Slashdot Mirror
IPv6 Application Competition - win $10,000
sneekz writes "The IPv6 Promotion Council of Japan has announced a competition for developers of IPv6-enabled applications. Various prizes up to $10,000 for ideas and actual implementations, and you keep the rights to your work. From their site: 'The contest will award developers of applications and software which helps to create new possibilities in the Internet world.'"
Better quit using TCP, then ...
:) -- and if their games are bandwidth hogs, maybe widespread QOS will force game-designers to write more efficient protocols.
unless you (and your ISP) are manually setting all of your TOS bits to 0, that is.
Protocols are mechanism, not policy. The reason why you haven't complained as yet of your IPv4 traffic being "slowed to a crawl" is that TCP's QOS features (minimal though they are) are typically only used when needed and ignored otherwise.
I would rather have the possibility open to shape traffic, should I want to, than to pre-emptively close the door because others might use it in ways I disapprove of. There are plenty of legitimate uses for QOS besides pissing off gamers (though I consider that an important activity in its own right
IPv6 is an attempt to return to the principles that gave the internet it's growth and democracy in the beginning:
Dumb network, smart edges.
When IPv4 was designed, there was no plan for exponential user growth outside of military/R&D/education. If there had been, addresses would've been 48+ bits from 1980 onward.
The failing with pre-existing networks which IP was meant to surmount is that the interior of the network was too intelligent. That sounds like a good thing, but it means that the network as a whole is less flexible- the inner nodes (routers) cannot be easily upgraded to support new applications and features. Under IP, all interesting computers are into hosts on the edge of the network. Each can be upgraded by an end-user, without supplication to the network templars- be they Bell Atlantic frame relay technicians, or Novell NOS admins. Those smart edges are served by a dumb cloud- the rest of the network just passes data from one place to another, without translating or modifying it in anyway. In the past, network application growth was slowed because users couldn't easily tell what was going on inside the cloud. IP made the cloud's job boring, so that you were no longer interested in seeing what went on there.
That change triggered the explosive growth of computer networks until they combined into the shared entity we all know and love.
NAT betrays this heritage
NAT boxes move intelligence back into the cloud- instead of IP packets being routed to the desired host and no other, there are now entities hidden in the cloud which waylay your packets. They seize them, pull them apart, inspect their innards- then, maybe, they'll deign to alter the packet and send it along further.
The damage isn't just a theoretical one- real end-users are being held back by NAT and other violations of the IP promise. New applications which would be easier to deploy with real per-host addressing are difficult or impossible to install reliably. This is things like high-speed game servers, file/web servers, P2P clients, cheap VOIP, videoconferencing, VPN, and prehaps things that haven't been invented yet.
The internet should be about giving power to the users on its edges. IPv6 would encourage that, but NAT hinders it. There are forces who don't want to empower users- major content providers and big ISPs. (Which may be the same thing). Fearful of losing control of mass audience's entertainment patterns, they want to keep mass creativity centralized. AOL doesn't want users to download ClickNRun IRC-like servers to create TeenTalkDaytonville chatrooms, they want to sell them as a value added service. Time Warner doesn't want 100s of cheap FTP servers passing out free copies of 56 year old TV shows (which by rights are public domain), they want you to wait for the DVD or PPV options.
The desire exists. A chicken in every pot, and permanent IP address in every study! The powers that be are fearful, though. The existing entertainment/datacomm oligopoly was harmed enough by the Internet. End-users sharing data amoung themselves could ruin them- but the exhaustion of IP addresses provided an excuse to keep end-users cordoned off from the real internet. They could download, but not serve files- as long as the people remain "consumers", the corporations can keep them under control.
NAT boxes bring the internet a tiny bit back towards the shape of traditional TV and telephone networks, which is just how big business likes it.