Re: [rtcweb] Prioritization

[Simon Perreault <[e-mail]>]

Le 2014-04-25 11:02, Harald Alvestrand a écrit :
> On 04/25/2014 03:53 PM, Simon Perreault wrote:
>> Le 2014-04-25 09:21, Harald Alvestrand a écrit :
>>>> Suggestion: leave it up to implementations to interpret priority levels
>>>> however they want. Reword the current text so that it becomes an
>>>> *example* of what an implementation might do.
>>> The problem with doing that is that it leads to completely inconsistent
>>> behaviour.
>> Understood. But is that something we should care about? I mean, if it
>> does not lead to application developers doing "if (chrome) ... else if
>> (firefox) ..." then behaviour inconsistency is of no consequence. I
>> can't imagine a situation where that would happen if we leave it open.
> 
> I can imagine that happening. Consider a scenario where complete
> starvation is considered a Really Bad Thing.
> 
> if (firefox) {
>    // higher priority channels will completely starve out lower priority
> ones. Don't.
>   priority = normal
> } else if (chrome) {
>    // lower priority channels will get some packets through. That's OK.
>    priority = high
> } else {
>    // I have no idea what will happen, let's play it safe.
>    priority = normal
> }

Ok, then how about "MUST NOT do starvation"?

I'll shut up now. I see your point. If nobody else feels this issue is
important, feel free to ignore my comment.

Simon
-- 
DTN made easy, lean, and smart --> http://postellation.viagenie.ca
NAT64/DNS64 open-source        --> http://ecdysis.viagenie.ca
STUN/TURN server               --> http://numb.viagenie.ca