Slashdot Mirror
IETF draft on different IPv4 addressing scheme
skuzbunny writes "The IETF [?] draft The Mathematical Reality of IP Addressing in IPv4 Questions
the need for Another IP System of Addressing has some really interesting comments on IPv6 [?] . Quote: "I was indeed successful in the
elimination of the problems associated with IP Address Flooding
inherent in IPv4 and the complexities of IPv6. In short, small
business and single family dwellings can now have the option of
having their own private IP Addressing Scheme," " Interesting, particularly if I understand the math correctly. Can anyone who's actually qualified to comment on this comment below?
The server seems to be a bit /.ed at the moment, so I haven't read the whole thing. I don't think there's any maths in the first few hundred lines though. In my experience you can tell a reasonable proof even if the language is poor. I couldn't see any coherent logic flow here.
I couldn't stand reading the whole thing, but here's what I got out of it: He's saying that first, Class D and E aren't being used, and we could simply use those addresses. It's actually not a bad idea. His second idea is idiotic. He's saying that the binary addresses don't need to be 8-bits. Right now, every address is 8-bits, even if the address is 32.32.32.32 (0001 0000.0001 0000.0001 0000.0001 0000 in binary) He's saying there's an alternate address of 10000.10000.10000.10000. Unfortunately, we'd have to re-write how TCP/IP works in order to do that (so why not just implement IPv6 is my question). If you didn't update, you'd choke when trying to get to the binary 10000.10000.10000.10000 website.
Winners tell stories while losers yell deal.
--
The problem that many of my detractors (who Should be Obvious to you by now). Is that They have more problems with, ( of course ) the subnet of my presentation ( table 1 ). Needles to say, Nevertheless. That they more than Likely do not comprehend ( of course ) the Fundamentals of the I'm a Fucking Retard Rule ( Needless to say, similar to my Octet rule ).
Never the less, it should be Obvious why I didn't ( or should i say, Couldn't ). Needless to say, pass the fucking Cisco exam because my head ( or never the less, what is on top of my head ) is so far.
Just imagine! Shoved up my ass, that this paper should be my addmitance paperwork out of computer ( or network ). Consutlting/IT Professional, and into scooping M&M's for Dary Queen.
if you read this hampsters paper all the way thru.. take off two points. Take off 3 if you printed it out to read it later.
guns kill people like spoons make Rosie O'Donnell fat.
There's absolutely no situation where having a clashing namespace is better in any way!
--
This paper reminds me of an article I read a long time ago (1988?) in Scientific American. The author (Professor Arlo Lipof) claimed to have invented a mathematical equation that allowed him to cut a 1"x5"x8" block of gold, and reassemble it into a 1"x8"x8" block (which resulted in a volume increase of about 1.5%.) The article was complete with diagrams and went on for 3 pages on the topic, very much like this paper.
:o). If this is a joke, he's a little out of season.)
(The SA article immediately activated my BS meter, but I got about 1/3 of the way through before realizing that it was published in the April edition try to make an anagram of Arlo Lipof, and see what you get
The main problem with IPv4 that IPv6 is trying to solve is a lack of address space. By using IP masquerading, that problem can be alieviated indefinately, at the cost of increasing the lag time. You get one IP address, which you then use IP masquerading to get up to 2^32 (minus oddballs like 127.*) addresses internally. And if that's not enough addressing for you, you can run IP masquerading on each machine of your internal network, increasing the layers indefinately.
:)
IPv6 is way too scary to actually work
--- A Jesus Fish eating a Darwin Fish only proves Darwin's point.
If his math is anything like his grammar, you can basically write it off straight away. And if it's not, it's still impossible to work out what he's really trying to say since he's not communicating with any sort of precision.
My translation:
Jen stepped over to the couch, slowly rocking her hips with each step, accenting the graceful curves of her body. She quickly move in next to him, noting the warmth coming from her lover. His warm hands started at her thighs, and crept up until they were her under her red sweater. He moved his lips next to her face, giving a quick nibble or her ear, and losing himself in the scent of her soft blond hair. She moaned softly, and brought her face closer to her man's ear.
