NNSquad - Network Neutrality Squad

NNSquad Home Page

NNSquad Mailing List Information

 


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ NNSquad ] BitTorrent bandwidth usage (from IP)


From: David Farber <dave@farber.net>
To: "ip" <ip@v2.listbox.com>
Date: Wed, 27 Feb 2008 06:23:28 -0800
Subject: [IP] Two positions at  FCC Comcast Hearing

________________________________________
From: Brad Templeton [btm@templetons.com]
Sent: Tuesday, February 26, 2008 6:50 PM
To: David Farber
Cc: ip
Subject: Re: [IP] Two positions at  FCC Comcast Hearing

> Note: Richard Bennett<http://www.bennett.com/blog> who was an expert panelist at yesterday?s hearings informed me that BianRosa claimed that BitTorrent didn?t exceed the contracted limit.  That however ignores the explicit ?no server? clause in the terms of service and no broadband service was built to be fully saturated 24×7.  This is why commercial grade T1 lines that offer less than half the speed of broadband connections costing 8 times less are $400 per month.
>

Actually, the prime reason T1s cost that much is the overpriced local loop.
When purchased at well-connected data centers, megabits of saturated bandwidth
are a great deal cheaper than the price quoted.

However, I feel it is important to point out that nobody denies that P2P
applications, especially Bittorrent, generate a lot of bandwidth usage,
more than most or all other applications.   This is not news of any kind.
(I am a director of BitTorrent Inc, which develops P2P software, though not
acting as a spokesman.)

There will always be low-usage applications, and high-usage applications,
and I think it's safe to claim there will always be a highest-use appliciation
which goes far, far beyond the average.

In this case, with BitTorrent, users trade their spare upstream
bandwidth -- which in many cases, such as the typical DSL ISP is otherwise
going unused and wasted -- to other users in exchange for their
upstream bandwidth in return.   (Or, in a "pay it forward/golden rule" situation,
they sometimes just do it out of philanthropy or in the hope of promoting a system
where they will be rewarded later.)  It is commonly incorrectly stated that
this is done to benefit the 3rd party (such as ubuntu.com) but the trade is
really mostly among the users.   The seed gets no means to reward tit for tat.

What is often missed is the question really comes out of this concept of
the user having spare upstream bandwidth.   Most ISPs sell a flat rate,
upstream package and as such the bandwidth is sold to the customer and
is theirs to use to further their usage of the internet.   In the case
of DSL, the upstream is truly otherwise unused and is lost forever if not
used.  With DOCSIS and wireless ISPs this is not as true.

Some ISPs want to claim you don't really have any spare bandwidth to trade,
that they didn't really sell it to you, that it is theirs, not yours,
in spite of what they advertise.   If so, there have been calls for them
to be clear in their advertising about these limits.

However there remains a deeper issue.  As I noted, there will always be
a heavy-use application at the flat end of the bell curve.  The 90-10 rule
will probably always apply.   Should we be concerned with a regimen that
wishes to "solve" that "problem" by beating down at whatever new
innovation becomes popular enough to be the heavy user, with application-specific
tricks such as protocol detection and forged resets?   Where does this
lead us?

Many value an internet where the smarts are in the "ends" and people come
up with clever new apps that use bandwidth to meet user desires.
In this case, people want big files.   Yes, because Bittorrent is the best
technology for publishing big files, it is used by infringers -- why would
they not seek out the best like everybody else -- but big files will be
sent, both for legit and infringing uses.   When a P2P cloud gets large
and has people exchanging data within a LAN, it actually reduces the bandwidth
load on an ISP compared to the traditional central server "hub and spoke"
approach.

So let's not argue about who is using the most bandwidth, but instead decide
how to set up an internet where there will always be a heavy bandwidth
user, and how to regulate that, if it needs regulation at all.

-------------------------------------------
Archives: http://www.listbox.com/member/archive/247/=now
RSS Feed: http://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com