Re: [rfc-i] Updates solution that you object to (was Re: draft-kuehlewind-update-tag/)

On 29 Mar 2020, at 16:45, Joseph Touch wrote:

> We’re still waiting to hear why this is needed.
>
> The IESG appears to have anecdotal evidence of the need for a 
> significant change to document marking but we have see NO evidence 
> that this is anything but a corner-case issue that will continue to be 
> a (possibly different) corner-case issue.

Not to be too snarky, but you want the IESG to spend time documenting 
more precisely how often the arguments about the Updates field come up 
that they say they are spending too much time on?

I can (anecdotally) say that from my experience on the IESG, it occurred 
approximately every other telechat, and that there were at least a few 
messages prior to telechat and occasionally a few minutes of the 
telechat itself for each one of these. The arguments came down to that 
people were using "Updates" when they meant "See Also", which wastes the 
time of implementers who are looking for real information on how the 
supposed "update" might affect their implementation. The use of 
"Updates" as "See Also" have now become so common that it has diminished 
the value of the header, because it's anybody's guess whether one will 
find useful information by clicking the link or not, that it's worth 
abandoning the current "Updates" and starting with one or more new 
headers that will have useful semantic value.

The plural of "anecdote" is still not "data", but the fact that numerous 
ADs (current and past) have been annoyed by this, given the number of 
other things ADs have on their minds day-to-day, gives at least some 
indication that this is something more than a corner-case.

I think Suresh did a bit of work checking to see the present uses of 
"Updates". I think it would be hard to find data on how often and 
"Updated-by:" link was clicked on the tools site or the RFC Editor and 
the person who clicked on it yelled, "That's not an update! Why are you 
wasting my time!"

Many folks have said that their impression is this is a problem. 
Impressions can be wrong. Joe, if you'd like to go collect the data and 
show whether or not the impressions are wrong, have at it. But this is a 
lightweight enough change for what appears to be a useful feature that I 
think it's worth pursuing one of the plausible options, whether one 
finds the particular proposal or some other variant the most appealing.

pr
-- 
Pete Resnick https://www.episteme.net/
All connections to the world are tenuous at best
_______________________________________________
rfc-interest mailing list
rfc-interest@rfc-editor.org
https://www.rfc-editor.org/mailman/listinfo/rfc-interest