That's it, they've all gone.
Iana to make final IPv4 allocations | ZDNet UK
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.
If at first you don't succeed, try writing your phone number on the exam paper.
It's hardly up to the user to ensure that their ISP is IPv6 enabled. If it wasn't, then they would surely be facing very significant legal troubles.
Sent from my iPadŽ
good thing my entire portfolio is in IPv4 addresses, I'm gonna get rich selling these things off once the next IPhone comes out
╔╗╔══╦╗
║║║╔╗║║
║╚╣╚╝║╚╗
╚═╩══╩═╝
Oops?
Anyways, odds are we we'll be trading IPv4 addresses back and forth for a long while before there's no other alternative than being force-fed the IPv6 pratfall. So yeah, keep those IPv4 if you own any.
... Or give 'em to me. I'm taking donations.
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
Are you ready?
Test your IPv6.
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.
If at first you don't succeed, try writing your phone number on the exam paper.
Fail
Failed here too, naturally. But I'm assuming this is just my ISP not having turned IPv6 only.
>> That page is from 2003. Many major .COMs have prepared themselves for this switch and from what I understand, Google and others have started testing.
I think the problem here is a lot more complex than simply having .COMs moved to IPv6 and your ISPs enabling it. I suggest you read slowly.
EDIT: I was trying to remember another more recent paper on the subject and only now I could find a reference to it in my notes. No wonder I was having trouble remembering about it; I don't have access anymore to the Burton Group research papers. If anyone in here does, this it it: IPv4 Address Exhaustion: An Inconvenient Truth. It's a research paper from late 2008 by Jeff Young, a senior analyst over there, in which he also predicts the extinction of IPv4 address by November 2011.
Basically it extends on Bernstein's, ermm lively, discussion that I linked above by addressing the issues of lack of compatibility and interoperability between IPv4 and IPv6 and discusses what's a likely outcome of IPv6 implementation for the near future, before the new protocol establishes itself firmly and IPv4 is all but entirely gone; i.e. an internet fragmented into islands of connectivity and an increase of security concerns.
Last edited by Mario F.; 02-03-2011 at 09:12 AM.
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
My problem is probably my router before the ISP. Looking at all the settings, it requests an IP from the ISP but only has them in IP4 format and all other IP settings are in IP4.
Might need to shell out on a new router to get it to work.
I'm assuming that once that was done and assuming the ISP was setup for IP6 I can still operate my subnet on IP4? Might get difficult with DNS and other stuff though....
Even 2008 isn't recent enough. The solutions for things like this start popping up in the 11th hour. It's very easy to point out problems in a structure that nobody had really bothered to start fixing, yet.
I'm going to make a not-so-bold prediction here... no major city in any major country is going to experience any problems what-so-ever when we make the switch. As for web-space owners... as long as you're with a reputable provider, then your website will be just fine. No fuss, no muss. The only real problem I might have is buying a new router because I have a cheap-o off brand router that almost certainly won't support IPv6 and definitely won't be providing any firmware updates.
I'm slight more concerned about the world ending on Dec 21, 2012.
Last edited by SlyMaelstrom; 02-03-2011 at 10:46 AM.
Sent from my iPadŽ
I'm not suggesting doomsday. Merely stating that without client and server changes (hardware and code) IPv6 will be a problem. An IPv4 application cannot talk to a IPv6 server and vice-versa. There's no interoperability between the two.
Because still today we are obviously not caring one bit about implementing IPv6 connectivity, these problems are merely being delayed. If however the push for IPv6 had been done earlier, altering software to deal with IPv6 would be today a common scenario. But it's not. Legacy applications will be hit the harder with everyone rushing to have them altered. But even modern software -- client and server -- will need to be changed.
There's no sense of urgency. Not because this is a problem, but because not very many realize this is a problem, or simply don't care about it until it bites them in their behind.
Last edited by Mario F.; 02-03-2011 at 11:23 AM.
Originally Posted by brewbuck:
Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.
See the thing I never understood is why they didn't just add another dot...
Old IP ... 64.122.43.244
New IP ... 0.64.122.43.244
This could be a gentle transition. Severs and backbone providers could implement it right away and so long as the upper character stays 0 for a while, the upgrades could be fed to various end user systems through forced updates and nobody would even notice. Then when current dotted quads are all used up simply start using the upper character in the "dotted cinq".
This entire ipv6 business strikes me as painfully illconsidered, perhaps even deliberately disruptive and quite frankly I have to wonder if it would have gotten even a sideways glance if Microsoft wasn't backing it... It really does impress me as the bully trying to re-arrange the club house.
Because IP addresses just don't work that way. The dot notation is just a text representation of what IPv4 addresses really are: four bytes of data, allocated to two integers according to the subnet mask.why they didn't just add another dot...
By "adding another dot" you extend this to five bytes, which already breaks everything. So if you're breaking everything, why not extend it to the point where you have enough addresses to give one to every wristwatch?
That said, IPv6 does more than just extending the size of IPs, but it's been too long since I've really looked at the topic.
All the buzzt!
CornedBee
"There is not now, nor has there ever been, nor will there ever be, any programming language in which it is the least bit difficult to write bad code."
- Flon's Law