ARIN Letter Says Two More Years of IPv4

dew4au writes "A reader over at SANS Internet Storm Center pointed out a certified letter his organization received from ARIN. The letter notes that all IPv4 space will be depleted within two years and outlines new requirements for address applications. New submissions will require an attestation of accuracy from an organizational officer. It also advises organizations to start addressing publicly accessible assets with IPv6. Is ARIN hoping to scare companies into action with the specter of scarce resources? This may be what's needed to spur adoption since there appears to be no business case for IPv6 deployment."

Comments

What about my toaster?

[MrEricSir]

When IPv6 was announced, one of the benefits was that everything could have its own IP address; even your toaster!

So as for a business case, what about the internet toaster business? If we don't switch to IPv6, what will they do?

Re:What about my toaster?

[Anonymous Coward]

Now to wait for the day that people hack into toasters because they're bored or mad at their neighbors.

Re:What about my toaster?

[Anonymous Coward]

I for one won't be satisfied until each of the pixels on my monitor has its own IP address.

Re:What about my toaster?

[Burkin]

Why would you be running Windows on your toaster?

Re:What about my toaster?

[Anonymous Coward]

He likes his toast burnt to a crisp?

Re:What about my toaster?

[omnichad]

IPv6 has 3x10^38 addresses.

Assuming that everyone in the world owns a 1080p monitor, that's about 1x10^16 pixels.

There would be enough IP addresses for each pixel, and still have more than enough IP addresses left to give every man, woman, and child's toaster an IP and also to replace IPv4 in its entirety.

Re:What about my toaster?

[shadow349]

So as for a business case, what about the internet toaster business? If we don't switch to IPv6, what will they do?

They can receive bailout funds from the stimulus bill under the guise of a "smart power grid" appliance.

You think I'm kidding, don't you?

Re:What about my toaster?

[Anonymous Coward]

Because chances are the toaster companies won't be smart enough to choose Linux for their Toaster Operating Systems, meaning that most toasters will unfortunately come with Windows Toaster Edition rather than Ubuntu Toaster Remix. ;)

Re:What about my toaster?

[aynoknman]

There would be enough IP addresses for each pixel, and still have more than enough IP addresses left to give every man, woman, and child's toaster an IP and also to replace IPv4 in its entirety.

Yeah, but do the toasters get their own pixels with IP addresses?

Re:What about my toaster?

[jd]

That's assuming packed addressing. IPv6 is hierarchical, which means that it's largely sparse addressing, so your theory doesn't hold up. However, since each home network has 48 bits of address space, you still have enough addresses for your monitor - you just won't be able to use the mobility option.

Re:What about my toaster?

[angelwolf71885]

would anyone like some toast ( TALKY TOASTER RED DWARF )

Re:What about my toaster?

[omnichad]

in HD!

Re:What about my toaster?

[timeOday]

Bailout and stimulus are different things, but I am all in favor of networked appliances to cut down on energy costs. I would start with heat, AC, and fridge rather than toaster though. I would like to get an email from my appliances each month saying how much power each one is using, since a broken appliance can end up running around the clock and wasting power.

At some point I would also like to get a plug-in hybrid that can talk to the power grid and charge up when energy is cheapest, which may be somewhat unpredictable as solar and wind power come online.

Re:What about my toaster?

[palindrome]

You know you're on to a loser when you have to state your pop culture reference in capitals.

Re:What about my toaster?

[Bandman]

HDoverIP

Remotely address the individual pixels of a monitor.

Talk about a thin client...

Re:What about my toaster?

[Joe The Dragon]

ISP like comcast will love to make you pay $5/m per system on top of $30 - $50 /m fee.

Re:What about my toaster?

[Anonymous Coward]

You use NAT. Same as you do for your pc and your gaming console and your mobile devices and etc. :)

Re:What about my toaster?

[Anonymous Coward]

Bailout and stimulus are different things

Perhaps you could tell the Administration that.

President Obama was introduced at his first speech after the stimulus package passed by the president of a solar power company.
His company is getting "stimulus" money from the package to stay afloat.

In my world, that's called a "bailout".

Re:What about my toaster?

[Anonymous Coward]

You can already get all that with the right kind of power meters you install at the outlets in the house.

But the important question is are you ok with sending all that data to the government run power company?

Do you want them knowing what appliances you have, when you use them, etc? And do you really think they will be able to resist selling that information to marketing groups, or giving it up to the authorities?

Now combine the ID with RFID tags in product packaging...

Just imagine the email you might get:
"Good day, Comrade!

We have noticed that you have switched from eating one bagel and two pieces of toast, cereal and milk, to simply drinking a cup of coffee each morning, and more alarming to us is the fact that there is no juice in your refrigerator. Your lack of caloric intake is probably why you have been running the heat more than normal, which is wasteful and we would like to remind you that waste is a violation of Health Ordinance 639462.
In addition, we've noticed that you haven't been leaving the house as often, as there have been less sudden drops in the ambient temperature from the doors opening, and we are concerned you may be suffering mental fatigue.

In the interests of your health & the well-being of the community, we have notified your local Health & Wellness officer, who will be at your house tomorrow to collect a mandatory urine sample. Anyone interfering with the collection of urine samples will face criminal charges.

Be well, Citizen!
- The Citizens Glorious Health Bureau"

Re:What about my toaster?

[jgtg32a]

well what ever OS base they decide to use for their TOS hopefully they have a good TOS.

Re:What about my toaster?

[adolf]

In the US, which is plainly the context here, power generation and distribution are almost entirely done by private (non-government) entities.

This small detail un-funnies the rest of your comment.

Re:What about my toaster?

[Miamicanes]

> However, since each home network has 48 bits of address space (snip)

The last time I checked (about 6 weeks ago), ISPs are supposed to assign a 48-bit address to each "customer" (read: site, household, office, etc), who'll have 80 bits, not 48, under his direct control -- from a block whose upper 32 bits are assigned to the ISP by the local coordinator (ARIN, RIPE, etc). In English, here's a theoretical IP address represented by placeholder letters (each letter represents 1 hexadecimal digit = 4 bits):

aaaa:aaaa:bbbb:cccc:dddd:dddd:dddd:dddd

where

aaaa:aaaa is a prefix assigned by ARIN/RIPE/etc to the ISP. For now, most of the addresses we see will have "2001" as the first 4 digits.

bbbb is a 16-bit value, representing 65,536 potential customers. This is the part the ISP gets to assign to customers.

cccc is another 16-bit value. This is the part you, the customer, are officially supposed to be able to use however you please

dddd:dddd:dddd:dddd is a 64-bit value. In theory, this value is supposed to be determined by your ethernet card's MAC address. Originally, it was "mandated". Due to privacy concerns (your ethernet card would be trackable out-of-band wherever in the world you used it from and would have effectively been the "tracking cookie from hell"), it was first softened to allow some randomization, and eventually made a "recommendation". More on this in a moment...

So... what does this mean for you, Joe DslCableModelCustomer? In theory, you will someday be getting a letter from them to the effect of, "Your new IPv6 prefix is 2001:3f87:991d:/48". What does this mean? In the real world, it means you'll plug the shiny new Linksys router you bought circa mid-2012 into it, and configure its address to be 2001:3f87:991d::1 You'll then verify that the rest of your network (192.168.x.x IPv4 addresses and all) is happily doing NAT, and forget about it.

To the rest of the world, your desktop PC (192.168.0.128) will either appear to be 2001:3f87:991d::1 (if the router is acting as an IPv4 proxy), or if you're extra-clever, will transparently be rewritten to something like 2001:3f87:991d:0::192.168.0.101 or 2001:3f87:991d:0::c0a8:0065. Ditto, for the other half-dozen computers and devices in your home that are connected to the internet.

A few weeks later, you get into an IPv6 fetish, and decide to abolish the IPv4 legacy and make everything pure IPv6. At this point, your public IP addresses look even prettier:

your firewall's new IPv6 address is set to 2001:3f87:991d::100
your desktop PC's new IPv6 address is now 2001:3f87:991d::101
your TiVO's new IPv6 address is 2001:3f87:991d::102
and so on.

Put another way, nobody is going to put a gun to your head and force you to use the lower 64-80 bits if you really don't want to. If you're a typical home user who just wants to plug things in and have them work, they'll autoconfig using the munged MAC address and publicly assume some horrific, ugly value its owner will probably never type directly anyway. If you want your network to be handcrafted, with addresses you can remember, you're perfectly free to collapse the 80 bits you control down to as few as 1 bit if that's what makes you happy. Maybe even ZERO bits (I'm not 100% sure whether 2001:3f87:991d:0:0:0:0:0:0 is a legitimate address, or whether the ::0 address still refers to the (sub)net as a whole).

As for privacy, I fully expect that most ISPs will eventually have a semi-anonymizing web proxy available for their customers to use. They'll keep logs for a few days to fight spammers, botnets, and criminals, but keep things sufficiently shuffled around to keep marketers from ever getting TOO comfy and intimate with your IP address. It'll make ISPs happy, because they can make it cache traffic and squeeze more use out of their upstream bandwidth.

Note that the allocation scheme I just mentioned IS radically different from what IETF envisioned circa 2000. Sometime in the past 2 or 3 years, they put down the crack

Re:What about my toaster?

[VGPowerlord]

I thought IPv6 split the network and local address segments right down the middle (i.e. each is 64-bit).

Geez, no wonder adoption of IPv6 is having so much trouble... they don't even include important information such as the size of the network identifier in the IPv6 Addressing RFC (RFC 3513: Internet Protocol Version 6 (IPv6) Addressing Architecture, obsoletes RFC 2373)

Re:What about my toaster?

[fractoid]

Thin client? I use a CRT, you insensitive clod! :P

Re:What about my toaster?

[timeOday]

I wasn't talking about emailing anything to the government anyways. I was talking about my appliances emailing *me* if they think something might be wrong, neither the govt. or power company has anything to do with it. Also if I left on a trip without e.g. turning my hot water heater off (which I currently never bother to do), I could email my home from my phone to go into "power saving" mode.

Re:What about my toaster?

[Anonymous Coward]

I like this box-size example...

Re:What about my toaster?

[hardwarefreak]

Only one of the survey respondents is closely tracking its IPv4 address usage, saying that it needs 130,000 new IPv4 addresses every three years.

Only spammers go through IP space at this rate, which brings up a very good point in the v6 vs v4 debate. That point is that there are dozens of "reserved" /8 IPv4 networks that have never been assigned/used and there are tons of "scorched earth" networks and netblocks that no one wants because spammers have caused them to be universally blacklisted in DNSBLs or firewalled by many entities (ISPs, Corps, EDUs, etc).

Many have known for a very long time that we don't need more address space. We need to make more judicious use of what we already have. But, then again, Internet Protocol and the net itself are American inventions. And it's the American way to throw out something perfectly good for something marginally better, simply because we're tired of having to properly maintain what we already have. Sad sad sad...

Then there's always the money trail. ARIN will bring in more revenue if they have more address space to 'lease'. Greed is a far more significant aspect of IPv6 deployment than most people realize. Yes, gasp!, there is greed in ARIN. They are not, nor have they ever been, a neutral organization.

Re:What about my toaster?

[hairyfeet]

Not to mention there are probably millions of routers that don't support IPv6 and can't be upgraded. That is going to be some serious metric tonnage heading for the landfills unless all the ISPs build all the new modems with IPv6 to IPv4 NAT built in. And then there are all those modems that will have to be chunked too. And the amount of routers out there is simply crazy. Everybody uses NAT nowadays so if they would actually take back all those "wasted" IPv4 network addresses we would probably get another decade. That would let us slowly replace the routers as they break down/become obsolete (assuming they stop making IPv4 only routers today) instead of having to chunk a large amount of functional routers now in mass.

Thanks to NAT I just don't see the need to switch ATM, and if ARIN would actually reclaim all those wasted addresses that are sitting idle we could probably get some decent mileage out of IPv4 and allow everyone to replace their gear as it fails with IPv6 ready gear. But just in my home town from what I've seen most of the routers folks have simply won't be able to handle IPv6. That is going to make for a serious load on the landfills if the switch is done quickly. Perhaps like TV we should simply say "no more IPv4 only routers allowed" to allow the public to slowly replace the gear with capable equipment so when the addresses is gone everyone's gear will be ready to go.

Re:What about my toaster?

[Anonymous Coward]

As for privacy, I fully expect that most ISPs will eventually have a semi-anonymizing web proxy available for their customers to use. They'll keep logs for a few days to fight spammers, botnets, and criminals, but keep things sufficiently shuffled around to keep marketers from ever getting TOO comfy and intimate with your IP address.

