[opensource] Support Network Neutrality!
hurley at todesschaf.org
Fri Jun 9 19:40:34 EDT 2006
On 9 Jun 2006, at 20:01, Jim Dinan wrote:
>> Imagine Google services loading slower because Yahoo! paid your
>> internet provider to favor it's information. Or imagine the videos
>> at www.youtube.com <http://www.youtube.com/> slowing to a crawl
>> because your telecom has a deal with some media conglomerate to
>> deliver primarily "media conglomerate friendly" videos (read: pop
>> videos and advertisements).
> Any idea how they intend to get the big networks to support this
> technology? Or is the packet prioritizing only supposed to be
> between ISP and end-user?
How do you mean "get the big networks to support this technology"?
All the major ISPs are on board with this, meaning basically the
entire internet will be prioritized depending on who your ISP decides
to let through the fastest. Generally speaking, this will mean one
thing: who pays the ISP the most.
This could come about in one of two ways, (1) the content provider is
connected to the internet through the same ISP as you are, in which
case the ISP doesn't have to bother with peering in order to get the
data to you, or (2) a content provider is connected to ISP1, who has
a peering agreement with ISP2, who happens to be your ISP. ISP2,
feeling they're getting the short end of the peering agreement with
ISP1 (who feels, in turn, that they're getting the short end of the
peering agreement), decides to give lower priority to all traffic
coming into their network from ISP1, slowing the data from any 2
nodes connected node1<->ISP1<->ISP2<->node2 to basically a crawl.
However, a large content provider (whose ISP is ISP1) decides that
they don't want this, so they make an agreement with ISP2 that ISP2
will give their content regular (or even higher than usual) priority,
so ISP2 changes the ACLs on their routers to say "All traffic from
ISP1 goes slow UNLESS it comes from this large content provider's
netblock". So, effectually, the content provider ends up paying twice
for access to one internet. ISP1 doesn't care that their customer is
paying more, because they'll be getting money from content providers
in a similar situation who have ISP2 as their ISP.
Of course, when one of ISP2's customers notices that their competitor
(connected via ISP1) is getting better bandwidth than they are,
they'll complain to ISP2, and ISP2 can say "well, ok, we'll give your
traffic priority if you pay us more money". In other words, the ISP
can't lose, because content providers HAVE to be connected to the
internet, and generally speaking, in any one geographical area, there
is only one, MAYBE two competing ISPs.
Basically, what's happening here, is that the ISPs are segmenting the
internet so that they can make more money. They all win, because the
big money content providers won't be cool with their traffic going
slower, so they'll pony up the bucks.
Oh, and let's not forget the other provision of HR 5252, making ISPs
into national franchises, similar to what happened with the gasoline
industry some years ago. What was *supposed* to happen with the
gasoline industry was make things cheaper for people who were buying
gas in an area that only had one gas station available (like on the
highway in the middle of nowhere, Kansas). This was supposed to
happen because the government said "you have to sell gas for
basically the same cost everywhere (within a certain margin of
error)". What *really* happened was, instead of dropping prices to
the lowest price (as the government said would happen), the gas
companies raised all their prices to around the highest price they
had. This *exact* same thing is happening with the ISPs now. The
government wants the ISPs to cut their prices so that people in the
areas where internet access is prohibitively expensive can get
internet access at a sane cost. What will most likely happen if HR
5252 passes the senate (it passed the house last night) is the same
thing that happened with gasoline - namely, the ISPs will raise the
cheaper prices instead of lowering the higher prices, because they
aren't allowed to have more than a small differential in service cost
across the nation.
OK, I think I've rambled on enough (and god only knows if I've even
answered the original question).
I'm looking for something that can deliver a 50-pound payload of
on a small feminine target. Can you suggest something? Hello...?
More information about the Opensource