLE: opinion - the design has unnecessary premature options

Open LPardue opened this issue 4 months ago • 4 comments

@gstrauss provided the following feedback on https://lists.w3.org/Archives/Public/ietf-http-wg/2025JulSep/0094.html wrt Section 11

This document is overly complex, in part due to unnecessary premature optimizations like this one. If an upload is small, the upload can be repeated, if safe. If an upload is not small, then an extra round trip to mark the upload as complete should almost never be a performance problem.

LPardue avatar Aug 11 '25 23:08 LPardue

Speaking as an individual: the opinion in https://github.com/httpwg/http-extensions/issues/3192#issue-3311909127 does not sufficiently reflect the variety of heterogenous networks that this document is designed for. For instance, a path with high throughput but long RTTs, such as a non-LEO satellite network. Features of the design, such as 104 (Upload Resumption Supported) that can help optimize, are optional to use.

LPardue avatar Aug 11 '25 23:08 LPardue

What's the proposal here?

mnot avatar Aug 12 '25 00:08 mnot

What's the proposal here?

@mnot, I think this is incorrectly split off from the previous paragraphs, which assert:

If the upload resource is simply a file and the client manages overall state and the client finishes the upload before sending Upload-Complete: ?1 with an empty body, then a client can send a HEAD request for size details. Upload-Offset would be redundant. There would be little need to keep upload resources around after Upload-Complete: ?1. There would be little need to make queries to try to approximate "transaction status" or "history" of the upload resource state.

In the event that the client did not receive a response to request with Upload-Complete: ?1 and empty body, the client could repeat that request.

The original comment is an attempt to pre-empt objections to the extra round-trip in the proposed model. The additional round-trip of sending the completion as a separate request after confirming all chunks were accepted is claimed to be negligible if the upload is actually large enough to justify using this feature.

(In fact, that's highly dependent on BDP, but likely true for most terrestrial use cases.)

MikeBishop avatar Aug 12 '25 19:08 MikeBishop

Proposal: close with no action

Acconut avatar Oct 02 '25 17:10 Acconut