BWAHAHAHAHAHAHAHA.

The only reason ISPs may want to hide that information from marketers is to sell it to them later.

Re:What about my toaster?

[Miamicanes]

> I thought IPv6 split the network and local address segments right down the middle (i.e. each is 64-bit).

From what I remember, that was more or less the plan circa 2002-2004. The main problem with the original address allocation scheme was that it left big gaps in places that made it nice to route, but a bitch to memorize and rendered the proposed shortcut notation all but useless. Originally, they planned to use the upper 3 bits as a grand macro-level version indicator, then leave the next byte zero for now, then hop and skip over the next few bytes using the lower bit or two of each byte until they got to the "meat" of the address somewhere around bytes 5-8. That would have resulted in lovely addresses like 100:103:401:3f7a:xxxx:xxxx:xxxx:xxxx which, if you actually used your MAC address to set the lower 48 bits, would have been all but impossible to meaningfully encode with the "::" zero-packing shortcut. At best, you might have ended up with 2 pairs of sequential zero bytes to compress, and had to pick one or the other.

Re:What about my toaster?

[pmarini]

as in POKE 192.168.11.12@1900:1060,#00FF00 to flip the bottom-right corner to green?

Re:What about my toaster?

[Anonymous Coward]

In the US, which is plainly the context here, power generation and distribution are almost entirely done by private (non-government) entities.

In that case, replace "Citizen" with "Consumer" and "criminal charges" with "violation of the EULA".

Re:What about my toaster?

[Omnifarious]

IPv6 address space has been very carefully partitioned out. The address space currently used for public addresses is only 1/8th of the total address space (still utterly mind-boggingly gigantic). There is currently also at least 1/4 of the total address space 'reserved for future use'.

So, yes, right now IPv6 addresses are sparse and might end up being used up fairly quickly, though I bet with the current addressing scheme that every pixel on every monitor could get its own IPv6 address and things would still work out fine.

But, even if the current addressing scheme's sparseness proves its undoing, there's another whole 1/4 of the space to try again with a different scheme. Adjusting routers to accept something other than 2000::/3 as publicly routable is not nearly as big a change as getting them to do IPv6 in the first place. I think we're good for the foreseeable future.

I do think that if we expand beyond the solar system and end up with an FTL communication scheme that IPv6 will prove to be too small.

Re:What about my toaster?

[Tony Hoyle]

Nobody uses /48, they use /64. RA doesn't work unless the prefix is /64, so ISPs and everyone else will end up using that. Also nobody manually sets ipv6 addresses.. one of the major features of ipv6 is that you don't do that.

Re:What about my toaster?

[Tony Hoyle]

What do you mean 'originally'. That's exactly what they did do.

The only change recently has been to deprecate site local addressing. They've also improved the RA specs to include sending other data like DNS (which it should have done from the start.. someone wasn't thinking straight when they missed that one).

Re:What about my toaster?

[Miamicanes]

As of early 2008, that was true. Things changed last year. IPv6's original supporters were furious, but the pragmatists won. The original stance was that NAT6 and manual configuration isn't just unnecessary, it's actively harmful. I suspect its proponents still believe that, but as of now they've decided to just throw in the towel and let people have NAT6 and human-manageable IPv6 addresses if that's what makes us happy. In ARIN-land, at least, requesting enough prefixes to hand out a /48 to any customer that wants one (if not everyone, period) is automatically approved and doesn't need to be justified.

To wit: for the next decade or so, at least, if you really want to make IPv6 look like IPv4 with slightly longer addresses, you can. The official stance is that you shouldn't, but nobody's going to forcibly stop you if you try. The original stance was that they were going to stamp out any and all IPv4 heresy, even if it killed them in the process (by motivating ISPs and others to say "No" and just keep extending IPv4 via NAT).

Re:What about my toaster?

[jonbryce]

You don't need IPv6 to do that. If you are controlling it by email, you don't even need a publically routeable IP address.

Re:What about my toaster?

[Just Some Guy]

Nobody uses /48, they use /64.

We do. We have a /48 from HE.net, and using the GP's example of aaaa:aaaa:bbbb:cccc:dddd:dddd:dddd:dddd, we use cccc to number netblocks on the router. For instance, aaaa:aaaa:1:: is the main LAN, aaaa:aaaa:2:: is the server block, aaaa:aaaa:3:: is Wi-Fi, etc. That way we still have 64 bits left for autoconfig, and the router doesn't have to memorize routes to every single host on our network.

Re:What about my toaster?

[RichiH]

> As for privacy, I fully expect that most ISPs will eventually have a semi-anonymizing web proxy available for their
> customers to use. They'll keep logs for a few days to fight spammers, botnets, and criminals, but keep things
> sufficiently shuffled around to keep marketers from ever getting TOO comfy and intimate with your IP address. It'll
> make ISPs happy, because they can make it cache traffic and squeeze more use out of their upstream bandwidth.

Bandwith is cheap
Routing is fast
Adding a new network egress point is easy

Proxying on a multi-homed network is none of that. I talked to the one in charge of doing this for the DFN, the German university & research backbone. The summary is "thank $deity we got rid of it".

Re:What about my toaster?

[Cajal]

The main problem with the original address allocation scheme was that it left big gaps in places that made it nice to route, but a bitch to memorize

Why do you need to memorize addresses? Use DNS (either static or link-local multicast). It's a lot easier.

Re:What about my toaster?

[GeorgeS]

Is there any way to re-flash some of these "home routers"?
In my limited experience I've seen a few that could just be re-flashed.
What about the big ones like Cisco's?
The idea of several million or worse, a billion routers getting tossed in the trash really bothers me so I hope there is some way to recycle or reuse them.

Nothing gets fixed until it breaks

[madbavarian]

Nothing gets fixed until it breaks so fully that people can't ignore it any longer. ARIN should just hand out the last of their IP assignment already and then we can move on with actually deploying IPv6.

Re:Nothing gets fixed until it breaks

[SnarfQuest]

Just do a HDTV conversion. Give a specific date when IPV4 support will be dropped, then extend the date when the timeout gets close.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

A nit, I know, but DTV conversion had little to do with HD...

That's partially why so many idiots were confused.

Re:Nothing gets fixed until it breaks

[vivin]

There are a number of corporations and organizations that own /8's

Here is a list

Here's a few from the list:

003/8 General Electric Company
004/8 Level 3 Communications, Inc.
008/8 Level 3 Communications, Inc.
012/8 AT&T Bell Laboratories
013/8 Xerox Corporation
015/8 Hewlett-Packard Company
016/8 Digital Equipment Corporation
017/8 Apple Computer Inc.
019/8 Ford Motor Company
034/8 Halliburton Company

Seriously... why does Ford Motor company need a /8?

The US government also owns a whole bunch of /8's

Instead of hogging these, they should just give them up. They don't need all these addresses.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

Seriously... why does Ford Motor company need a /8?

They've been keeping it in reserve for a rainy day.

Do you know how much a /8 is worth in today's market? It could pull Ford out of its financial problems!

Re:Nothing gets fixed until it breaks

[dew4au]

Level 3 Probably uses a good bit of those /8s. I mean, they do have one of the largest backbone nets, and they use allocate their IPs to content providers on their networks.

Re:Nothing gets fixed until it breaks

[morcego]

IBM used to use 9.0.0.0/8 address for their internal network. Computers that didn't have access to the internet or anything.

This was back in 1995, so I can't guarantee it is still true, but it is likely.

Re:Nothing gets fixed until it breaks

[aynoknman]

Seriously... why does Ford Motor company need a /8?

They need it so they can sell it when they go into Chapter 11

Re:Nothing gets fixed until it breaks

[jd]

Nah. Ford already uses most of their /8 in assigning each nut and bolt in each of their cars its own IPv4 address.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

TRUE!! Japan(the people who developed HD over 20 years ago) still doesn't do digital.

Re:Nothing gets fixed until it breaks

[Blue6]

A lot of the companies that have class A networks had them issued before CIDR. Also don't underestimate the size of some of these networks. Ford has a half dozen datacenters spread out around the world thousands of VOIP phones, Desktops / Laptops, routers, switches, AP, servers. Not to mention most modern manufacturing plants PLC's run on a IP network sure you will never use the whole space but do you really think they are going to re-IP a network that size. Ford also owns a class B network :)

Re:Nothing gets fixed until it breaks

[Bandman]

And there's absolutely no reason that those devices can't be assigned an address from the 10.x portion of RFC 1918. None at all, except for the magnitude of the problem.

They should have planned for that so, so long ago.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

..do you really think they are going to re-IP a network that size.

If given proper notice that they will be losing the class A license, then I'm sure they would. There is almost no justification for a corporation to have public IP addresses for VOIP phones, Desktops, Laptops, and many network components (switches, routers, etc) which strictly reside on their internal network.

Re:Nothing gets fixed until it breaks

[timeOday]

It's like land - what was initially given away to ranchers and farmers in vast swaths will now be sold back to us in ever-smaller and more expensive blocks.

Re:Nothing gets fixed until it breaks

[Blue6]

execpt they are paying the fees for the range so they can use it as they see fit. Why should these companies be punished for ARIN's lack of planning.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

Go ahead, yank 'em all back. Worldwide, the five RIRs (AfriNIC, ARIN, APNIC, LACNIC, RIPE) go through 12-14 /8s per year. Don't give yourself a charley-horse patting yourself on the back because you managed to move out the exhaustion date by 8 months.

BTW, the US Government *gave back* several /8s.

IPv4 is terminal. Get over it and get your IPv6 on.

Re:Nothing gets fixed until it breaks

[tchuladdiass]

What about when companies merge, or otherwise have to connect networks? Two companies using 10.x could have overlapping IPs.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

There are a number of corporations and organizations that own /8's

Here is a list

Here's a few from the list:

016/8 Digital Equipment Corporation

Must be an old list. DEC was purchased by Compaq a decade ago. Compaq has since been purchased by HP.

Re:Nothing gets fixed until it breaks

[DaleGlass]

Instead of hogging these, they should just give them up. They don't need all these addresses.

"Should"? Why exactly should they?

I thought America was the land of Capitalism, and as good capitalists of course they will keep them until the address space runs out, then make big bucks reselling the unused chunks.

Re:Nothing gets fixed until it breaks

[idiotnot]

THIS. Mod A/C parent up.

Reclaiming class As only delays things slightly, and doesn't fix the inescapable math.

But it's much easier to bitch and point fingers at evil corporations like Ford, than it is to pick up a damn book and learn how IPv6 works.

Re:Nothing gets fixed until it breaks

[Just Some Guy]

Here's a few from the list:

[snip 10 /8s]

OK, that takes care of 4% of the IPv4 address space. What's your next great idea?

Re:Nothing gets fixed until it breaks

[supremebob]

They still used the 9. network when I worked there in 2008, so I doubt that anything has changed.

Re:Nothing gets fixed until it breaks

[iris-n]

+5 Insightful again?

Can't we just let this die? There are plenty of unused IPv4 addresses, sure. Most are hard to get due to political problems. No company will re-IP all their network just out of goodwill. So what?

The sooner IPv4 addresses end the better. Any quantity that is salvaged is just delaying the inevitable, and hurting IPv6. We could be in a much better infrastructure today if it wasn't for all this whining and "business case"ing.

So what that these companies can make a buck selling the addresses? Anyone dumb enough to pay instead of upgrading does not deserve the money.

Re:Nothing gets fixed until it breaks

[sl3xd]

Or, given the number of them that are tech companies, it could be that they are "hogging" them to increase pressure to move to IPv6... because they believe moving to IPv6 is the "right thing" to do.

Sure, it'll piss off the MCSE's who still struggle with IPv4 and think that NAT is a firewall, but the rest of the world won't really care. Vista, OS X, Linux, and *BSD all support IPv6 out of the box now, as do most browsers. IPv6's stateless autoconfig is a beautiful thing.

All the ISP's have to do is start routing IPv6 to their customers. The Tier-1 and most tier-2 providers already use and route IPv6.

Re:Nothing gets fixed until it breaks

[Kaboom13]

What about when companies merge, or otherwise have to connect networks? Two companies using 10.x could have overlapping IPs.

Holy shit, so the network admins will have to do their jobs? It's not like just patching two completely different networks together makes a whole lot of sense anyways.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

There was NEVER an HDTV conversion. There was a digital TV conversion. Digital TV does not equal HDTV. Digital TV was forced on people so the government could sell off the analog TV bandwidth.