"Rob," she moaned, "show me your Commander Taco."
How's my translation?
--------------------------
I can't believe someone could actually write prose like that. My best guess right now is that the author wanted to post an encrypted message and used a high-order Markov model to encode the ciphertext as a plausible English document.
;-).
The training set for the model might be real RFC's, or possibly the U.S. Congressional Record
Blech, I haven't read such bad prose since I took technical writing in college. And no, it wasn't from other students, but the postmodernist drivel the teacher forced us to read as part of the class.
Which gives me an idea.. what if this article is in fact a hoax, à la Alan Sokal, but directed toward the Internet community by some spiteful English lit student? Take some bogus mathematics, sprinkle in some jargon with a rudamentary understanding of network architecture, and mix it together in a dense, grammatically flawed style. "Ha! Those nerds will never know the difference! Now the jokes on them! *cackle* *cackle*"
I was playing fast and loose with words in my post because I am trying to explain highly mathematical concepts to readers in a way they can understand. A friend of mine once said that sometimes the incorrect explanation is just clearer.
You're right of course. Strictly speaking the problem is that real life doesn't let you apply the axiom of choice. And uncountability is not the feature of the axiom of choice that leads to the BT paradox.
But if I sit here and try to explain all the finer mathematical details, my post becomes 3 times as long and 1/100 as clear.
I've made it through most of the article. AFAICT, he postulates adding bits as a method of getting around the number-of-addresses problem, and proposes a different way of organizing subnets.
The *one* (1) saving grace that his article has is that his proposed organizational scheme makes it relatively painless to increase the number of bits down the road, without having to reassign addresses. OTOH, it's easy enough to do that with the present system too (treat your 4-byte IP as the _least_significant_ part of a larger address).
The article was poorly organized and incredibly obfuscated. I really do hope that this person isn't really a member of any decision-making organization. I could give a summary containing all of the useful information on it in a tenth the space, and more clearly.
In fact, I'm seriously considering doing this just so that nobody has to wade through this monstrosity in its original form.
This is, of course, why the 10.x.x.x and 192.168.x.x networks are there to begin with; they're specifically setup as non-routeable addresses for firewalls (NAT/IPMasq or otherwise).
---
"'Is not a quine' is not a quine" is a quine.
"'Is not a quine' is not a quine" is a quine.
Quine "quine?
I happen to like NATs - they are a good way of making sure that the network inside my workplace or home isn't visible to the outside world. As far as the ISP is concerned, my house consists of the firewall machine, and my workplace consists of a firewall and a mail server, which IMO is as it should be.
I readily agree that using NATs as a means of packing more machines into the address space is a Bad Idea - I'd like to have the potential for more than a few billion world-visible boxes. They're also a bad idea on an internal network that has to be able to see all parts of itself from all parts of itself, and for cell phone networks. However, I don't see why they're intrinsically evil.
I haven't had a problem running games behind a masquerading firewall. Tribes 'net play works fine. Quake 'net play works fine.
Look at table 4, and you'll see he's getting the extra combinations from the subnet bits. He's seems to be operating under the delusional assumption that the subnet mask somehow floats alongside the IP address. Combine that with his delusional ramblings about decimal vs. binary vs. hexadecimal. (Dude, those are just representations, just like 90 degrees is PI/2 radians is a right angle! I could write the numbers in octal if I wanted to and it wouldn't change their values.)
The reality, of course, is that we can get at most 2**32 (~4.3 billion) globally unique addresses if we completely remove any artificial partitioning and special encodings that would use up encoding space. This guys "mathematical proof" reads like some of those "random data compression" patents that Jean-loup Gailly (of Info-Zip/ZLib fame) likes to discredit on his homepage.
Of course, having partitions and special values does simplify things alot, which is why we don't get all ~4.3Billion addresses. Just look at RISC computers which use a 32-bit opcode. They sure as heck don't implement 4 billion different instructions.
--Joe--
Program Intellivision!
