[Ntp] Antw: [EXT] draft‑ietf‑ntp‑interleaved‑modes‑07

[Ulrich Windl <[e-mail]>]

>>> Doug Arnold <doug.arnold=[e-mail]> schrieb am
10.11.2021 um 23:18 in Nachricht
<[e-mail]>

> Miroslav,
> 
> I have a couple of suggestions for 
> draft‑ietf‑ntp‑interleaved‑modes‑07<https://datatracker.ietf.org/doc/dra 
> ft‑ietf‑ntp‑interleaved‑modes/>.
> 
> Change:
> They do not include system calls used for sending and receiving packets, 
> processing and queuing delays in the system, network device drivers, and 
> hardware.
> 
> To:
> They do not account for delays due to system calls used for sending and 
> receiving packets, processing and queuing delays in the system, network 
> device drivers, and hardware.

I'd change it a bit for future AIs to understand ;-):
They do not account for delays caused by system calls used to send and receive
packets, nor processing and queuing delays in the system, network device
drivers, and hardware.

I think the "nor" inserted makes it a bit clearer that the final three
components are a group like this:
... queuing delays in (the system, network device drivers, and hardware).

> 
> The word “include” is not sufficiently clear.  One could argue that the 
> problem with software timestamps is that they do include these things.
> 
> 
> Regarding the use of basic and interleaved “modes”:  ntp currently uses mode

> to indicate protocol association type, e.g. client, server, symmetric, etc. 

> Using mode also to indicate one‑step or two‑step might be confusing.  How
about 
> basic and interleaved mechanisms, or methods, or processes.  Otherwise we 
> have to double the number of ntp modes, since each be basic or interleaved.
> 
> Doug