Did Internet Founders Actually Anticipate Paid,

Jack Bates jbates at brightok.net
Fri Sep 17 15:11:13 UTC 2010


On 9/17/2010 9:49 AM, Joe Greco wrote:
> So if I'm now downloading my latest FreeBSD via BitTorrent, my basic
> expectation is ultimately that I'll get fair treatment.
>
And this is always a debate. You might say letting someone with voice or 
video have queue priority during saturation as being unfair, yet your 
p2p will work when it's running slower, where as their voice or video 
might fail and be completely unusable.

If a provider has to deal with saturation, they have to make such 
decisions. Their goal, ideally would be to have a majority of the 
customers able to do what they need to do during saturation, allowing 
traffic to slow down that can afford to, and giving priority to traffic 
that to be usable must maintain certain QOS.


> What's "fair treatment" though?
>
> I think at the end of the day, it means that I've got to have reasonable
> access to the Internet.  That means that if I can get packets in and out
> of your transit without fuss, that's probably good enough.  If you've
> short-circuited things with peering that gives me faster access, that's
> great too.  However, if your transit is 100% saturated for 20% of the
> day, every day, that's NOT good enough.

I'm all for rules to limit saturation levels. This has nothing to do 
with neutrality, but to me it is the more important point. Consider 
telco world and voice communications. I could be wrong, but I seem to 
recall there be rules as to how long or often or what percentage of 
customers could experience issues with getting a line out.

I'm also a strong believer in enforcing honest business practices. If 
you sell prioritization to one company, you should offer it to all 
others. The practice itself doesn't scale, so given an all or nothing, 
it is a business model that will burn out.

The short-circuits and QOS applications are just methods of improving 
service for a majority of customers (those who use those 
destinations/services). This means, of course, p2p will usually be the 
loser. As an ISP, p2p means little to me. The QOS to the sites that hold 
a majority of my customers captive is the issue. Even without 
saturation, I have an obligation to see how I can improve video and 
voice quality in an erratic environment. This includes dealing with 
things such as microbursts and last mile saturation (which for me isn't 
shared, but customer's run multiple applications and the goal is to 
allow that with a smooth policy to assist in keeping one application 
from butchering the performance of another, ie, p2p killing their video 
streams from netflix/hulu/cbs/etc).


Jack




More information about the NANOG mailing list