It looks to me like this draft is saying:
A) The author feels nobody explains IP Addressing well;
B) There is some discrepancy between the standard decimal representation of an IP address and the standard binary representation of it;
C) The original class A/B/C method of assigning IP addresses is obsolete;
D) The 32 bit IPv4 system could be used for another hundred years without upgrading to IPv6 if you use some obscure addressing scheme that appears to depend on B, above, and hiding some of the address in the subnet mask;
E) Adopting this scheme will be easier than teaching people how to use IPv6.
Well, point A is obvious, if he considers this draft to be a "logical...explanation", than no previous documentation would quite pass muster.
He provides no clear evidence for point B. The number 119 is the same if you represent it in decimal (119) or binary (01110111). If this is not the case, I want to hear it from a mathemetician, not an IETF draft.
Point C is true, that's why we no longer use it. He apparently has either not read or not understood RFC 950, which describes how to get away from the unnecessarily coarse class A/B/C system, without using his equally coarse class A-1/A-2/A-3/B-1/... system.
Point D is not adequately documented to be of any use to anyone. The current IPv4 address allocation scheme still has a lot of wasted addresses, which could extend its life if tapped. I can't even tell if this scheme taps them, or if it just pushes big words around on the page.
Point E is false in this instance, since fully grocking this draft is much harder than understanding and implementing IPv6. Even if it is translated and better explained, I doubt any scheme to tap a significant number of wasted IPv4 addresses would be easier than just upgrading to IPv6. This is because most of the waste is considered "expansion space" by the owners of the network addresses. Any use of these addresses would require not only reprogramming many routers, but spending a lot more time maintaining the resulting routing tables as addresses here and there get used.
The bottom line, IPv4's not dead yet, but IPv6 is still inevitable, and this paper proposes nothing coherant.
----
----
Open mind, insert foot.
There was actually a Discover magazine article a few years back that talked about the possibility of space being quantized. While I haven't heard anything about the subject since, I assume that's because the question is still open, not because it's been settled one way or the other.
Energy and matter are quantized, so it is certainly conceivable that space and time are also quantized. Again, unless you have clear evidence to the contrary, I don't think the possibility can be dismissed.
In any case, there is an additional difficulty with applying the Banach-Tarski paradox in real life: you do have to make an uncountable number of very exacting, precise choices at once. Considering that there are only countably many seconds (and possibly even finitely many) in the lifespan of the universe, it seems like it would be difficult to pull that off.
Just because something is out there mathematically doesn't mean we'll ever see it in the real world. For instance, the decimal expansion of pi is infinite nonrepeating. We will never see all the digits of pi laid out in sequence, since there are only finitely many atoms in the known universe, and hence only finitely many sheets of paper to write it on. The B-T paradox is the same kind of thing. I'm quite confident that you will never be able to achieve it in real life.
--
I think everyone is forgetting that IPv6 does more than give us ~2^128 IP addresses. IPv6 also tries to make performance improvements. For example, in IPv4 any router is allow to fragment packets to squeeze them through the hardware's MTU. In IPv6, fragmentation is only allowed at the source of the packet. This means that the MTU for the entire path must be determined ahead of time and packets fragmented accordingly. This will lighten the load on the routers in between the source and destination because fragmentation would have already been done and packets won't need to be broken up/reassembled. There are other improvments as well but the point I'm trying to make is that IPv6 is a result of years of learning experience with the current IP protocol and is much more than simply solving an address space problem.
--
"What do you want me to do? Whack a guy? Off a guy? Whack off a guy? Cause I'm married."
The catch (of course there is one!) is that you need to accept the axiom of choice, which basically allows you to make arbitrary choices even if those choices are too many to count. The cuts you have to make along the sphere involve choosing an uncountable number of unknown real numbers in each of the three spatial coordinates all at once.
In real life you could not make such choices, since you are constrained to splitting a gold bar along gold atoms, which are discrete units. This lack of applicability of the axiom of choice to real life has led many in the field to reject the axiom of choice as invalid ... but that's a whole other story.
The netmask is a per-computer think. It basically controls the broadcast address your computer will use for that subnet.
.127 will be the broadcast for that subnet, and it will not see the packets in the 128 - 254 subnet (unless there is a n explicity connection). It has to do with logical hiearchies, etc.
