Neutrinos point to a four-year-old experiment with QoS at Internet2 as proof that QoS can never work. The report on this experiment is very critical of the network equipment of the day, and clearly biased in favor of over-provisioning as an alternative to QoS. Yet the report’s author doesn’t support current efforts to enact network regulation in the name of Net Neutrality:
I have not seen the bill that is reportedly shaping up. However, what I know of it indicates that it is likely to do much more harm than good. A good bill needs to have two things: (i) sunset provision and (ii) possibility to price the network in an economically efficient way. It appears, from what we know so far, that the bill will permanently prohibit economically efficient pricing and mandate a flat pricing structure, where users pay a monthly fee only. This would be extremely harmful, as it would instill the incentive to regulate the amount of traffic a user sends by giving him a tiny straw instead of a fat pipe. It would be tragic if we started with the belief in fat cheap network pipes and ended up outlawing them.
[Update, May 2, 2006: It appears that the draft Steven’s bill satisfies both conditions: it does have a five-year sunset provision and does not prohibit economically efficient pricing. It appears to be very good, at least for now.]
The Stevens bill does not enact “neutrality” provisions.
He writes: “Tens of dollars a month can buy a 1-Gb/s connection; at such speed, no congestion will happen.”
10 million compromised machines X 1 Gb/s connections = potential for targeted victims to see lots of congestion.
This is what these folks never get. I like to simplify by describing the Internet as hub and spoke. If the hub is faster than the spokes, no congestion (unless multiple spokes are streaming to a single spoke, um.)
But every time you speed up any spoke, you have to speed up the hub by an equivalent amount or you get congestion. And any time any one spoke is faster than any other spoke, you have congestion.
So the arms race continues until the network has more capacity than all the attached devices can use if running at full speed at the same time. That would be lots of bandwidth.
Yes, as we all know, the Internet has ground to a halt over the past few years through sustained attacks from the million-plus DDOS zombies controlled by cyber-criminals.
Oh, wait, that’s right – there are products out there to handle that – like Cisco DDOS Guard — which are used heavily by Pipex for its casino businesses.
DDOS attacks love Net neutrality, dirblue. If the law says “all packets are equal” than the DDOS packets are just as decent as all the others; same for spam, phishing, etc.
What does the casino have to do with this? It looks like you’re trying to show off.
I agree, and am confused as to what the casino has to do with anything? Please clarify your statement so a proper response to this absurd argument can be rendered.
I’m saying the Internet has suffered from massive, broadband, distributed denial-of-service attacks for years (e.g., http://it.slashdot.org/article.pl?sid=03/11/12/1923203&tid=172&tid=187&tid=95%3Cbr%20/%3E )
An example is the blackmailing of a major online casino a while back. The blackmailer controlled a botnet (army of zombie computers), most of which were broadband-connected. He directed the bots to attack the casino just before the Superbowl when the owner refused to pay the blackmail.
When this occurs, ISPs will now use technology products to isolate and repel the traffic. The Cisco DDOS guard is one such product.
I had just assumed everyone had heard of these cases. My bad.