Re:Nothing gets fixed until it breaks

[jgtg32a]

Most of them don't just support it most look for an IP6 address and then fail to IP4

Re:Nothing gets fixed until it breaks

[Tubal-Cain]

IBM used to use 9.0.0.0/8 address for their internal network. Computers that didn't have access to the internet or anything.

If they're not connected to the Internet, then what does it matter?

And why 9.0.0.0/8?

Re:Nothing gets fixed until it breaks

[jibjibjib]

Unlike the TV situation, there's no central authority which can stop people from using IPv4.

Re:Nothing gets fixed until it breaks

[Bashae]

Because they own all of those...

Even if you are not connected to the internet, you should not use IP addresses that are not reserved for local networks. However, since they own the space, they can do whatever they want with it.

It's still a huge waste, though. A single /8 could provide enough addresses for the entire population of my country to have a computer connected to the internet... And there would still be millions left over.

Re:Nothing gets fixed until it breaks

[drinkypoo]

Lots of those machines have access to the internet, but not vice versa. And lots more are accessible from other hosts on the 9 net which fall into that first category.

Re:Nothing gets fixed until it breaks

[adolf]

I know, right?

Just like with telephone extensions: Imagine the fallout if the five-digit telephone extensions overlapped! Couple that with overlapping assembly lines, and overlapping administration, and what you'd get would be mass hysteria.

Or, rather, it wouldn't be. It's just another problem that needs addressed (pardon the pun) with a change as large as a merger.

Re:Nothing gets fixed until it breaks

[adolf]

We used to sell some H.323 IP stuff from Altigen, early in the VOIP game. These phones, when we first started with them, we an odd breed: They'd be happy enough to deal with the latency and packet loss associated with a public network, but were totally unable to deal with NAT. At all. Not even a DMZ was helpful until different firmware came out years later which finally addressed a few of these problems.

So: Perhaps Ford simply has 16.7 million Altigen IP phones. :)

(Yes, yes. A VPN would work, but my experience with VPN+H.323 hss been pretty dismal as well.)

Re:Nothing gets fixed until it breaks

[VGPowerlord]

Go ahead, yank 'em all back. Worldwide, the five RIRs (AfriNIC, ARIN, APNIC, LACNIC, RIPE) go through 12-14 /8s per year. Don't give yourself a charley-horse patting yourself on the back because you managed to move out the exhaustion date by 8 months.

[citation needed]

Actually, the citation should point to IANA's master list, but that disproves your statement, not supports it.

Re:Nothing gets fixed until it breaks

[QuantumRiff]

Problem is, nothing will break, just nobody new can get address blocks. NOthing will change, because you and I won't notice people we don't know about, not coming online..

Re:Nothing gets fixed until it breaks

[ekhben]

http://www.iana.org/assignments/ipv4-address-space/ipv4-address-space.xml is a bit easier to parse mechanically. There were 9 /8 allocations to RIRs in 2008. Not 12-14. 2007 was 13 allocations. 2006: 10. 2005: 11. 2004: 9. 2003: 5. 2002: 4. 2001: 7. So far in 2009: 4.

So, yeah, 12-14 is hyperbole. Not by so much as to fundamentally change the point, which is, it would take more effort to convince those holders of a precious and dwindling resource that they should just give it up out of the goodness of their steel-and-concrete hearts, than you could justify by the amount of time you'd gain. Spend that time and money on IPv6 promotion and education, instead.

Re:Nothing gets fixed until it breaks

[morcego]

If they don't accept incoming connections from the Internet, there is no reason for them to use a 9/8 address. They should just go for a 10/8, 192.168/16 or whatever.

Re:Nothing gets fixed until it breaks

[Nevyn]

arstechnica 2008 year end report in ipv4 implies that 4% is actually a pretty big chunk ... not that I think GM/MIT/etc. are stupid enough to just give the IPs back.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

If they don't accept incoming connections from the Internet

You mean if none of those machines will ever accept even one connection. Because that's how brain-damaged NAT is.

And in return for breaking every interesting IPv4 application, you're postponing the inevitable for, what, a few months?

Re:Nothing gets fixed until it breaks

[RCSInfo]

Don't forget that HP bought Compaq, who previously bought Digital Equipment. You can put both the 015/8 and the 016/8 from your list under HP.

Re:Nothing gets fixed until it breaks

[dodobh]

There is not justification for violating end-to-end connectivity. It's a network of peers.

Re:Nothing gets fixed until it breaks

[TheTurtlesMoves]

Another problem is that people think NATs are a feature rather than symptom of the inadequacy of ip4. Put something like IPsec in there with the NAT traversing and you really start to see how much easer life would be if we were on ip6 now.

Re:Nothing gets fixed until it breaks

[Omnifarious]

Except for, you know, that the whole 10.x.x.x thing is a stupid, brain-damaged way to go about doing things in the first place. In fact, if you read the RFC you reference it as much as says that. The only real justification it gives is the preservation of the global address space, since it's clearly way too small. Just fix the real problem instead of applying a contorted broken patch to solve it.

Re:Nothing gets fixed until it breaks

[petermgreen]

What will almost certainly change is that home users will be forced to go behind ISP level NAT (ISP level NAT is much worse than a NAT you control because you are unlikely to be able to port forward through it) or pay extra (perhaps a lot extra).

This will then free up address space for more lucrative customers.

Re:Nothing gets fixed until it breaks

[kasper37]

Get over it and get your IPv6 on.

I'm ready, but what isp will give me ipv6 connectivity now? My isp doesn't (and I thought that if any isp would offer ipv6 services, it would be Speakeasy).

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

Why should IP administrators punish Ford (or Xerox or HP ad nauseum)? Because the alternatives will be:

a) Stop assigning IP ranges. Punish 2^8 companies that could responsibly use /16 networks if Ford weren't hoarding a /8 network.

b) Stop assigning IP ranges. Punish 2^16 companies that could responsibly use /24 networks if Ford weren't hoarding a /8 network.

b) Transition to IPv6. Punish the millions of businesses and consumers with IPv4 hardware.

Before reading your comment, I felt some sympathy for the Ford IT team. Now that I'm looking at this as an emotional matter of "who to punish", this issue is a no brainer: punish Ford.

Re:Nothing gets fixed until it breaks

[slack_justyb]

Wow, hindsight is indeed 20/20.

Re:Nothing gets fixed until it breaks

[drinkypoo]

If they don't accept incoming connections from the Internet, there is no reason for them to use a 9/8 address.

( You == Idiot || Liar ) == True

First of all, there's the various applications that don't work through NAT. They are numerous, and to expect IBM to toss them over because we're out of IPs is fucking ridiculous. Then there's the issue that NAT is just an extra piece of work; it leads to headaches. Life is much simpler if you don't have to mess with multiple addresses for a single host, or have a system providing such a service. Finally, IBM is entirely justified in not giving up these addresses, because we have a solution, and that solution is IPv6. NAT has been a fucking terrible headache and created a ton of unnecessary work for programmers and IT professionals alike that prevented actual progress.

Re:Nothing gets fixed until it breaks

[Just Some Guy]

They should have planned for that so, so long ago.

They did, by acquiring a /8 so they never had to worry about it again.

Re:Nothing gets fixed until it breaks

[Rich2k]

Because my car definitely needs it's own IP address, just so I can browse the Internet using IE6, get infected with a virus and crash the car!

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

b) Transition to IPv6. Punish the millions of businesses and consumers with IPv4 hardware.

Ford followed the rules. It's not their fault you chose hardware from an incompetent vendor. Why should they cripple their network with NAT? You're going to have to upgrade anyway, you're just putting it off for a few weeks. It's not as if we have enough /8 blocks to keep going until you retire or die.

Re:Nothing gets fixed until it breaks

[RichiH]

Having migrated a /17 recently, I can only tell you to shove it. If they actually use the /8 (i.e. not fully, but with high fragmentation), there is no way in hell they could do this now.

In an ideal world, they would have started the migration five years ago at the command of IANA and the local registries. Which is not realistic as there is IPv6 and they would never have been forced to do so.

So while I agree with your gut feeling, it's not practical from the technical and adminstrative side.

Re:Nothing gets fixed until it breaks

[dilvish_the_damned]

What about when companies merge, or otherwise have to connect networks? Two companies using 10.x could have overlapping IPs.

Double nat with local DNS fixes manages the problem. Its how I isolate testbeds that require the same address ranges. I see no reason it could not be scaled to 10.x.

Re:Nothing gets fixed until it breaks

[Cajal]

Even if you reclaimed all of the legacy /8s, it wouldn't do you much good. The IANA allocates between 10 - 12 /8s per year. At best, reclaiming the legacy /8s would only get you another 18-24 months in the IANA pool.

Rather than spend efforts trying (in vain) to reclaim legacy space, why not deploy IPv6?

Re:Nothing gets fixed until it breaks

[GeorgeS]

You have to remember who owns Speakeasy now(BestBuy).
Sucks for me too because I'm a Speakeasy customer myself.
I did ask about it a few months ago and they didn't have any idea or
plans for IPv6 that they could tell me anyway(probably going to switch to Verizon's FiOS partly because of this).

Here's a list of IPv6 providers http://www.sixxs.net/wiki/IPv6_Enabled_Service_Providers/
Some of them are in the USA(8 I think) but, only a few offer IPv6 to end customers.

Re:Nothing gets fixed until it breaks

[Anonymous Coward]

theres no reason the corps need them anyway. this is not about ipv6 not being good or people not being able to learn ipv6. i'm sure an elitist asshole like you can say "o i know ipv6 i rule!" who gives a fuck? this is about disproportionate allocation of resources.

get over yourself douchebag.

It's to try to get some attention

[kevmeister]

I just got back from the ARIN meeting this week and the letters are, indeed, a "scare tactic". Network providers keep reporting that PHBs won't spend any money on IPv6 even though engineers are begging for it. Most corporate officers probably think IP is only Intellectual Property and this is an attempt to draw their attention to the fact that the network world as they know it is going to end soon and that the only way to avoid serious problems is to either stop growing or to start IPv6 deployment. PHBs sometimes get the idea when they realize that not spending some money will lead to big problems in a few years. Others figure that if it's over a year away, it really does not matter because it won't impact their bonus this year, so it may not work, but we can hope.

Re:It's to try to get some attention

[jd]

That makes sense, but really they would be better off just saying to customers "sorry, we have no bananas today" and telling them that they can get a great deal on IPv6 addresses instead.

(If the PHB's need to know more, IPv6 is like the GruntMaster 5,000 - including all necessary wormholing technology.)

Re:It's to try to get some attention

[Anonymous Coward]

kevmeister (979231)
...
I just got back from the ARIN meeting this week
...
the network world as they know it is going to end soon

Mitnick, is that you?

Re:It's to try to get some attention

[moon3]

IPv6 is flawed and nobody is eager to switch, at least for now.

If you have a bunch of equipment on IPv4, that would cost thousands if not millions to upgrade.

From the security standpoint IPv6 is a nightmare, no kidding. Now we have multiple NAT and no-default-gateways etc., we are happy about it, in fact to not have IP facing internet is an advantage. So you tell us to throw this away and opt to routable IPs ? Just ask any ISP what they think about their customers running direct IPs.

Re:It's to try to get some attention

[TheTurtlesMoves]

IPv6 is flawed and nobody is eager to switch, at least for now.

How so? Really.

However NAT is broken. Perhaps your happy about it because someone else does all the dirty work for you?

Re:It's to try to get some attention

[Tony Hoyle]

Just ask any ISP what they think about their customers running direct IPs.

I did.. and they said 'how many do you want?'. I got 16 but could have as many as I can justify.. and that's not even a business account.

I don't get where you're coming from. You're running a firewall right? Or are you trying to run a company behind some cheap ass linksys and think that NAT is giving you security.

I want IPv6 support, but ...

[Midnight Thunder]

I want IPv6 support, but there are lots of pieces still not in place. I am actually using Miredo (Teredo implementation) when I am on the move and Sixxs when I am at home. These are more stop-gap solutions and until the necessary entities start allowing to get on board properly.

My parents live in France and they are with Free.fr who offers IPv6 as a standard option. On the other hand I am living in Canada and not one of the service providers offer IPv6 in any shape or form. One questioned about it they blame their up-stream provider. Even if they are ready the only IPv6 ready router for the home is the Apple Airport Extreme, and even then there is a blocker issue for connecting to Sixxs.net (Apple's bug). Linksys, D-Link and Buffalo are still not ready with a public release and you are left trying to see if the version of DD-WRT you need for IPv6 supports your router. Chances are you will be looking at eBay for a router that has enough flash to support it.