:-)
Example: 255.255.255.128
means that 192.168.0.1 through 192.168.1.126 will be valid, and fine. But
If this man is saying he can use Netmasks as extra address bytes, he has clearly pointed his ass at the computer and spewed forth bullshit.
Class C subnets are Of the form net.net.net.node
(and have a netmask of 255.255.255.0)..
There are more defined in the applicable RFCs. Like class A, and B.
The problem with IPv4 was that class C was 255 addresses, class B was 65,025 addresses, and class A was 16,581,375 addresess. If your corporate network had more than 65,000 PCs (possible if you had many servers, and happened to be a huge accounting firm), you basically had to take 16 MILLION addresses away from the global pool because that's how IANNA assigns IPv4 numbers.
Ludicrous! But logical, and in fact proper. This is why IPv6 is good. We *can* piss away IP addresses easily
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
That's a good insight because you were able to relate to the author. I see how the concepts might be confused. I can tell you that the subnet number, even if it were transmitted, cannot be used to augment the address. The only real purpose of having a subnet number is for multicasting to all machines in a subnet. Think of each machine as having two IP addresses, one being the multicast address. If a machine has the address 10.20.30.40 and its subnet number is 255.255.0.0, its multicast address is 10.20.255.255. When it wants to broadcast to all machines on the subnet, it simply sends to 10.20.255.255. All machines on the same subnet will listen.
An example: I like to use class A addresses (10.x.x.x) in my masqueraded network. Within the little network, I set up Samba to communicate with my laptop. Initially, I set the subnet number of the Linux box as 255.0.0.0 while I set the laptop to use subnet 255.255.255.0. Samba has to use multicasting to perform some of its functions. When broadcasting, the Linux box was broadcasting to the address 10.255.255.255 while the laptop was listening for broadcasts on the address 10.0.0.255. Thus Samba did not work.
On the other hand, when I did not understand the subnet number, I set up many computers that should have been 255.255.255.0 as 255.255.0.0. Nothing ever went wrong! The computers were able to browse anywhere on the Internet and log in to the IPX-based Novell network, which was all that seemed to matter.
Does that say that families and businesses can have their own subnets? That's what it sounds like to me, but then again, I'm ignorant. :)
With adequate equipment, an otherwise monolithic candidic legume may be segmented vertically, or horizontally, into smaller, more easily manipulated fragments.
-- What you do today will cost you a day of your life.
...but that is some profoundly lousy writing.
In fact it sucked so much that I was suspicious of it being a genuine IETF draft. I couldn't imagine releaseing to the public a "professional paper" with the horrific language use therein.
Silly me...
My university has a thing called the WEPT (Written English Proficiency Test) that ALL undergrads must pass before receiving a degree. I used to think it was foolish...
This guy would have failed.
License: By reading this you are agreeing that you agree with me.
If I understand it correctly (and I'm not sure that I do, due to the incredibly obfuscated language) he is claiming some expansion of the IPv4 address space by using multiple instances of the same IP address, differentiated by subnet mask.
I gave up after Chapter 3, as my head was starting to hurt.
His mathematics is extremely suspect, both in his calculations and in his apparent amazement that binary and decimal notations do not coincide. Competent mathematicians writing for a technical audience do not generally point this out three times a paragraph.
If someone finds a kernel of truth or reason in this article, please speak up. But don't go in there without your brain firmly strapped in.
Rupert
--
E_NOSIG
Nearly everyone has remarked how extremely awful his writing is, so I won't add to the pile here. People have also noticed his startling revelation that "The distinction [between decimal and binary] is that, this is a Logical expression, that has no Equivalence. [LOL]"
:).
If you actually want to read his paper, just skip to the bottom where he displays his amusing tables . Any ideas what those small numbers in the last column mean (the 1, 10, and 110 ones)?
However, if you were at all like me and dissected his "paper" for what he was really trying to say, you may have actually noticed (if you were successful) that he considers the subnet mask part of the address (look at Table 1 in his "appendix"). Since TCP/IP fundamentally routes a IP datagram around using only the destination IP address, this won't work at all. Datagrams don't keep a subnet mask around with them, they are nodal notions only. His scheme will actually yield several thousand hosts which have the same IP address, which definitely won't work.
