h a l f b a k e r yThe leaning tower of Piezo
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
BandwidthSwitcher
Create a central repository of wasted bandwidth to be used for short amounts of time. | |
Sometimes, people don't use all the bandwidth they are given.
For example, i am allowed to transfer 6 gigs of information per month at my domain, http://www.loveofcolor.org . Currently, I only use about 100 MB/ month. It would be great to be able to monitor my domain, and then let others use it when
they needed it.
For example, perhaps someone is going to get large media coverage for their website.. they would quickley contact the bandwidth repository, which would find people who had space and bandwidth available. The [movies/ images, files] would be transferred to my webhost, links changed and then when their site recieved massive traffic, they would do fine.
Kontiki
http://www.kontiki.com/ ... and plenty of similar P2P bandwidth-sharing schemes are all quite baked. [egnor, Aug 29 2001]
[link]
|
|
Unfortunately the bandwidth limits are calculated to already account for people not using their full quota each period. The technique has a fancy name ("statistical multiplexing", I believe) but it's the same technique that allows N+M telephones to share N long-distance telephone lines. It assumes that not everyone wants everything all the time. |
|
|
So your provider offers B bandwidth to each of N+M users. In reality your provider only provides N*B bandwidth (at a substantial cost savings) total to all users because they are betting that M*B of the promised bandwidth will go unused anyway. By allowing you B bandwidth instead of (B*N/(N+M)) bandwidth, your provider is already silently implementing the functional equivalent of a shared bandwidth repository. |
|
|
[to Big Brother's comment], When hosting companies say you can use X amount of bandwidth, you are allowed to use X amount of bandwidth.. I agree that they don't expect everyone to use all of their bandwidth- but if we do, we won't be charged. |
|
|
This system can work, easily- just code it in php/ coldfusion or ASP. |
|
|
Right, but his point is that they depend on not everyone using all their allotted bandwidth. That way, although they were paid for all of it, they don't have to pay their upstream for all of it, and make a profit. If everyone used all of their allotment, they wouldn't make any money. You don't get charged for using what you've paid for it. |
|
|
If this idea became widespread practice, bandwidth prices would most definately change. Not that it's a BAD idea, but it isn't going to work as is... |
|
|
This is the paradox of ceteris parabis. It doesn't work if
many or all members of the ISP use it. It does work if few
use it. |
|
|
In any case, freenet, gnutella and true peer-to-peer
models may prove these points moot. Broadcasting from
one computer to another (gnutella) and propagating
once's information across entire netowrks of individuals
(freenet) lessens the burden on any one server. In a
client-server world, there are limitations that only serve
to favor the ISP. In a client-as-server world, the benefit is
the publisher's. |
|
|
Is that publisher as contrasted with author? Or publisher as contrasted with sponsor? How does a publisher/author fare under the new rules? Do they get any relief from the current epidemic of $ales pollution? What is gnutella, anyway? Does it have any non-napster-like functions? I've been trying to find any. |
|
|
Let's use Napster as a model. As a user you have the
option of downloading and sharing files. It's design permits
you as a consumer of a song to also be a publisher of that
song. That is, as soon as you download it, it can also be
shared. |
|
|
So "publisher" in the old definition of an agent that
disseminates written or recorded works still aplies. The
"producer" of that work is the artist. And inteh case of
Napster, an artist can produce a work and publish it both.
So essentially the network of fans becomes a network of
publishers and the artist has all of the sponsors he/she
needs. |
|
|
As far as sales pollution goes, you hit on an interesting
point. Since the first usenet SPAM to today, there is no
real way to limit uninvited sales pollution in what amount
to open systems. A music file disguised as a Pearl Jam song
could just as well be a 4 minute commercial for Pearl's
Jam. However, once the fraud is uncovered, the user
usually chooses to trash it, instead of increasing the value
of that fake song by also publishing it. |
|
|
Gnutella is a protocol, not software. It is a way that
infomation is shared and transferred between devices
that have internet access. Limewire uses the gnutella
protocol to allow peers to find and share any electronc
file. Programs, music, pictures, text. Litterally anything on
your computer can be shared with any other computer
running a gnutella client like Limewire. |
|
|
Non-napster? Well, limewire is pretty mucht the same
thing as napster in use. The difference under the hood is
that there is no central source making the matches
between peers. That is, napster had central servers that
routed requests for songs. ONce you found a peer with
the song you were looking for, the central server got out
of the way and let you transfer the files directly with a
peer. With limewire/gnutella, you bounce your requests
from your closest peers on the the gnutella network to
their closests peers and so on. Each degree of separation
is called a "hop" and if you want to extend the number of
hops, you can potentially search all the peers using the
gnutella network. As a peer yourself, you are both routing
searches for other users as well as potentially sharing files
with people who found something you have and they
want. |
|
|
go to http://download.com and search for limewire. It's
fee to download. |
|
|
". A music file disguised as a Pearl Jam song could just as well be a 4 minute commercial for Pearl's Jam. However, once the fraud is uncovered, the user usually chooses to trash it, instead of increasing the value of that fake song by also publishing it. " |
|
|
This happened, pretty much exactly like that. Barenaked Ladies put out a file that was a short sample of the song, then them saying 'This was a sample, if...' |
|
| |