Like the Swine Flu outbreak, I get the feeling that few entities are going to be rushing to do any work until there is media frenzied panic.

There is no killer application for IPv6, since its just infrastructure. On the other hand the lack of a NAT can make certain application solutions easier to implement, since you don't need to do any NAT busting or other fancy tricks. Of course since internal addresses are now all routable, you will certainly need to make sure that you have a real firewall on the gateway device.

Once you are on IPv6 you can start playing around with IPv6 torrent and http://ipv6.google.com/ , if you are curious.

Re:I want IPv6 support, but ...

[entrigant]

So then my linksys router with built in fully functioning out of the box 6to4 support is what... ? I didn't even know about it until I noticed I was connecting to ipv6 irc servers.

Re:I want IPv6 support, but ...

[madbavarian]

If you are truly using IPv6 couldn't you just plug a plain old Ethernet switch into your cable modem or dsl modem?

(Well, I guess that assumes one is running an OS that has a firewall just as capable as the ones in home "routers".)

Re:I want IPv6 support, but ...

[Midnight Thunder]

So then my linksys router with built in fully functioning out of the box 6to4 support is what... ? I didn't even know about it until I noticed I was connecting to ipv6 irc servers.

What model is that? Engineering probably didn't tell sales, since when I contacted them they said none of their routers support IPv6.

Re:I want IPv6 support, but ...

[gbjbaanb]

That 6to4 support is bundling IPv6 packets and transmitting them inside an IPv4 packet. So technically, the poster is still using IPv4 with his linksys router.

I think the home router issue is the one that matters. I want IPv6, but simply cannot have it (unless I cough up lots of cash for a serious router). I think the home router manufacturers are missing something here, they just need to say they cannot release firmware updates, and that you need to buy a new router to get IPv6, which is obviously better. They then sell loads more routers.. I don't understand why they don't do this.

Mind you, a firmware update would be better for me :)

Re:I want IPv6 support, but ...

[jd]

IPv4 is still gaining features that should have been in place to start off with. IPv6's biggest selling point is that it's designed to be retrofitted, whereas IPv4 is not.

Besides, if people were forced to use IPv6, how long do you REALLY think it'll take for the network companies to finish the protocol? A weekend at most, at this point. They're dragging their feet because R & D sees this as a cash cow they can milk forever if they never actually complete anything.

Re:I want IPv6 support, but ...

[jd]

A layer 2 or layer 3 switch, yes. A router or a switch/router would have problems unless IPv6 is supported OR you are tunneling over IPv4.

Re:I want IPv6 support, but ...

[Anonymous Coward]

Not sure about his but my friend has a RVS4000 and it seems to have it, LAN does anyway.

Re:I want IPv6 support, but ...

[Bandman]

I think your layer 3 switch would have problems unless it supported ipv6, too.

Re:I want IPv6 support, but ...

[Anonymous Coward]

On the other hand I am living in Canada and not one of the service providers offer IPv6 in any shape or form.

If you're in eastern Canada, you might want to look into teksavvy. While they don't directly offer IPv6, they are looking into it. In the mean time, there's a 6to4 tunnel broker is in the same datacenter as where they aggregate their DSL lines. So, very little extra overhead.

Re:I want IPv6 support, but ...

[mellon]

There is no killer application for IPv6, since its just infrastructure.

Not true, and you mentioned the killer app in the very next sentence: end-to-end connectivity. Having real, working end-to-end connectivity is a big deal, but most people don't know it because they're accustomed to living on a network where there is no end-to-end connectivity.

So if you want to see more IPv6 deployment, start developing apps on top of Miredo/Teredo that really make use of it. When there's enough encapsulated IPv6 running across your ISP's network, it'll actually save them money to switch to native IPv6.

Re:I want IPv6 support, but ...

[mellon]

DD-WRT. Of course, this assumes you aren't running one of the crippled Linksys routers that don't have enough memory to support a Linux kernel...

Re:I want IPv6 support, but ...

[Anonymous Coward]

The newest Linksys RVS 4000 has a full IPv6 stack, you have to click into the setup and turn it on. The poster may indeed be doing 6to4 but it is incorrect that no Linksys routers support IPv6.

Incidentally, that model router also has the capability of blocking the popular P2P file sharing protocols, unlike all their prior models.

Re:I want IPv6 support, but ...

[stu72]

99.99 % of the people on the internet:
- do not read /.
- have no idea what end-to-end connectivity means
- have no idea they don't have it with ipv4

Re:I want IPv6 support, but ...

[Anonymous Coward]

Apple's Airport and Airport Express both have IPv6 support (and 4to6/6to4) right out of the box.

Meanwhile, DD-WRT and most other OpenWRT-based firmware still don't support IPv6, even though most of the underlying software does.

Re:I want IPv6 support, but ...

[SBrach]

And VPN and IPv4/IPv6 dualstack and a bunch of other features for $100. Great router for the home.

Re:I want IPv6 support, but ...

[Yaztromo]

I think the home router issue is the one that matters. I want IPv6, but simply cannot have it (unless I cough up lots of cash for a serious router).

Get yourself an Apple Airport Extreme. You don't need a Mac to run one, although the configuration software is Mac or Windows only (although it might run just fine under Wine). It has support for both true IPv6 packets to and from your ISP, and 6to4 tunnelling. Clients that connect to it need merely be IPv6 enabled -- on my home network (running gigabit/802.11n/IPv6) I've got OS X, Linux, and (virtualized) Windows clients all running IPv6, both within my private network, and to the public Internet at large.

Yaz.

Re:I want IPv6 support, but ...

[mellon]

Of course not. They just know that their app doesn't work.

IPv6 is depressing...

[wandazulu]

...because whoever is in charge of it does such a crummy job of explaining what it is and why I should care, and more importantly, why my folks should care.

I got my router set up to use IPv6 (an Apple Time Capsule), and I went searching for some IPv6 love and found practically none. Yes I got to Google, and yes I found a few websites that seemed to do little more than blink(!) "hooray, you are connecting using IPv6! Your address is ..."

IPv6 needs both a killer app (IPv6-only Twitter, anyone?) and some ready-to-explain-why-you-can't-get-to-it documentation that will get the people to *demand* that they have IPv6 addresses.

Until then, it's a 32-bit address space world.

Re:IPv6 is depressing...

[jd]

There's apparently free porn on offer in New Zealand for those who are using IPv6 as an incentive to switch.

Re:IPv6 is depressing...

[gbjbaanb]

IPv6 killer app... how about video-on-demand phone calls that appear on your TV set?

TVs are becoming internet-enabled, if each could be addressable, then you could add a webcam and use it as a scifi-style video phone, for free calls anywhere in the world.

You could also have your ISP push programmes to your set-top box instead of you going and fetching them.

The only 'killer apps' I can think of that'd make sense are for entertainment purposes; that are your ISP refusing to connect you to the internet because they have no IPs to hand out anymore.

Re:IPv6 is depressing...

[poopdeville]

URL?

Re:IPv6 is depressing...

[compro01]

.

Re:IPv6 is depressing...

[compro01]

I have your killer app right here.

Re:IPv6 is depressing...

[Anonymous Coward]

IPv6 killer app... how about video-on-demand phone calls that appear on your TV set?

The technology for video phones has been around for decades. Lots of people like the idea of it, but nobody wants it in practice. Saying "ah, but now it uses ipv6 for no particular reason" isn't going to make a difference to that.

Re:IPv6 is depressing...

[jd]

http://www.ipv6porn.com/

Re:IPv6 is depressing...

[paul248]

That project has been vaporware for years. And more importantly, the site has been offline for the last couple months.

Re:IPv6 is depressing...

[asdfndsagse]

that site is ipv4-only

Re:IPv6 is depressing...

[Tony Hoyle]

It's also about 3 years old and hasn't been updated in a long long time... and doesn't contain any porn.

I think it was someones idea of a joke that got out of hand.

Re:IPv6 is depressing...

[Raenex]

...because whoever is in charge of it does such a crummy job of explaining what it is and why I should care, and more importantly, why my folks should care.

I think the people pushing IPv6 are doing the best they can do. The fact is IPv6 is a technology waiting for a resource crunch that is a couple of years away. As a consumer, you should just demand that any network products you buy like routers or cable modems are IPv6 ready. Other than that, just wait.

IPv4 Address Exhaustion Is Always Be 2 Years Away

[afabbro]

Case in point. Thought it was supposed to be 2010? Now it's 2011.

IPv4 addresses won't magically be exhausted one night. They'll just start getting more expensive.

As I keep pointing out

[wowbagger]

As I keep pointing out on each IPv6 story, there will be little motivation to move to IPv6 until you can hit major sites, like cnn.com and slashdot.org, using nothing but IPv6 packets.

We've made a bit of progress, in that now, if you have IPv6 connectivity to "the Internet", you can in theory do the name resolution entirely by IPv6 packets, now that the root name servers support IPv6.

Note to the "a little knowledge is a dangerous thing" crowd: yes, you can form an IPv6 packet with an IPv4 address, but that doesn't mean the target machine will actually be able to understand it - it is still a completely different packet type than an IPv4 packet.

So, does slashdot.org have IPv6 enabled? Does the colo housing slashdot.org's servers route IPv6 packets from the Internet to the slashdot.org servers? Can "the Internet" route IPv6 packets to the colo?

If a tech site like slashdot.org doesn't have the ability to handle IPv6 traffic, then why should I get all hot and bothered about trying to get IPv6? And if I'm not going to demand it, then why should my ISP spend the effort to supply it?

Re:As I keep pointing out

[Ceiynt]

Exactly. Just like the conversion from radio to television. Once enough stuff started to get snipped from the radio, like a lot of the serials and soaps, and started to be put on television, television took off. Start making Yahoo.com and Google.com junk with IPv4, and advertise on the page why you get such crappy service and why they should upgrade, in plain enough English to the non-tech people to understand. Then wait a few years for OEM computer to ship with IPv6 compliant NICs, and offer rebates or whatnot for IPv6 routers.
Or just do like the entertainment industry does, wait for the porn to be IPv6 enchanced, then IPv4 will be dead over night.

Re:As I keep pointing out

[gbjbaanb]

Start making Yahoo.com and Google.com junk with IPv4

so *that's* Microsoft's plan to get some users for their search engine!

Re:As I keep pointing out

[Bandman]

I'm not sure you could get the tail to shake the dog like that.

Those sites are important because they are easy to use and good at what they do (ok, Google is, anyway).

Users typically follow the path of least resistance. If Microsoft Live Search was the only search engine available to people who had ipv4 and ISPs were still only handing out /32 addresses, guess which search engine those people would use.

Of course, that wouldn't happen, because Google and Yahoo would retain their /32 addresses, because they're businesses designed to get money, not force social or technological change.

Re:As I keep pointing out

[grumbel]

As I keep pointing out on each IPv6 story, there will be little motivation to move to IPv6 until you can hit major sites, like cnn.com and slashdot.org, using nothing but IPv6 packets.

That is never gonna happen, because enabling IPv6 for major sites has zero advantages and will break plenty of stuff. With IPv6 it isn't a "you have it" or "you don't have it" thing, for lots of consumers its a "you kind of have it, but routing is pretty broken" thing, meaning half the servers out there will not be reachable over IPv6, even if both parties have IPv6. And since plenty of software out there does the "clever" thing to default to IPv6, this means servers that worked before on IPv4 will no longer work once they have a IPv6 AAAA record or will only after a long timeout. Result of all this is that nobody is using AAAA records for their main DNS names, i.e. you get an AAAA record for ipv6.google.com, but not www.google.com. This kind of connection trouble of course would quickly fade away when IPv6 is actually used instead of just an experimental toy thing, but for quite a while peoples only IPv6 access will be via broken 6to4 tunnels.

Or to look at it from a different angle, there simply is no need for IPv6 for major sites, as there are so few of them that they aren't running out of IPv4 addresses anytime soon, they simply pay premium as soon as they get really rare.

The only way I can see IPv6 moving forward is if you start with it on the client side (torrent, ip telephony, whatever), since client to client communication is what is truly broken in IPv4 due to all the NAT'ing and other hacks and using IPv6 could make things there better instead of worse, since the status quo with IPv4 is already completly awful. And of course we already ran out of IPv4 addresses on the client side a long time ago, which is why the whole NAT thing become necessary in the first place.

