NNSquad - Network Neutrality Squad

NNSquad Home Page

NNSquad Mailing List Information

 


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

[ NNSquad ] Re: Broadband Infrastructure (from IP)


Lauren Weinstein wrote:
You, me, and many others have made this point over and over again.
A key reason that there's so much skepticism about ISP deployment
promises is that we've been screwed on this score repeatedly.  Unless
there are ironclad requirements for truly neutral, open access
deployments, there's no good reason to assume that ISPs will behave
any differently than they have in the past.  To wit: cherry-picking
locations, giving themselves the lion's share of bandwidth for their
own content, blocking ports, arbitrarily prohibiting servers and other
applications, and nowadays even modifying user data, spying on user
data, trying to charge unaffiliated Web services for access to
customers, and of course implementing arbitrary and anticompetitive
bandwidth caps.

Any new public funding/tax incentives for ISP broadband deployment
should be predicated on a reasonable regulatory approach, including
disclosure, transparency, neutrality, and an understanding that we're
not going to throw money at the ISPs the way we just did with the
financial services sector.

Agreed 100%. In fact, I'm a lot happier with this idea than with flat-out regulation. Here, you can have this subsidy for setting up cable to more places, provided:
. You reach at least X% of the households in the designated geographic area, with Y% signing up.
. At least X% of the cable bandwidth is allocated to Internet access (vs. TV cable, PPV, etc.)
. Bandwidth caps must be demonstrably reasonable based on median and 90th percentile usage.
. Protocol and port-agnostic carriage. [Exceptions: users must explicitly opt-in to allow access to the Windoze filesharing and to allow _outgoing_ connections to port 25. In the latter case, users explicitly agree to accept responsibility for spam sent from their machine.]
. No modification of data sent to or from the user's machine except for:
+ those explicitly allowed by the RFCs (e.g., decrementing TTL, changes involved in packet splitting)
+ caching,
. Users are allowed to run any application -- specifically including servers -- except those things generally recognized as malware (spam or spim generators, botnet directors, port scanners without explicit consent of the scanned site, attempts to exploit known security holes). [Note: I would have no objection if users have to opt-in before remove access to servers is allowed, as long as there is no additional charge for this. That protects the vast majority of users from having their systems exploited in various ways, at a small inconvenience (one phone call) to the relative few who actually _do_ want to run web/ftp/filesharing servers.]