dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
28

funchords
Hello
MVM
join:2001-03-11
Yarmouth Port, MA

funchords to lolerskatez

MVM

to lolerskatez

Re: yay sandvine

Crap -- I often do that mistake. In that post, I meant 200? instead of 199?. Add 10 years to everything there.

DPI was not even a market segment in 199?

Thanks for pointing that out.

lolerskater
@rogers.com

lolerskater

Anon

To add one more interesting point to all of this. If comcast weren't scared to deploy the sandvine product in the critical network path, you wouldn't even know what was going on. Packets would slow down, and no one would be any wiser.

Instead, they put them in an offline situation (like most reporting software), and are required to use RST packets instead of just dropping packets outright.

I think you even identified this in your previous posts.

I am curious what happens when the RSTs stop being sent. What will everyone complain about then? There would be no packet forgery, just 'queueing'.

funchords
Hello
MVM
join:2001-03-11
Yarmouth Port, MA

1 edit

funchords

MVM

said by lolerskater :

I am curious what happens when the RSTs stop being sent. What will everyone complain about then? There would be no packet forgery, just 'queueing'.
Do you want to hear something funny -- they probably weren't too scared, they were too cheap. To install them inline would have required ordering a lot more.

Even on Comcast and pre-merger Adelphia, users knew something was going on. More than a year before my first message on this subject, users were questioning the unusual performance characteristics they were seeing and some had even questioned the RST packets. The only credit I can take in this manner is by coming up with a way to demonstrate it and to eliminate possible explanations until only the one remained.

As a result, Comcast has taught us a method to detect and prove ISP interference -- and it works even without overt evidence such as an RST packet.

Also, it taught the blogs and freedom advocates that even highly-paid high-echelon staff are fans of Network Neutrality and an open, free Internet. They despise their bosses' notions of fattening the bottom line by cheating the customer. And they're smart enough to anonymously expose it and even describe how it works.

Even so, the ISPs really are stupid enough to try this again. Between a public who is wary and watching, an FCC who is pissed about getting fooled, bloggers who love this story, and employees who work for the ISPs who actually are fans of the Internet -- they don't stand a chance.