Opened 22 months ago

Last modified 11 months ago

#27 new defect

Terminology improvements

Reported by: wes@… Owned by: draft-ietf-tsvwg-l4s-arch@…
Priority: major Milestone: L4S Suite - WGLC Preparation
Component: l4s-arch Version:
Severity: - Keywords:
Cc:

Description

Use of terms "traditional", "classic", and "legacy" should be carefully checked.

See Michael Scharf's comment: https://mailarchive.ietf.org/arch/msg/tsvwg/8_gdxerN-99oou9ILas5OKl8viU

Change History (2)

comment:1 Changed 20 months ago by jholland@…

Also perhaps worth noting was a comment about reducing "hype" in the text, giving examples from the abstract:

comment:2 Changed 11 months ago by g.white@…

In response to Michael Scharf's comment, there was quite a bit of discussion on the mailing list, and no clear consensus to remove the word "classic". As a compromise, it was proposed to eliminate the usage of the term "classic TCP", but to continue to use "classic congestion control", defined to mean 'Reno-friendly'. These changes were made in the Feb 2020 updates of the drafts.

The term "traditional" appears to be used appropriately in the current drafts ("DCTCP uses the traditional TCP Reno additive increase" and "Traditionally ultra-low latency has only been available for a few selected low rate applications").

The term "legacy" is used once in DualQ to refer to senders that are not compliant with RFC8311. It is used 5 times in L4s-Arch. These instances probably need a final scrub.

The drafts were also revised in Feb & March 2020 in an attempt to eliminate hype.

It would be good to get feedback from WG participants on the status of this issue. Can this be closed, or is there still concern about the terminology?

Last edited 11 months ago by g.white@… (previous) (diff)
Note: See TracTickets for help on using tickets.