Re: [rtcweb] DTLS, DTLS-SRTP, and 5-tuples

[Christer Holmberg <[e-mail]>]

Hi,

>> I think there is some confusion between the underlying link protocol 
>> used by candidates and the protocol running on top of ICE. I do not 
>> think TCP is defined as a protocol that can operate on top of ICE. It 
>> is defined in RFC
>> 6544 using framing from RFC 4571, as a protocol for underlying links.
>> Whatever protocol is defined on top of ICE should support operation on 
>> top of packetized communication protocol provided by ICE and should 
>> define how its packets can be demuxed from STUN packets used for ICE 
>> connectivity checks. So far things like RTP, DTLS, and SCTP fit such 
>> requirements. New things can be defined in the future. When they do, 
>> they should treat ICE a virtual communication channel that provides 
>> unreliable packet transport with no order guarantees which can span multiple 5-tuples.
>
> Exactly. Please Christer, don't go to layer 4 protocols. Neither IPv6 can run over an ICE initiated "virtual transport" ;)

I am not the one saying "whatever protocol" :)  

My point is that we need to be more clear about the scope - we can't just say "Take whatever protocol and do a "5-tuple"/"virtual connection" search-replace." :)

Regards,

Christer