| Quote: |  | |
| The landmark move to the networking technology upgrade, known as IP Version 6 (IPv6), will open the door for billions more devices to be connected to the internet.
In turn, this will trigger a boom on top of the $2.4 trillion worth of commerce and banking transactions already carried out on the internet. | |
| | |
I have to disagree; I don't see IPv6 as something that will be any great revolution. Sure, it's a good thing... but something that is going to cause a $2.4 trillion boom? I think not.
My reasons:
Proven technologies like NAT and UPnP can already open up the possibility for all your home devices to be directly addressible from the outside world. The need for all your gear to have a public IP simply isn't there. In fact, it's quite the opposite: hiding your LAN behind a firewall/router helps protect you from outside attacks. Consider this: how long does it take an unpatched Windows XP RTM machine to become infected if it is directly connected to the internet? Not long. (I've known a couple people that found that out the hard way. On the other hand, it's always good for a laugh when people try it.)
Also consider: Web servers often host multiple websites under the same IP. It is not always necessary for each and every public server to have it's own IP.
Ok... True, we are running out of addresses, and we do need more. But the mere availability of the addresses does not mean that all of a sudden we'll have more websites and devices on the net.
There is also the aspect that a good number of ISP's block residential users from running web and email servers. The reason is largely financial: bandwidth costs money, and if you plan on running a server they want you to pay extra. The shift to IPv6 will not change that fact. There is another reason: home users rarely have the knownledge to securely run a web-/mail-server, and the last thing your ISP wants is a large block of its users infecting the rest of its userbase with nasties.
Further: Although it will certainly be cheaper to purchase a large block of IP addresses in IPv6 (once it becomes mainstream), it isn't really that expensive to own several IPv4 addresses right now.
Example: My job is as sys-admin for a small company. We "own" a block of 16 public addresses (and really only use 8). The cost is marginal compared against the cost of the line, itself. The price of IPv4 addresses is just high enough that you don't go purchasing a bunch of them unless you'll actually use them, but still low enough that if you have good reason to need them you won't balk at the price.
At the end of the day, IPv6 will be a nice thing... but it isn't really going to foster some kind of radical changes. The savings it will (eventually) bring will effect the bottom-line on very few companies (webhosts are about the only ones that I can think of). On the other hand, the cost of buying new equipment, implementing it and testing it can be quite high. Personally, I estimate that it will take a good long while before you'll see any heavy use of IPv6. There just isn't enough pay-off for the cost involved.