Oh and I love : "To render a more pointed fact, I needed to pass a CISCO Certification Examination." That says it all
The opinions I post here have nothing to do with my employer.
His math reminds me of something I saw about 10 years ago - there was some stir in the comp.compression newsgroups over a press release by a company called WEB something-or-other (short for Wider Electronic Bandwidth) anyway, this company claimed they had 'almost perfected' a breakthrough compression algorithm that could losslesly compress any file by a ratio of exactly 16:1. They claimed you could even do this recursively on the output of their compressor, until you reached a size of 'about 1k'. Imagine it! They actually believed that they could take absolutely any n-byte file, and map it one-to-one with some file of m bytes, where m is 1024 or so.
:^)
you could argue, I suppose, that with godlike foreknowledge, you could 'number' all the files humanity will ever produce, and the serial number for any document ever produced could fit into under 1k, but, of course, you decompression tables would be *enormous*. -- oh, and I guess that table would be a file, so it would need a new serial number, and thus a new table, ad infinitum.
As I recall, they even issued press releases announcing they had received VC, and were about to release a product as soon as they figured out how to solve the 'highly unusual situation when four identical numbers are at the corner of a matrix' -- they never explained this cryptic gobbledygook, and never released any details of their scheme.
But the really amazing thing was how many yoyos in the newsgroups bought it, hook, line, and sinker, and spouted nonsense such as: "people thought Galileo was crazy, too, but it turned out he was right! Maybe there are things about your precious number theory that we haven't discovered yet!"
Some poor soul tried to explain that there is no "advanced number theory" involved, just plain counting - there is no way to do a one-to-one mapping from one-byte to 16-bytes. You would think a reasonable person could generalize this principle to understand that you also can't do a one-to-one mapping from 1kb to 16kb, but alas, many pundits wrote back, calling the first guy an idiot for not 'noticing' that the company had 'already admitted' you could only carry out the process until a size of 1k.
The whole thing was pretty funny, but rather pathetic at the same time...
Ghod, that was one of the most unreadable pieces of crap I've seen in a while. I hope it didn't say anything important, I couldn't finish reading it.
The guy needs to go back to grade school and, relearn basic. Rules of English punctuation. He sprinkles commas. At random with, no apparent clue about where periods belong ( to say nothing of the strange spaces around parens ) .
I don't trust his math, either.
IPv6 is coming, anyway. Doesn't almost everything that counts already support it?
-- Alastair
He actually did propose extending the number of bits in IP addresses. The main point of the new subnetting scheme, AFAICT, is to make it easier to add these bits while keeping older addresses valid. However, his new scheme isn't necessary for that (click on "user info" to see my previous response).
You're right, the netmask is not transmitted in the datagram. I've tried very hard to include the diagram from RFC 791 here but Slashdot does not allow me to do so in a legible way :-(
See p.10 of the RFC for the table.
Erik
Has it ever occurred to you that God might be a committee?
Has it ever occurred to you that God might be a committee?
--- Jubal Harshaw
Also, I was surprised to not find an mention of CIDR in the entire document. The IP class system has been obsolete for nearly five years....
--
I always thought that 0.0.0.0 was the loopback address
No, 127.0.0.1 is the loopback address (localhost) 127.0.0.0 is the loopback network, and 0.0.0.0 is default gateway.
$ route -n
Destination Gateway Genmask Flags Metric Ref Use Iface
192.168.20.18 0.0.0.0 255.255.255.255 UH 0 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 1 lo
0.0.0.0 192.168.20.1 0.0.0.0 UG 0 0 138 eth0
Is jwz testing dadadodo again, or did someone else write their own dissociator?
...or perhaps someone was feeding zippy the pinhead and emacs doctor too many RFC's...
...and I have a feeling they never passed that Cisco exam.
pb Reply or e-mail; don't vaguely moderate.
"There yet remains a value in the IPv4 addressing Scheme, which surpasses the promises of IPv6, and could conceivably satisfy our needs indefinitely without an expansion beyond the 32 Bit address range. That is, if it were distributed with country and or state codes as its prefix."