In the end I however wouldn't expect miracles, this isn't something that is going to move much forward by free market forces anytime soon, this is something that could need a serious government push to get momentum. Or if that doesn't happen, I guess we have to wait till China, India and all the other counties see benefits of IPv6, before somebody here realizes that it might be a good idea to start with it too.

Re:As I keep pointing out

[Anonymous Coward]

IPv6 compliant NIC

WTF is that? Unless you've got an extremely fancy NIC with TCP offload, your NIC doesn't care what high level protocol you're running on it. Even if you live in a parallel universe where you can buy a cheap-assed home PC with a TCP-offloading NIC, it will still run IPv6. NICs are not a problem.

Re:As I keep pointing out

[Tubal-Cain]

Those sites are important because they are easy to use and good at what they do (ok, Google is, anyway).

search.yahoo.com isn't bad. Too bad it isn't their default.

Re:As I keep pointing out

[Midnight Thunder]

there will be little motivation to move to IPv6 until you can hit major sites, like cnn.com and slashdot.org,

I would love to see Slashdot show itself being a true "news for nerds" site, by being IPv6 accessible ASAP. There is one big issue: Slashcode is not IPv6 ready. Investigating why it turns out that certain major Perl libraries are not IPv6 ready, including libwww. I see someone has already made an attempt at offering a patch, but it has not been acted upon. I have other issues with CPAN and Perl, but that is another story.

If any there any Perl developers out there that could make the changes that would make Slashcode IPv6 ready, then that would be cool.

As for the other sites, their source is closed, so I can't comment.

Re:As I keep pointing out

[evilviper]

Start making Yahoo.com and Google.com junk with IPv4, and advertise on the page why you get such crappy service and why they should upgrade

I don't recall the evolution from radio to television require a giant conspiracy that actively degraded the quality of radio service to compel people to switch...

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Lord Ender]

ARIN really is the most trustworthy source you could have for a claim like that, though. Sure, many have made the claim before, but this is the next best thing to having Jesus, Moses, Mohamed, Buddha, and Thor all sit down with you around a burning bush and explain the importance of implementing IPv6.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[georgewilliamherbert]

They're already more expensive. The expense increase has been down in the noise for customers - that will no longer be true by the end of the year, and it will hurt by mid 2010.

IPv4 is no longer too cheap to meter. If that's not a business case for IPv6 I don't know what is.

ipv6

[robpoe]

I can get IPV6 from my co-lo provider, but my server control panel (Plesk) doesn't support it ..

So I can serve up "You see this page because you just installed Apache" ... in IPv6..

(wooo!)

We've heard this same song for years

[flipper9]

Every few months, another prediction comes out that IPv4 is doomed and that we are going to run out of addresses. Those dire predictions never come true, and the predicted date keeps getting extended. When are we truly going to run out? Nobody knows. What will happen is that IPv6 won't become a priority until things start breaking. That's just how it goes.

Re:We've heard this same song for years

[paul248]

We know *exactly* how much unallocated IPv4 space there is. Every day, that number decreases.

It's not like oil, where we can cling to the hope that we'll discover bigger reserves someday.

Re:We've heard this same song for years

[Tony Hoyle]

Not true.. it increases as well. I've seen the 'exaustion counter' go as high as 1000 days (around January IIRC).

The number of days until ipv4 runs out remains fairly constant, at around 2 years - which seems to suggest that the number returned to the pool is balancing the number of newly allocated addresses.

http://atchoo.org/ipv4/

Re:We've heard this same song for years

[Dan Ost]

You're saying that we've reached a steady-state where new IPV4 allocations are balanced by deallocations? That's hard to believe.

I would bet that what's really happening is that the rate of allocations is decreasing so the existing supply will last longer.

Now what am I going to do?

[bobbuck]

With this junk IPv4 toaster?

Re:Now what am I going to do?

[Big_Monkey_Bird]

There's a new NetBSD out...

Re:Now what am I going to do?

[Anonymous Coward]

I dunno. But six is definitely quite the upgrade over four.

Class A Address Space

[Anonymous Coward]

How about they take back the Class A address space owned by companies who probably aren't even utilizing it. Here's a list of a few companies who have class A licenses and you wonder how much of it they are even using:

General Electric 3.0.0.0 - 3.255.255.255
IBM 9.0.0.0 - 9.255.255.255
Xerox Palo Alto Research Center 13.0.0.0 - 13.255.255.255
Hewlett-Packard 15.0.0.0 - 15.255.255.255
Hewlett-Packard (originally DEC, then Compaq) 16.0.0.0 - 16.255.255.255
Apple Inc. 17.0.0.0 - 17.255.255.255
Massachusetts Institute of Technology 18.0.0.0 - 18.255.255.255
Ford Motor Company 19.0.0.0 - 19.255.255.255
Royal Signals and Radar Establishment 25.0.0.0 - 25.255.255.255
Halliburton Company 34.0.0.0 - 34.255.255.255

Why the hell do some of these companies even need 16+ million addresses? I can't see them utilizing the space available, but maybe someone here can enlighten me on how that is done (aside from trying to justify a public IP address for every workstation).

Re:Class A Address Space

[Bandman]

They missed the RFC 1918 memo

Re:Class A Address Space

[azrider]

Why the hell do some of these companies even need 16+ million addresses?

Can't answer to the others, but IBM uses it's address space for all of it's equipment worldwide (desktops, labs, wireless, etc). All of the equipment is accessible via internal LAN's for each and every building IBM is in (and access can be had via VLAN if approved). The others may have similar needs.

Re:Class A Address Space

[LackThereof]

a public IP address for every workstation

Yeah, that's right. Even if they're on a private LAN, or firewalled to hell, all the workstations are using legitimate public IPs. And back in the olden days, when most of these companies bought their blocks, if you needed more IPs than a class B there was no other option. Remember, NAT was still a long ways from being trivial to implement back then.

And it's not just workstations. I imagine for Ford, all their assembly robots have their own IP addresses. You'd need a few thousand IPs per factory. Similar situation at Halliburton, I bet there's thousands of network connected controllers in a single chemical plant.

How about they take back the Class A address space owned by companies who probably aren't even utilizing it

Sure you can say "they don't need them", but so what. They've been purchased. You can't just take back their address space.

Car analogy: That's like telling a soccer mom that because doesn't NEED her mammoth Hummer H2 to go buy a gallon of milk, that you are going to repossess it and give her a Chevy Aveo5 instead.
Sure, it might be wasteful, but it's their legal right to own and use what they have purchased.

I'm sure once we're out of IPv4 addresses, some of these companies with old class A allocations will start selling off chunks of them for a tidy profit. But until that day, they've been purchased and allocated, and are not coming back.

Re:Class A Address Space

[gregg]

Ford Motor Company 19.0.0.0 - 19.255.255.255

The addresses held by Ford might open up soon.

Re:Class A Address Space

[Anonymous Coward]

None of these companies "own" IP space because IP numbers aren't property. They have a justified need for the /8 and that is why they have it. All that is required to get and retain a /8 is proving need, and these companies did that when /8's were still available.

At the burn rate that the world's RIR's are using up /8's, even if you retrieved these /8's from these companies through the claim that they don't anymore meet the burden of proof of showing need for a /8, (which would be possible) your not going to do much more than add a few years to the deadline, then we would be out of IP numbers again. And if you were to do something like that you would merely end up causing a lot of fighting and arguing from these companies which would undermine the effort to go to IPv6.

People need to face the fact that the Internet is BIG and I mean REALLY REALLY BIG. I think it's far larger than most people's imaginations can handle, anymore.

Re:Class A Address Space

[drmerope]

Right. Most people are sitting on unaddressable addresses. The ANT census is pretty explicit on this point. Roughly 4% of the IPv4 address space is in use, 30% is not allocated at all, and the remainder (66%) is trapped due to inefficient allocations.

Re:Class A Address Space

[POTSandPANS]

Sure they could take them back, but there would be no point. Assuming they got back *all* the IPs from the companies on that list, they would get enough IPs to buy us maybe a few months at most.

Re:Class A Address Space

[DaleGlass]

How about they take back the Class A address space owned by companies who probably aren't even utilizing it. Here's a list of a few companies who have class A licenses and you wonder how much of it they are even using:

How would they take it back, after having sold it?

Should Russia just take back Alaska, because some time back they sold too much land too cheaply?

I'm afraid that giving somebody too much of something and regretting it years later doesn't entitle you to get it back.

Re:Class A Address Space

[Anonymous Coward]

Even if they're on a private LAN, or firewalled to hell, all the workstations are using legitimate public IPs.

All workstations needing Internet access might be using public IPs but you can take hundreds or thousands of workstations that will only utilize a Class C network using NAT/PAT. There are many other reasons that workstations or other devices won't don't need access to the Internet, so if they are using properly assigned RFC 1918 addresses then they aren't ultimately using public IP addresses.

I imagine for Ford, all their assembly robots have their own IP addresses ... Similar situation at Halliburton, I bet there's thousands of network connected controllers in a single chemical plant.

If they have equipment like this assigned public IP addresses, then the government would be doing everyone a favor by buying back the Class A address space. I don't care if they have firewalls in between. If someone has a network connected controller at a chemical plant with a public IP address, that is just retarded (although I wouldn't put it past government contractors to do something like that). Maybe 15 years ago that made sense - if at all.

I'm sure once we're out of IPv4 addresses, some of these companies with old class A allocations will start selling off chunks of them for a tidy profit. But until that day, they've been purchased and allocated, and are not coming back.

If there became a serious issue with IPv4 address depletion (which isn't the case right now), and these companies refuse to hand over their Class A licenses or try to charge an exorbitant fee for them, then the government can always just use the eminent domain claim and take back the address space.

Regardless of how the address space changes hands, I still think it should be done in a calculated manner so that these companies have the time to move all the necessary devices to private IP addresses.

Re:Class A Address Space

[againjj]

And they can't use 10.0.0.0 because ...? And can't use 9.0.0.0/9 because ...? And HP needs two /8s because ...? And these companies should get (not "do get") preferential treatment because ...?

Re:Class A Address Space

[m.ducharme]

Or maybe not...Ford looks like it actually might make it through the recession. GM and Chrysler, on the other hand....

Re:Class A Address Space

[againjj]

How about they take back the Class A address space owned by companies who probably aren't even utilizing it

Sure you can say "they don't need them", but so what. They've been purchased. You can't just take back their address space.

Actually, the addresses are not "owned" by the companies. They are just allocated. So, theoretically, ARIN could deallocate them. The problem is that people would object and file lawsuits; besides, ARIN has no way to enforce the deallocation of addresses, as ARIN could simply be ignored. If that happened, you now have more than one machine per address, which is bad. Besides, it would only postpone the inevitable, and not by that much.

So, the proposal won't work because it would be a lot of work, be destabilizing, and not actually have a worthwhile payoff.

Re:Class A Address Space

[The Cisco Kid]

RFC-1918, issued February 1996, came a bit late for even the 34/8 network, which was originally assigned in 1991.

You can safely assume that all the lower-numbered networks than 34 were assigned earlier than 34 was.

Note: There is no longer any such thing as a "Class A network". This has been the case for over a decade, despite far too many educational institutions having continued to teach it that way. The proper term for what used to be called a "Class A" is "/8"

Also, I'm not sure why you are calling them 'licenses'. This isn't a radio network.

Re:Class A Address Space

[Bandman]

Maybe you meant to reply to the parent?

I didn't say anything about licenses or Class A (though Class A is still really standard jargon and everyone knows what you mean when you say it).
Licenses is a little weird, yea.

Re:Class A Address Space

[bendodge]

We don't want stopgaps. They just postpone the problem a few years. A total solution is much better.

Re:Class A Address Space

[witherstaff]

Gotta love how a business class line from many telcos give you 8 IPs. I used to be in the field often working at various clients and it was very rare to see even 2 of those actually used.

I used to feel guilty using 4 IPs to have real IPs routed via openVPN back through my office. Then I got over that as I realized even a few Cs of junk wouldn't compare to the crap that's being allocated already.

Re:Class A Address Space

[LackThereof]

Ford has serious stockpiles of cash and is in no danger of going out of business any time soon, especially with the strength of their European sales.

You must be thinking of GM and Chrysler.

Re:Class A Address Space

[Anonymous Coward]

Class A is still really standard jargon and everyone knows what you mean

"I haven't read an RFC since 1993 because I'm ready to retire."

Re:Class A Address Space

[pmarini]

They still probably think that it adds to the glamour having a "Class A" network, like having an AAA score would have prevented the fall of the markets... (for your information: it didn't)

Re:Class A Address Space

[petermgreen]

Not responding to ping (or even responding negatively to ping) does not nessacerally imply not in use.

