News for hackers. Stuff that matters.
Re:I'm on the FCC. Please stop us from killing net neutrality by Joeri on Thursday November 23, 2017 @03:26AM (#15763689)
Even the language is wrong. Getting rid of net neutrality doesn’t introduce fast lanes, it introduces slow lanes. Nobody will get faster service than today, but some will get slower service.
Re:I'm on the FCC. Please stop us from killing net neutrality by derefr on Thursday November 23, 2017 @03:34AM (#15763734)
Some people could get faster service than they do today, in the sense that a given flow (e.g. VoIP data) might go from having equal QoS priority to having higher QoS priority, meaning that under congestion, their flow might go from being slow (before) to being the-speed-they're-paying-for. Of course, that would come at the expense of other flows (e.g. BitTorrent) having lower relative QoS priority, making them slower under congestion. It's zero-sum.

This type of flow management is already how VoLTE flows are prioritized on LTE (see https://en.wikipedia.org/wiki/IP_Multimedia_Subsystem); I wouldn't be surprised to see land-line ISPs implement it.

Re:I'm on the FCC. Please stop us from killing net neutrality by nerdponx on Thursday November 23, 2017 @05:25AM (#15764156)
It's zero-sum

Not so fast. It's zero-sum if and only if call Internet subscribers have the same ability to pay for the service. If any user has a very high willingness to pay, but does not have the resources to pay it, then this amounts to a transfer of "wealth" (i.e. bandwidth) from that user to another who might not value it as much. It's not hard to construct utility curves in which this decreases total welfare.

That said, QoS is probably not the concern here. This is very much not a zero-sum proposal.

Re:I'm on the FCC. Please stop us from killing net neutrality by Retric on Thursday November 23, 2017 @06:23AM (#15764369)
Only with a poorly designed routing, networks, and VoIP systems. Routers can't cache meaningful amounts of network traffic, so assuming ~zero dropped packets you degrade bandwidth not latency. VoIP works fine with even minimal bandwidth so that's not an issue.

Dropped packets are a sign of poorly designed networks at which point you can't assume QoS will improve things. The added complexity might make things worse for all traffic.

Re:I'm on the FCC. Please stop us from killing net neutrality by derefr on Thursday November 23, 2017 @04:10PM (#15767815)
Keep in mind, cellular carrier stations (cell towers) don't just become congested; they're oversubscribed, pretty much all the time. Each level of cell service has essentially an SLA for minimum voice quality per line—which can be modelled as each customer having a fixed bandwidth allocation. So, at a given service tier, you can sacrifice latency (which doesn't work for real-time protocols) or drop packets (which is what ends up happening), but you can't just lower the available bandwidth of the channel. The [hardware!] voice codecs are built to assume the bandwidth when they're on a given cellular network standard—they're essentially emulating circuit-switching!

All that being said, that's why cell networks, until 4G, used actual circuits for voice, rather than having it be packet-based; and that's why 4G+ networks use IMS (a layer 1+2+3 protocol suite that all middle boxes are aware of for QoS purposes) rather than just relying on layer-3 https://en.wikipedia.org/wiki/Differentiated_services QoS fields (which are handled by full layer-3 switches within a given LAN, but get dropped the moment the packet crosses onto the public Internet and are just ignored by layer-2 switches.)

(And even on large LANs—like your cable provider's ATM network—DiffServ tags aren't relied upon, because QoS is just so much easier when layers 1 and 2 are aware of it. If you get a IP phone line from your ISP, it'll be using https://en.wikipedia.org/wiki/Multiprotocol_Label_Switching tags instead, to give its packets QoS priority across even the dumbest middleboxes.)

Re:I'm on the FCC. Please stop us from killing net neutrality by Retric on Thursday November 23, 2017 @05:11PM (#15768143)
That's more an abstraction than reality. Frequency hopping means N senders may overlap but the abstraction is independent the IP part of TCP/IP or UDP so the device knows things failed instead of waiting for a timeout on a dropped packet. Thus IP software never sees latency spikes just lowered bandwidth or dropped connections.
Re:I'm on the FCC. Please stop us from killing net neutrality by Retric on Friday November 24, 2017 @10:14AM (#15771262)
To clarify, in the past the over subscription might have required dedicated bandwidth for voice but that's no longer the case.

Voice is very low bandwidth as long as the minimum bandwidth per device is over 100 kpbs and the device to prioritize voice data towers don't need to bother with QoS for traffic. They still do this for various reasons, including the fact they are sending the voice data to a separate network not the internet.

TLDR; Total bandwidth / devices > 100kbps and it's not actually over subscribed.

Re:I'm on the FCC. Please stop us from killing net neutrality by Fnoord on Thursday November 23, 2017 @05:39AM (#15764204)
The net neutrality in the EU isn't perfect, but it does allow for QoS for things like VoIP.

EDIT: Example: Free Music Streaming by T-Mobile [1] is active in EU as well. So if you stream from Spotify its "free" yet when you stream from your home server, via a VPN, or an unsupported service it isn't "free".

[1] https://techcrunch.com/2014/06/18/t-mobile-stops-counting-da...