Re: [RTG-DIR] [babel] [Last-Call] RtgDir review: draft-ietf-babel-information-model-11.txt

[Geoff Huston <[e-mail]>]

I’m was not raising an issue with the use of units of one tenth of a second - I was expressing the thought that the term “centisecond” is an unusual term in English, and as a native speaker I haver also followed a convention of using the units of “tenths of seconds”, “hundredths of a second” and then heading to milliseconds, nano seconds. Until now I had never seen the term “centiseconds” and I would claim that I am familiar with scientific English.

Its as if I used terms such as furlongs, chains and perches in an RFC - yes they are =units of measurement with a defined meaning but its not a commonly used term and it would look strange. I have a similar reation of centisecond (and decisecond).

But it's not my RFC and frankly I don't care all that much. The review template wanted my nits with the document so I provided my nits. I think the string nit is more critical than the one about the latin prefix for seconds. Why define some intergers as unsigned 16 bit values yet leave strings of undefined maximum length. Can I stuff a novel into the field? How about all of wikipedia? The point is why is the document inconsistent about its treatment opf the range of acceptable values in terms of size?


> On 20 Oct 2020, at 11:49 pm, Juliusz Chroboczek <[e-mail]> wrote:
> 
>>> "centiseconds"? Really? "tenths of seconds" would be clearer to most
>>> readers!
> 
>> Perhaps this was intentional, but a centisecond is a hundredth of a second,
> 
> The wording is consistent with Section 4.1.2 of draft-ietf-babel-rfc6126bis:
> 
>   Relative times are carried as 16-bit values specifying a number of
>   centiseconds (hundredths of a second).
> 
> It then goes on to justify this somewhat unusual choice of units:
> 
>   This allows times up to roughly 11 minutes with a granularity of 10ms,
>   which should cover all reasonable applications of Babel
> 
>> maybe spelling it out as you suggest is reasonable.
> 
> I agree, a parenthetical remark to define the unit would be helpful,
> simlarly to what is done in rfc6126bis.
> 
> -- Juliusz