Many home routers and machines with firewall software will not respond to ping dispite actively participating in the network. I strongly suspect many companies also have machines on public IPs but with either no access to the internet, access via statefull packet inspection firewalls or even access via NAT.

Re:Class A Address Space

[TheRaven64]

Absolute nonsense. I know of several large sites that block inbound inbound ICMP packets. None of the machines on these sites will respond to pings originating outside the local network, but that doesn't mean that they don't exist.

Re:Class A Address Space

[Just Some Guy]

Gotta love how a business class line from many telcos give you 8 IPs.

That's mainly because a 4-IP block is hugely wasteful. Remove the router and broadcast addresses and you're left with 2 usable IPs. At least with a /29, you get to use 6 of 8 of those addresses.

Re:Class A Address Space

[Rich2k]

Can't we just get all IP block holders to once again demonstrate their need for each and every IP, if they don't reply or bother, simply recover them.

Re:Class A Address Space

[Yaztromo]

Addresses in very large organizations aren't necessarily used in ascending order. I'm sure IBM didn't just start assigning numbers contiguously from 9.0.0.1 (having worked there, I can virtually guarantee it). Indeed, big multi-national companies may be using their /16 networks to divide their network by geographic regions or business units, and then further sub-divide their /24's into local workgroups other other logical units.

As such, without some major renumbering work at such organizations, you can't just "reclaim" the wasted space in any reasonable manner. Routers still need to be able to get packets to the destination address (and know how to route them there), and without some form of compaction, you're not going to be able to easily reclaim addresses here and there without some serious routing table explosions. The unused addresses won't be contiguous, and getting a big organization to compact them into sufficient contiguous units that are going to be worthwhile to redistribute is going to be a very costly undertaking that they would (rightly) resist.

Re-distributing big /8's that are owned by large corporations just isn't going to happen. Getting perfect utilization for all the routable IPv4 address space was never going to happen, and never will. At least IPv6 seems to be designed with this fact in mind -- an organization can have an entire IPv4-sized address space all to themselves if they want to, and use it as densely or as sparsely as they desire, and we won't run out of assignable subnets.

Yaz.

Re:Class A Address Space

[Cajal]

No one is getting preferential treatment. There have been a handful of legacy /8s reclaimed. But even if we reclaimed all of them, it would only buy 1.5 - 2 years of extra life for the IANA IPv4 pool. Reclaiming "unused" IPv4 addresses isn't a viable long-term plan.

Re:Class A Address Space

[azrider]

Because ALL of the addresses (and all of the subnets) are accessible from the outside world through VPN's.

Cool

[Anonymous Coward]

At least we got two more years. Then we can start using phone numbers. IPv6 is insane, unless everybody carries their own DNS cache. Talk about vulnerable!...

Re:Cool

[shentino]

Massive DNSSEC usage
IPv6 rollout
Duke Nukem Forever

The order in which these will happen is probably significant.

I would hope that DNSSEC gets rolled out before IPv6 owing to the increased reliance

Set the Way-Back Machine 2 years...

[volxdragon]

...wait, didn't they say the same thing then??!?

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[oojah]

You're right. The ipv4 address report at potaroo is a prediction based on modelling and it does change. A while back I started recording the reports and plotting the changes in predictions. It's a bit disappointing that I didn't start before the world began to end because I bet the graph would be a much more interesting shape. Anyway, current predictioned date are getting further away - the number of days remaining at the time the report is made remains roughly constant.

Graphs at http://atchoo.org/ipv4/

Why not?

[cmtonkinson]

It's not just some small alarmist group raising a bell over nothing. We will run out of IPv4 at some point. Some estimates say two years, some say eight. Could be more, could be less. At any rate, we WILL run out. Why not adopt a clearly better alternative while we've got time? How can it be disadvantageous?

We need ipv4.5

[Twillerror]

I think ipV6 is to much of a move. IP addresses are nice and easy to remember like phone numbers. Yes IPv6 has short hand, but it is still harder.

Why couldn't we just add another octect. So my new IP is 1.24.101.1.15. That gives use 2^40 (~1 trillion) versus 2^128(unfuckincredibly big). We made way to big of a jump.

There is also virtually no need to upgrade to v6 for internal communications. We have 10, 172 and 192 which is more then enough for even the largest companies.

I guess we are going to become even more dependent on DNS for everything. I can't imagine someone actually typing a full ipV6 hex address. Mabye the easy ones ::::::b00b:8008

Re:We need ipv4.5

[georgewilliamherbert]

Why couldn't we just add another octect. So my new IP is 1.24.101.1.15

Fortunately, nobody in their right mind would let Slashdot design a new network protocol.

Re:We need ipv4.5

[Bandman]

Awesome idea. We'll give Google 1/40, The government can 2/40, IBM will get 3/40, etc etc etc

Same problem. The ipv6 is not a "bad" idea, it's just sort of like...imagine in 1950s if the phone company decided "we could go with area codes to subdivide numbers to prevent running out, or we could use letters AND numbers".

Can you imagine the upheaval?

In a lot of ways, that would have been even easier to deal with, because everyone's phone was owned by AT&T. New phones could have been issued without too much problem.

No, imagine it instead in the mid 1980s. Ma Bell doesn't own the phones any more, in fact there are tons of cheap phones available, cell phones are starting to come out, and there are still rotary AND push button phones.

That's more like what the IPv6 switch is like. Do you give the new people 2 numbers, so that grandma can still call them? How long is it before you stop accepting legacy phones that only have 10 dialing options? How the hell do you get DTMF to work with 36 numbers? Do we need area codes? It would be weird without them, but we don't really need them.

The equivalent of these questions are still being asked. Just a couple of months ago, there was a huge to-do about NAT and IPv6. "IPv6 is a world without NAT". The hell it is. My internal routers don't get publicly routable IP addresses, even if I have to NAT back to IPv4.

When the wrinkles get ironed out, we're going to wonder how we ever did without it. During the transition, it's going to be hell for everyone (with the possible exception of the clueless end user, who might have to buy a new router at most).

Re:We need ipv4.5

[snaz555]

Just a couple of months ago, there was a huge to-do about NAT and IPv6. "IPv6 is a world without NAT". The hell it is. My internal routers don't get publicly routable IP addresses, even if I have to NAT back to IPv4.

I agree with the sentiment - however, it's one of policy, not mechanism. NAT is a pretty poor substitute for a router that implements policy (known as a firewall). NAT has literally an all-or-nothing granularity. For instance, I might want to specify that an internal host can enable BitTorrent via UPnP, but under no circumstances can CIFS be allowed through - in either direction. An internal host sending a CIFS solicitation out does not mean a pinhole should be opened and some set of hosts (depending on cone of restriction) free to respond. NAT is just not a practical policy tool. It's an address space recovery tool. Reverse NAT, however, has some redeeming qualities for load balancing and failover - I'm not versed well enough in IPv6 to understand how they'd be implemented without NAT. (Anycast addressing, I suppose.)

But you can implement NAT in IPv6 just as much as in IPv4 if you wish. A router could appear to have a single interface ID and translate to/from that. It's largely unnecessary though since instead of a handful of IPv4 addresses you have an entire 64-bit space to yourself (and maybe even the SLN prefix, not sure about that).

IPv6 really is a major cleanup and simplification from IPv4. I'm slightly disconcerted by the increased dependency on DNS however.

Re:We need ipv4.5

[Just Some Guy]

Why couldn't we just add another octect.

Because if we're going to completely break networking, we might as well switch to something that fixes a lot of IPv4's problems (such as, say, IPv6).

Re:We need ipv4.5

[characterZer0]

That was a joke, right?

Re:We need ipv4.5

[Samah]

I guess we are going to become even more dependent on DNS for everything. I can't imagine someone actually typing a full ipV6 hex address. Mabye the easy ones ::::::b00b:8008

Although I always liked the MAC address c0:ff:ee:c0:ff:ee.

Re:We need ipv4.5

[ion.simon.c]

"IPv6 is a world without NAT". The hell it is. My internal routers don't get publicly routable IP addresses, even if I have to NAT back to IPv4.

Hi. You're ignorant. Let me educate you.

RFC3513 gives us Link-Local (fe80::/10) IPV6 addresses.
http://tools.ietf.org/html/rfc3513#section-2.5.6
These are addresses that *must not* be routed to the outside world.

RFC4193 gives us Site-Local (fc00::/7) IPV6 addresses.
http://tools.ietf.org/html/rfc4193#section-3
These are addresses that you *may* choose to not route to the outside world.

You don't need NAT. :)

Re:We need ipv4.5

[paul248]

Your claims are a bit inaccurate.

Link local addresses are local to the local network segment. They can't traverse a router at all. Site-local addresses are considered deprecated, replaced by unique local addresses: http://tools.ietf.org/html/rfc4193

Neither site-local nor unique-local addresses should *ever* be routed to the outside world. They're only for use within private networks, like RFC1918 for IPv4.

Although, if you were going to set up an IPv6 NAT (most people would consider this a bad idea), then you would most likely use unique- or site-local addresses on the private side.

Re:We need ipv4.5

[Bandman]

Thanks for the links, I'll check them out.

Re:We need ipv4.5

[doshell]

I think ipV6 is to much of a move. IP addresses are nice and easy to remember like phone numbers. Yes IPv6 has short hand, but it is still harder.

So let me guess. You type all your addresses by hand when you browse the web? There's something called Domain Name System that may be useful to you.

Re:We need ipv4.5

[ax13]

we switch to IPv6 and i know for a fact that there's gonna be hell loose at lan parties -

"Hey, what's your ip address?"

"......"

Re:We need ipv4.5

[ion.simon.c]

A link-local address can allow a computer on a LAN to reach a router. That same link-local address (or another if the router needs to use another interface) can allow that router to reach a router that's Internet-facing and has its own globally-routable IP. Right?

WRT "site-local" vs. "unique-local":
You should double-check the RFC that I linked to... I got the terminology confused, but nothing else.
Also, I read this as saying that unique-local SHOULD NOT, but may be used to bridge multiple "private" networks. (Why else would site-local addresses have a low probability of collision?) ;)

Re:We need ipv4.5

[ion.simon.c]

Also check out the my conversation with paul248. He might be correcting some of the slightly inaccurate points of my post.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Blakey Rat]

Which one of them set fire to the bush? I wanted to eat those blueberries!

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Anonymous Coward]

What would most likely happen is the backbones switch to IPv6, and then the whole IPv4 range can be NAT'd to end users. Most home routers all have the same IP address, 192.168.10.1, and they all run fine. The only difference is your "real" outside IP will be IPv6. Each ISP could assign the whole IPv4 range to their customers.

I would be more worried about IP ports, when computers and networks get fast enough to handle 65,535 connections at a time.

How much of ipv4 is dark right now?

[Marrow]

If we had a measurement that said that only 25% of the entire address space is in use at any one time, then maybe would would rethink our choices.

Re:How much of ipv4 is dark right now?

[Wesley Felter]

If there's no way to convince people to give up "dark" address space then it doesn't matter.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Bandman]

Peak Oil for the internet :-)

Re:What about my monitor?

[Anonymous Coward]

According to my maths, 2^48 addresses on a home lan is enough to individually address every pixel on 32 million 4096x2048 HD displays...

Why would businesses care?

[OrangeTide]

I have 6 IPs just for personal use. Every big networking company that controls some portion of the Internet is set for IPv4 space for a while. There just isn't room for anyone new to enter into the market. This is a huge advantage for those already established companies. I don't think they intentionally planned it this way, but the scarcity of address is a short term advantage for too many businesses for us to simply ignore that and keep pushing IPv6 as if is of some automatic benefit to everyone. Don't get me wrong, I would be thrilled if Comcast and others moved me over to IPv6. Maybe with a massive address space scanning IP blocks for SSH logins and open firewalls would no longer be as a productive use of botnet time.

Re:Why would businesses care?

[Anonymous Coward]

Brand new companies can always get IPv4 numbers from those established companies. In fact, the entire addressing assignment mechanism is setup to force them to do this. It is like when your kid first gets his drivers license you don't give him a 400HP sports car, you give him a 75HP Yugo that doesn't go more than 55Mph, downhill, with a tailwind. Once he's demonstrated he isn't going to crack up the Yugo then he can graduate to a more powerful car.
Any new Internet business needs to demonstrate longevity and need, using IP addresses assigned from an existing business, before they are going to be approved for their own numbers. This is just as true with IPv6 as it is with IPv4.

Re:Why would businesses care?

[paul248]

I have 6 IPs just for personal use.

