[ NNSquad ] Re: [IP] Does AA VoIP usage violate the "federal Internet policy"

[Rahul Tongia <tongia@cmu.edu>]

Brett has raised an interesting idea of signaling. [I ignore his claim 
of most p2p being stealing]

I recently was introduced to foldershare, an encrypted p2p application 
for synchronizing files. Unlike skype in audio mode, this can take up 
enormous bandwidth. 

So, instead of just the network signaling peak vs. off peak, etc., apps 
need to adapt to such issues as well.  Certainly p2p applications that 
come on whenever online can be an issue.  But I can count on so many 
more.  Is a simple hierarchy of tagging "active" vs. "passive" bandwidth 
consumption sufficient and appropriate? What about timers (e.g., how 
long can I delay downloading a large patch)?  I could envision something 
that can sense spare capacity and adjust accordingly - but to work well 
it needs mechanisms for granular, honest, and updated information on 
this. Will carriers help?

When I travel, I'm sometimes on a VPN (or dial-up), and I can see my 
total MB uploaded/downloaded. I would wager that the bulk of my 
bandwidth is fluffy graphics, ads, background/passive etc.  Dave farber 
had asked about a widget for measuring that - did anyone find anything 
that could add more value of not just total up/down but also the 
specific app/port?

If we're debating NN for broadband, mobile wireless is a great example 
of where "unlimited" really doesn't mean unlimited. And where consumers 
are more accepting of such restrictions.  So if I accept restrictions on 
ports/usage on a wireless device, is it not fair there could be similar 
(or different) restrictions on more traditional broadband? Here, I am 
differentiating between usage caps and other restrictions. Just thinking 
out loud...

If we accept that a plane satellite/wireless link is bandwidth 
constrained, what are fair mechanisms to share the bandwidth? 
proportional capacity? Pay more get more/priority? Port/app 
restrictions? Resets or technical "fixes" that gum up certain 
applications or services?  I also don't know the answer to this issue.

Rahul

************************************************************************
Rahul Tongia, Ph.D.
Senior Systems Scientist

Program in Computation, Organizations, and Society (COS)
School of Computer Science (ISR) /
Dept. of Engineering & Public Policy

Carnegie Mellon University
Pittsburgh, PA 15213 USA
tel: 412-268-5619
fax: 412-268-2338
email: tongia@cmu.edu
http://www.cs.cmu.edu/~rtongia



Brett Glass wrote:
> At 02:10 PM 9/16/2008, Lauren Weinstein wrote:
>  
>   
> > I have not yet taken a "formal" public position on the issue of phone
> > calls from planes.  This seems to be an issue that has only come to
> > a head with the threat of largescale phone usage -- it did not seem
> > to be a big controversy when restricted to (expensive and limited)
> > use of Airfone services.  
> >
> > However, blocking of protocols is going to open up the same can of
> > worms we're dealing with now in the ground-based Internet.  Will
> > people try to do high-traffic P2P from the air?
> >     
>
> You bet. In fact, they may not be able to help it. P2P apps (the
> authors know they're stealing anyway, so why should they give a fig
> about the user's wishes?) often start up in the background every 
> time the computer boots, without notifying the user that they will 
> be eating his or her bandwidth. 
>
> In any event, this situation reflects the vagueness of the FCC's
> recent ruling against Comcast. What is permitted? What is not?
> Under what conditions? The FCC, ignoring the Constitutional prohibition
> of vague laws, doesn't say. In fact, no one knows, at this point, whether 
> ANY network management practice might rile the FCC, and small broadband
> providers such as myself are worried that we might be next to be 
> pilloried, without warning, as was Comcast. See my filing at
>
> http://tinyurl.com/5gfn6p
>
> for 20 more situations in which no one knows how the FCC would rule.
>
> --Brett Glass
>
>
>
>