W3C home > Mailing lists > Public > public-webrtc-logs@w3.org > January 2017

Re: [webrtc-stats] Define terminology for "stats object" et al.

From: Harald Alvestrand via GitHub <sysbot+gh@w3.org>
Date: Wed, 25 Jan 2017 09:12:16 +0000
To: public-webrtc-logs@w3.org
Message-ID: <issue_comment.created-275055989-1485335534-sysbot+gh@w3.org>
I observe that you disagree with the overall direction.
The bugs filed came from the problem that we sometimes need to talk 
about a stats value changing - in a way that is a bit shorter than 
"the difference between the attributes of two Stats Objects that have 
the same "id" value". This led to the questions raised in the 
specified bugs.
Another approach would be:
- Say that each object in the object model of WebRTC has some 
associated stats values (some of the things that have stats values 
aren't in the object model, but if the approach makes sense, we can 
make our way around that)
- State that a stats object (of a specific stats type) is the snapshot
 of the stats values of the object at a specific moment in time (the 
timestamp)
I think this would be a smaller textual change, and might achieve the 
same goal of being able to talk about a value changing (on the 
stats-owning object) while still being clear that a returned value 
never changes.

Does that sound workable?

-- 
GitHub Notification of comment by alvestrand
Please view or discuss this issue at 
https://github.com/w3c/webrtc-stats/pull/139#issuecomment-275055989 
using your GitHub account
Received on Wednesday, 25 January 2017 09:12:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 4 June 2019 15:32:41 UTC