In 20 years, when everyone uses IPv6, this comment will be analogous to "640k ought to be enough for anybody."

You have six IPs, whoop-de-doo. I have 2^80 of them, and the address space is so expansive that nobody even cares.

Re:Why would businesses care?

[OrangeTide]

I would have him walk until he can buy his own damn car.

(why are there all these car analogies on Slashdot when geeks usually don't know jack about cars?)

Re:Why would businesses care?

[OrangeTide]

Well they are assigned to 4 different servers. Do you think an individual needs more than 4 colo'd servers?
That was a rhetorical question, the answer is No, they do not.
Most people have zero IPv4 IPs because they don't need their very own permanent address to use the internet.

I agree

[Anonymous Coward]

People have had plenty of warning already, they should just one Saturday give away the remaining IPv4 addresses. And perhaps several key organizations should turn off IPv4 on all equipment they control. Yes, it will cause plenty of havoc and people only have themselves to blame, and moreover can't expect the people who actually implement things (do actual work, as opposed to the PHB's) to work around the existing crap forever.

No killer app needed, just sensible migration path

[Estanislao Mart�nez]

IPv6 is depressing, because whoever is in charge of it does such a crummy job of explaining what it is and why I should care, and more importantly, why my folks should care.

Actually, I would claim that that's not a big deal. The big problem is that IPv6 just doesn't provide a sensible migration path from IPv4. The idea that we're all going to wake up one day and switch off IPv4 at once just doesn't cut it. More precisely, an IPv4 node just has no way of talking to an IPv6 node. If we built some sort of standardized IPv4-to-IPv6 NAT technology that was invisible to existing IPv4 nodes, then IPv6 could be adopted gradually and incrementally with minimal cost (the cost could be rolled into the cost of general network gear upgrades).

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[fbjon]

This all sounds like peak oil. Addresses won't run out, but the business case will magically spring up when IPv4 addresses become more expensive than implementing alternatives.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[characterZer0]

If the Pope declares ex cathedra that thou shalt use IPv6, I will convert to Catholicism immediately.

Why v4 - v6?

[Anonymous Coward]

Why no IPv5?

Re:Why v4 - v6?

[petermgreen]

IIRC a value of 5 in the version field was used for some special purpose protocol (I can't remember the details right now and I CBA looking them up, if you really care look it up yourself). So ordinary IP went straight from v4 to v6

Why?

[NetCow]

They can pry IPv4 from my tired, RSI-affected-due-to-:-overuse fingers. I'm not about to rely on DNS when debugging and testing my networks. Sure, it's a petty, localized, small scale view, but I don't care.

Re:Why?

[idiotnot]

I manage two /48 IPv6 netblocks. I can remember them just as easily as I do v4 addresses. While autoconfiguration is the preferred method for v6 devices, you can assign addresses manually. So, the host that I have on (my.ip.prefix).20, is also (my.ipv6.prefix)::20.

Schools and universities should start now.

[SigNuZX728]

Most schools and universities are not affected so much by the economy, and most schools and universities don't need a business case to make this switch.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Ironchew]

To a monopoly, IPv4 is a better business case. The artificial scarcity of less than 2^32 addresses will keep the cash rolling in.

Latest report

[mrstrano]

Actually I've just read a report from Netcraft that shows that IPv4 is dying!

economics as usual

[shentino]

Again, the problem is hoarding of unused IPv4 addresses.

We'd be just fine if it weren't for folks like MIT that have way more IP's than they need.

Of course, when a resource gets tight, the folks who have it become kings. You can bet your behind no company is going to give up it's v4's without a fight.

I'm glad that IPv6 is based upon a stewardship model rather than an ownership model. And also that the v6 guys are leaving 87 percent of the potential v6 namespace unallocated

Re:economics as usual

[davburns]

No, the problem is that we really do need more address space. IP addresses include identification information and network topology information. We really do have almost that many computers, and almost that complex of topology.

Forcing the holders of large legacy allocations to give them up would hurt more than moving to IPv6, and it'd only get us a few more years of IPv4 growth. Opening up the class-E space would also hurt more than moving to IPv6, and still only give us a few more years.

NAT effectively adds 16 more bits to the address, but does so on a per-connection basis, not a per-node basis. It requires the network to be stateful, instead of just passing packets while the end hosts carry all the state. (This means that the end hosts can't just route around problems.) NAT is messy, but it happens to work because it can steal some bits of TCP or UDP to make up for not having enough in the IP header.

IPv6 adds way more address space than anyone can think of a use for. So it can encode a lot of information about the node's position in the network, plus keep an address unique for (practically) ever.

Re:economics as usual

[shentino]

My point was that had v4's been from the beginning treated like the scarce resources they are now proving to be, we wouldn't be in this mess right now.

Perhaps later we'd want v6, but better stewardship from the beginning would have prevented the present crisis.

And also, notice that there is no BUSINESS case for it, even though there's plenty of technical merit.

Business, not technological merit, is vetoing v6 right now. Translation: The current holders are greedy and won't budge unless we answer "what's in it for them"

I will happily admit that v6 will make things much easier. However, the corporate entitlement mentality is definitely making things worse than they need to be.

Spam blocking

[Anonymous Coward]

if you think spam is bad now wait till the spammers go ipv6, those DNSBL lists will be impossibly large (unusable)
when the spammers have trillions of addresses to choose from the stream will become an ocean

Re:Spam blocking

[Miamicanes]

> if you think spam is bad now wait till the spammers go ipv6, those DNSBL lists will be impossibly large (unusable)
> when the spammers have trillions of addresses to choose from the stream will become an ocean

In theory, yes. In all likelihood, no. It looks like ISPs are going to be expected to assign 48-bit prefixes to "users" (site, customer, household, office, etc). So even if a spammer picks random 80-bit IP addresses in his subnet to spam from, it'll be obvious where the ultimate source is located. Also, for the near future at least, most IPv6 addresses will have "2001" as their first 16 bit prefix, which basically leaves 32 bits of unique addressing space to share among internet users of the world. Put another way, "zero day" IPv6 is basically going to be the same real address space we have now (32 bits), but tweaked a bit to make the routing and allocation easier. Even when the 2001 prefix gets outgrown (I'm guessing 2-5 years), there won't be any compelling need to massively expand the first 16 bits beyond a few more values (say, one per regional authority... 2100 for ARIN, 2200 for RIPE, 2300 for APNIC, 2400 for AFRINIC, etc).

As long as ISPs are somehow restrained from assigning fewer than 80 bits of address space to customers, the total number of values that need to go into a blacklist table will be roughly the same as now (while the "2001 era" lasts), maybe 4-8 times as large as it is now once addressing expands out into a second-gen prefix per regional coordinator, and at worst 65,536 times as large as it is now if the upper 16 bits were fully and arbitrarily utilized.

Hunt my posting history and read the post I did before this one... it explains why IPv6's ridiculously huge address space isn't likely to be quite as gruesome to deal with as it was originally intended to be.

Re:Spam blocking

[paul248]

Also, for the near future at least, most IPv6 addresses will have "2001" as their first 16 bit prefix

False. Off the top of my head, I've seen at least 2001, 2a01, and 2620.

For reference:
http://www.iana.org/assignments/ipv6-unicast-address-assignments

Might be closer than we think...

[Anonymous Coward]

Networks need to start migrate to IPv6 as soon as possible. The site www.ipv4depletion.com is a good starting point to see how bad the problem actually is.

And hording addresses will not do you any good. What are you going to use them for when the rest of the world is using IPv6?

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[ekhben]

All you wanted to know about IPv4 exhaustion, and more.

Predictions aren't facts. They're guesses. The assumptions that go into them can change, and given the number of factors that affect Internet usage growth, they *do* change. The current best guess suggests 2012; the past six months have seen a bit of a reduction in growth, likely due to some sort of global recession.

And, IPv4 exhaustion is a fairly well defined term, meaning either the date IANA allocates its last /8 or the date an RIR allocates its last free block. Price won't go up until *after* exhaustion, because before then, all you need to do is demonstrate a need and you get your allocation from your RIR for the same annual fee everyone else pays. Exhaustion will be an *event*: it will happen at a specific time and date.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[oldhack]

"this is the next best thing to having Jesus, Moses, Mohamed, Buddha, and Thor all sit down with you around a burning bush and explain the importance of implementing IPv6."

Yep, but that's not like hearing cthulhu's ghetto hoopti pulling up your drive way.

A message from our sponsors...

[jonaskoelker]

When IPv6 was announced, one of the benefits was that everything could have its own IP address; even your toaster!

Wait a minute... Is IPv6 just a clever marketing scheme for NetBSD?

Short-sighted management?

[jonaskoelker]

Others figure that if it's over a year away, it really does not matter because it won't impact their bonus this year, so it may not work, but we can hope.

Emphasis mine. Don't they think about next year's bonus too?

I don't know about you guys, but my attitude to work is this: do good work that's valuable to your employer for a reasonable compensation, and prefer to do The Right Thing(tm) when justifiable.

Do PHBs have a different attitude?

Re:Short-sighted management?

[Anonymous Coward]

Next year they'll STILL get their bonus even if the company doesn't make as much money, or starts losing it because 'They upgraded their equipment for the future!'

Network effects prevent the killer app

[jonaskoelker]

IPv6 needs [...] a killer app

And if I own the killer app, can you please explain to me why I don't offer it over via IPv4 also, and multiply my ad revenue by $BIGNUM?

Maybe if I'm someone like Mark Shuttleworth who is willing to gamble money on cool technology with a hope (but no certainty) of making it self-sustaining (or possibly breaking even).

But there are only so many of those people.

Re:I want IPv6 -- swine flu ...

[Anonymous Coward]

The top level authorities are working like mad to arrange the infrastructure for the new flu shot. It might not be here until next September. At lower levels they are trying to calm everyone down, and quietly hoping lots of people get this while it is a non killer flu. That way if it becomes stronger next year it will spread more slowly and infect fewer people. They all are really working quite hard now. Although the ones slaughtering pigs are pretty misguided.

Killer app

[coryking]

I'd argue there is never going to be a killer app for IPv6 because it is nothing more than window dressing on the same old, boring protocols. The true killer app will be on a protocol that is nothing like TCP/IP... say a working mesh protocol where there is no notion ports, IP addresses or any of that nonsense. Where you don't care where the data you get comes from as long as it is authentic. That is the future. Bit torrent is the closest we have to that future and bit-torrent is nothing but a hack of TCP/IP. If the protocol stack was built from the ground up to not care about the source of data, only that it is authentic, *then* you'd have a killer app.

IPv6 is boring and it isn't even mainstream. How about we cook up something new. Remember when TCP/IP was the new kid on the block and most games had dual or tri-network stacks (TCP/IP, IPX/SPX, Netbeui)? It only took like a few years before all that nonsense went away and we all settled on TCP/IP. Basically, overnight. The same thing *will* happen again. Only it will *not* be IPv6. Mark my words. We've outgrown what IP gives us... The mesh is the future.

Re:Killer app

[daveime]

Yup, I can just imagine the future.

The seed/leech ratio on public trackers is abysmal already, even most private trackers suffer from this depending on the zealousness (or lack of it) displayed by the admins and the ratios and rules they set.

And despite whatever killer app materialises, within a week there'll be leech mods available for it to ensure you don't waste your precious upstream helping serve "the web" to your neighbors and friends.

It'll never work on a "trust" model ... it has to be something enforced, just like IP4 and IP6.

Re:Killer app

[Anonymous Coward]

What the fuck?

That's stupid.

Re:Killer app

[TheTurtlesMoves]

Where you don't care where the data you get comes from as long as it is authentic.

So what does authentic mean in this context?...With no address and no ports it didn't come from anywhere and, well, can't be going anywhere either.

The mesh may not be TCP, but it will be packet based. A widely deployed globally addressable packet protocol. Say like IPv6?

There is a lot more to a network stack than TLA. Though I am happy to see ATM getting phased out now.

IP itself is hierarchical, that is the problem

[coryking]

Nobody will adopt IPv6 because it is just a larger tree. It doesn't scale the way we are now using it. The way we are starting to use our network is peer-2-peer--dare I even say "cloud-like"?

