Schaller: Well then you’re arguing about nothing, because i’ve taken up no other position.
Howk: Or at least the original idea for it.
Dorazio: Other than the fact that your characterization of net neutrality as some kind of free market solution is bull****.
Kofahl: I think qos is harmful.
Coshow: Currently they prefer certain types of traffic like voice and video, etc.
Overson: But that means they have to know that it is voice or video.
Astry: Anyway, i’ve got to bounce.
Pelayo: Which means they need to continuously update that software.
Godde: If someone develops a novel form of compression for either, then even considering gains from compression or whatever then it will not get qos’d.
Hirko: Or another format entirely, whatever.
Zadroga: The point is that it duplicates responsibilities.
Garofano: Just let it be slower.
Bittle: Improve the network elsewhere.
Lightford: You can get insanely fast connections in many countries.
Sucher: I was thinking denmark.
Schnitzler: And other countries like that.
Deschamp: That’s a pretty weak argument. Any network is constantly changing and always needs to be updated. There are new protocols like ipv6, etc.
Deschamp: Qos is more about latency, not bandwidth.
Zogopoulos: That doesn’t mean we should just add more and more crap to it.
Nickell: And duplicate effort everywhere.
Thonen: I still think it’s a false sense of speed.
Horney: Dubai ment to have fast internet.
Horney: Hong kong best in the world.
Elser: There was a ranking i saw recently.
Hollings: I forget the link though.
Efron: The u. S. And canada are **** though.
Stuemke: It’s still a heuristic though, and not a great solution.
Casiano: There is no standardized way of doing it.
Biren: Compressed frames can still be qos’d by just refering to ip dscp.
Deschamp: Why should everyone be forced to spend money for expensive hardware when software can fix it easily? Surely that would have a bigger negative impact on the overall economy?
Biren: However similar to the ‘evil bit’ one might consider for internetwide end to end qos.
Siaperas: That’s a legitimate argument.
Alaibilla: I still think qos is brittle though.
Biren: Can we have a meta converstion about why qos gets to have a little tiny o?
Guarnera: If there were an open standard that everyone agreed on.
Calver: Then i would agree with it.
Biren: See wikipedia discussion page on qos.
Ryhal: Arbitrary capitalization rules.
Littrell: For example if the ietf came up with a way to speed up video and voice traffic using free formats of course then sure.
Domanski: Because then it’s fair.
Rainge: And it’s standardized.
Biren: Using free formats of course.
Foresman: Because the rfc would have to tell you how to implement those too.
Biren: The open standards exist.
Blaze: I see a qos working group that concluded in 1997.
Biren: It’s literally built in to the ip stack.
Biren: The whole problem is about governance and scalability.
Biren: Ultimately, who are you to tell me what to do? Router.
Biren: How does one prevent malicious tagging on a wild landscape? Where do you set your trust boundaries?
Biren: As far as protocol inspection itself it’s fairly trivial to identify rtmt type traffic with simple protocol ****yzers.
Biren: This isn’t really an argument that is a protocol discussoin.