We dont care where the information comes from, only that it is the real deal. It could come from some data center, some server pool, a microwave, the cell phone, the car stereo, or your neighbors TV... doesn't matter. As long as I know the data is authentic, the source doesn't matter. That is exactly what bit-torrent is about. Only bit-torrent needs a tracker because of the deficenies of TCP/IP. If the network was all about data and how to get to it, rather than maintaining connections between two devices, we wouldn't need trackers or bit-torrent. And when you think about it, this is how it needs to be. Otherwise all the traffic has to aggrigate through larger and larger "central" links--down the tree and back up the tree to the other side. That is what we have now--it is the mindset of IP... you start and the edge node and work in than out to another edge... This doesn't scale and it gives a lot of power to the guys with the big pipes (i.e. your cable company or mega-ISP). Bit-torrent is really a mesh of interconnected goo. That is how it should be--only as a fundamental feature of the network. Focus on the data, not on end to end connections.

IPv6 is more of the same. The fact that it is hierarchical is a bug, not a feature.

Re:IP itself is hierarchical, that is the problem

[Anonymous Coward]

Are you sure you aren't confusing your network and application layers?

Re:IP itself is hierarchical, that is the problem

[jd]

What you are talking about is called "Content Addressable Memory", when it is at the computer level. I'm not sure CAM at the grid/cloud level is well-enough developed yet for there to be any formal term for it.

Re:IP itself is hierarchical, that is the problem

[Dan Ost]

I believe that it's already a part of Plan 9. Perhaps they already have a word for it.

Re:IP itself is hierarchical, that is the problem

[jd]

They probably do, but I don't speak Martian.

I'm intentionally confusing them

[coryking]

Because I believe they are becoming a hindrance to moving onto something better. The future will have layers, yes, but it might not look like the ones in the OSI model.

Re:I'm intentionally confusing them

[Anonymous Coward]

We can't even migrate from IPv4 to IPv6. What the hell leads you to believe that we're going to replace the OSI stack?

Stupidity

[xous]

Why the hell doesn't ARIN just refuse to issue IPv4 allocations to providers that aren't ready for IPv6.

They could also just set a hard deadline for anyone with an allocation to support IPv6 or their IPv4 allocation is forfeit.

We NEED IPv6 because the asses that have huge allocations will never willingly give them back and probably involve lawyers to fight it out.

I've been harassing my ISP for years to provide native IPv4/IPv6.

Ha Ha

[Midnight Thunder]

Instead of hogging these, they should just give them up. They don't need all these addresses.

Sounds like you have never worked with large corporations? Believe me this isn't going to happen, since even if it was technically feasible the politics and bureaucracy would just you insane. Anyhow, who is to say that whole blocks aren't being used? Chances are you will find many partially used blocks instead of lots of block that are completely unused, and that different bickering departments are holding on to different blocks. Believe me, moving to IPv6 is a much more realistic and simpler way forward.

Re:No killer app needed, just sensible migration p

[Midnight Thunder]

The big problem is that IPv6 just doesn't provide a sensible migration path from IPv4.

In many ways this is true, but at the same time there is only so much a sensible migration path can do. As some point you will get to the IPv6 haves and IPv6 have-nots. Those who don't have IPv6 will have to decide whether they need it or can find a way to getting it. In many ways this is no different than when people were still moving away from propriety network protocols to IPv4, the difference is the number of users.

If we all start having systems using dual IPv4/IPv4 stacks now, and having real IPv6 connectivity now, then things won't break when suddenly IPv4 addresses can no longer be allocated. The thinking here is that at that point people won't know whether they are using IPv6 or IPv4, unless they are specifying numerical addresses. This is why the delay in making IPv6 available to everyone increases the chance of breakage becoming evident.

BTW I believe Vista is already using Teredo tunnelling standard, though I would like someone to be able confirm this?

Re:No killer app needed, just sensible migration p

[ajuricaba]

IPv6 is depressing, because whoever is in charge of it does such a crummy job of explaining what it is and why I should care, and more importantly, why my folks should care.

Actually, I would claim that that's not a big deal. The big problem is that IPv6 just doesn't provide a sensible migration path from IPv4. The idea that we're all going to wake up one day and switch off IPv4 at once just doesn't cut it. More precisely, an IPv4 node just has no way of talking to an IPv6 node. If we built some sort of standardized IPv4-to-IPv6 NAT technology that was invisible to existing IPv4 nodes, then IPv6 could be adopted gradually and incrementally with minimal cost (the cost could be rolled into the cost of general network gear upgrades).

Haven't you heard about Dual-Stack Lite (DS-Lite)? http://tools.ietf.org/html/draft-ietf-softwire-dual-stack-lite-00

An extra octect is not much different

[Midnight Thunder]

Why couldn't we just add another octect. So my new IP is 1.24.101.1.15.

If you are going to add another octect, which will invariably still break many things, you might as well just accept that going with the extra 12 octects is not much worse.

Maybe it's because IPv6 is the wrong solution

[godglike]

There are 2 problems with IPv6 that are common problems with non-solutions:
1: It is highly disruptive
2: It doesn't solve the problem

The FUNDAMENTAL problem with IPv4 is that it has a LIMITED amount of addresses.

To fix the problem the solution needs to remove the limit, NOT make the limit further away.

IPv6 still has a limit to it, so when you explain to people why they should move to IPv6, you also explain why they should distrust IPv6 as a solution.

When you then explain that it will require a complete re-structuring of the internet infrastructure, they will realise that it can't work. Even if it could, they can safely wait for everyone else to do it first.

So, does anyone have a suggestion for an enhancement to IPv4 that is backwardly compatible and allows for an ever-increasing address space?

Re:Maybe it's because IPv6 is the wrong solution

[elysiuan]

ipv4 uses 32bit addresses giving a total maximum of 2^32, or 4,294,967,296 addresses.

The human mind is not good at thinking in exponential terms, and also possesses no intuitive feeling for very large numbers. Thus it is not immediately apparent (unless you've some familiarity with it already) that just 'adding 1' to the exponent doubles the final result.

So if ipv4 used 33 bits for addresses rather than 32 we would have 2^33, or 8,589,934,592 addresses available.

ipv6 on the other hand uses a 128 bit address. 2^128 gives us 3.403 x 10^38. Using a rough estimate of the current world population (http://www.census.gov/main/www/popclock.html) this gives us 5.02 x 10^28 ip addresses *for every man, woman, and child on the earth*.

In essence, every human on the earth could have 1.17 x 10^19 *ipv4 internets' worth of addresses* allocated to them before we run out.

So yes there is a limit. There is also a limit on the lifespan of a proton.

Neither of these limits poses any difficulty for the far foreseeable future.

Where did the addresses all go?

[zmollusc]

Odd, according to the news, 'OMG teh economy are collapsing!!'. Since thousands of multination corporations are going bankrupt every day, why can't their address space be recycled?

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[paul248]

Running out of ports isn't a very significant problem.

The OS tracks a connection as a tuple of (local ip, local port, remote ip, remote port). You'll only run into problems if you try to create a lot of connections to the *same host*. A typical server makes a small number of connections to any given client, so the port numbers can be reused.

If you really do want to create tons of connections between a pair of hosts, you could just add more IPv6 addresses to one of the endpoints to get virtually unlimited port space.

Re:IPv4 Address Exhaustion Is Always Be 2 Years Aw

[Tony Hoyle]

Who pays for ipv4 addresses? They're handed out free FFS.

There are some ISPs who see it as a cash cow (some of the more unscrupulous ones even try to charge you *monthly* for them) - but it's got nothing to do with scarcity.

No, we need Real-IPv4

[Anonymous Coward]

Instead of using numbers from 0 to 255, we could use four Real numbers. Or how about going real imaginative and use a tuple consisting of complex numbers as an IP address?

Thank you

[JSBiff]

Thank you, for putting the lie to this oft-repeated non-sense. Every time there is a discussion about IPv6, someone (usually multiple someones) repeats this rubbish about "IPv6 is not a solution because it too is limited - just a larger limit). Yes, it is limited but it is so large that mankind will not run out of addresses until we either literally spread to many millions of planets throughout the galaxy, each populated by Billions of people, and/or start creating unfathomably large numbers of Nanoscopic electronic devices and individually addressing them with IPv6 addresses. I suppose when we reach the day when your body is filled with nanites that repair damaged cells, destroy cancer and virus infected cells, supplement the immune system, etc, so that you can have almost-immortality, at that point, we *might* start coming close to exhausting the IPv6 address space.

Because, even if everyone has TVs, cars, computers, cell phones, bicycles, refrigerators, stoves, microwaves, home security cameras and security control systems, etc, etc, individually IPv6 addressed, we wouldn't come anywhere near to running out of address space with IPv6.

The problem is...

[SecurityGuy]

...they keep saying that in $SMALL_NUM years we'll be out of IP addresses, and $SMALL_NUM years goes by without incident. The sky persistently fails to fall.

Call it the peril of poor predictions, but I'm now officially not worried because the claims have so often been false.

Hoarding and .gov

[Spyder]

The point about hoarding is a big one. The amount of address space held by US governement entiies ie huge. I've worked with/in several .gov networks and the address allocation in most can chartiably called ineffecient. There are networks of 25K - 50K hosts that use multiple class B and class C allocations, with everything using routable IP addresses, regardless of need.

IPv6 will create some serious growing pains. We have more 20 years of the world wide web and IPv4 w/VLSM experience as an industry. There's a number of things we take for granted in the conventions, and even the protocols that IPv6 can put into question.

ZOMG WEER GUNNA RUN OWT UV IPZ!

[Benfea]

How many years has it been that we've been getting these "we're gonna run out of IP addresses" articles? It's no wonder no one takes these warning seriously: we've all been subjected to a dramatic demonstration of the boy-crying-wolf parable for some time now.

Re:ZOMG WEER GUNNA RUN OWT UV IPZ!

[Anonymous Coward]

Yep; they told us years ago it was going to happen; gave us updates every few years, had the new protocol ready in time, and everything.

Should of just saved telling everyone until there was only 90 days left, and then watched the fun!

For the last 16 years or more...

[cardpuncher]

... the IPv4 address space has been "about" to run out. And indeed, the "officially" allocated address space is depleting, though the correlation between "allocated" and "necessarily in use" is tenuous.

However, even if the number of addresses available for "necessarily in use" interfaces were disappearing fast, it does not automatically follow that IPv6 is the solution.

The real drawback with IPv6 is that, as long as 32 bits are enough, there's no incentive to consider all the training/testing/form-filling that's required to deploy something you won't use - but unless everyone (or pretty much everyone) deploys it before a 33rd bit is inevitable, the "post apocalypse" guys on the network are in some sort of partitioned wilderness. You're not going to get a wide uptake of IPv6 by IPv4 users unless there's a real threat that their IPv4 address will stop working and that's not going to happen.

It's also not the case that we currently have a fully-connected and addressable network of endpoints. In fact it's pretty disconnected and some truly atrocious hacks are required to get application-layer protocols to deal with address-translated networks. IPv6 isn't going to make that go away.

It's much more likely that IPv4 will stay as a network access protocol and there'll be some further application-level hackery (faking DNS results to return temporarirly-assigned IPv4 addresses, etc) which allows the network to transition to a larger number of endpoints in an evolutionary fashion. It won't be entirely pretty, but at least it will be (humanly) possible.

Unfortunately, the more time people spend peddling the idea of IPv6 as a solution, the less time there'll be to cobble together the inevitable grubby hack.

Stop whining & add IPv6 addrs to your public s

[Anonymous Coward]

You don't need to change your internal network. You don't need to believe in IPv6.

You just need to add IPv6 addresses to your email and web servers, and if this is beyond you, just get someone with real skillz to do it, instead of whining like some unemployed mainframe programmer.

Actual stats on assignment

[jcurran]

Actually assignments from RIR's to customers are shown here:
<http://www.nro.org/documents/presentations/nro-jointstats_03-31-09.pdf>, on page 4.

This has been growing steadily, and was more than 12 /8 equivalents. This indeed is not "12 to 14", but shows that the reclamation of several 'Class A" addresses would still only give us more months, not years.

nice try - predictions right on track

[Anonymous Coward]

...they keep saying that in $SMALL_NUM years we'll be out of IP addresses, and $SMALL_NUM years goes by without incident. The sky persistently fails to fall.

Call it the peril of poor predictions, but I'm now officially not worried because the claims have so often been false.

Actually, quite incorrect... the number of years left has changed because we're actually getting closer. The IETF working group looking at this said [in 1994]:

The linear growth model, presented by Tony Li, included these last two data points while the logistic model, presented by Frank Solensky, did not. Both models currently suggest that IPv4 addresses would be depleted around 2008, give or take three years.

http://mirror.switch.ch/ftp/doc/ietf/ale/ale-minutes-94dec.txt

Not bad for 15 years ago... Now what excuse are you going to use for not being prepared when we run out?