Ask Internet Visionary and Pioneer Vint Cerf 109
As co-designer of TCP/IP (along with Robert E. Kahn), and former chairman of ICANN, it is no exaggeration to say that Vint Cerf is certainly one of the fathers of the internet, and is often referred to as simply the father. His lifetime of network engineering accomplishments — meriting, among many other laurels, the Turing Award — leaves little doubt as to why he's now a full-time internet visionary for Google (and formerly with WorldCom) as well as a Google VP. Now, Cerf has graciously agreed to answer Slashdot readers' inquiries about the past and future of this little thing called the Internet, and his role in it thus far. This short call for questions is inadequate to sum up his contributions to engineering the data flows that entangle and enlighten us in 2011, but read through a few of these capsule descriptions to get a sense of them. In accord with the interview guidelines, please try not to lump together unrelated questions. (You may find that your questions are moderated downward if they aren't concise; if you have several distinct questions, simply submit separately as many as you'd like.)
Hindsight is 20/20 (Score:5, Insightful)
Re: (Score:2)
The maximum allowed packet size.
At least, that's what I would guess. But there are other things, just look at the differences between IPv4 and IPv6. Of course, allow more than 2^32 IP addresses. Also, the way subnets are organized (/8, /16, /24) is very handy, but did waste huge amounts of address space. IPv6 could easily run short of address space if we're careless with the bits. 128 bits is not much if they're parceled out in ways similar to IPv4.
Re:Hindsight is 20/20 (Score:5, Informative)
IPv6 could easily run short of address space if we're careless with the bits. 128 bits is not much if they're parceled out in ways similar to IPv4.
Nonsense. IPv6 provides enough addresses that each person on earth could have their own IPv4 internet, which would only account for only about .000000000000000005% of the available addresses.
Re: (Score:2)
Re: (Score:2)
http://xkcd.com/865/ [xkcd.com]
Re: (Score:2)
It's not just the blocks for organizations, it's also the addresses reserved for special purposes. IPv4 also uses a few of what could be called "8/". That is, as I'm sure most everyone knows, many addresses ending with .00 were reserved to indicate the whole subnet. Addresses ending in .255 are for broadcasting. Class C networks don't have 256 addresses, they have 254. Also spoils the purity of the address space. Can't make the blanket statement that any set of 4 bytes always represents what should ap
Re: (Score:2)
All this discussion seem to prove is that people don't handle really big numbers well.
Allocations are, for a variety of reasons, assigned such that a subnet is almost always a /64. So the reality is that there are NOT 2**128 addresses. There are really 2**64 ne5works which will always be overwhelmingly sparsely populated. This is not waste. 2**64 is 18*10**18 or (using American terms for big numbers) 18 quintillion or 18,446,744,073,709,551,616 networks.
A /32 is in incredibly tiny bit of this space.
Yes,
Re:Hindsight is 20/20 (Score:4, Interesting)
I think Vint Cerf has already on multiple occasions mentioned the two things he would have liked to have done differently when designing IPv4 and TCPv4. The two things were the size of the addresses and mobility support. At the time there was discussion about the size of the addresses, some people wanted 32 bits, some people wanted 64 bits (and AFAIR some people wanted variable size). Vint was responsible for ending the discussion and deciding on 32 bits. He has publicly admitted that turned out to be an unfortunate decision.
If IPv4 had been designed with 64 bit addresses, chances are we would never have gotten IPv6 with the additional improvements it offers.
Vint Cerf has explained that leaving out mobility from IPv4 was due to it being considered appropriate for a lower level in the stack. And to some extent mobility can be handled by WIFI or cell phone networks. But it turned out that in some cases that isn't sufficient, and mobility at a higher level in the protocol stack would have been better.
I don't think there is anything wrong with the question, but I think that apart from the above two points, what answers you will get is mostly what got changed between IPv4 and IPv6 anyway. Maybe he would also have wanted something designed differently in TCP to begin with (such as knowledge of mobility). The upgrade from IPv4 to IPv6 didn't really change TCP, so TCP still does carry around some stuff that could have benefited from a redesign.
I seem to recall that Vint Cerf also at some point pointed out authenticity as a point that would have been a good idea to have in the protocol from the start. However at the time when IPv4 was designed public key cryptography was still too young to be properly understood, and any authenticity put into the protocol at the time would likely have turned out to be flawed.
Considering how much I can write from recollection of what Vint Cerf has said about that question in the past, maybe it is worthwhile having him repeat it again in his own words.
But there is another question that I would really like to see answered. It is about the IPv4 to IPv6 transition. If IPv4 had been designed differently to begin with, we might not have had to go through this transition. But you never have the necessary experience to design something right the first time, so that is not really a worthwhile discussion. And I don't think discussing the specific design of IPv4 and IPv6 is really that interesting either. The changes from IPv4 to IPv6 all seem for the better, and with the requirement for larger addresses they couldn't have been made much more compatible anyway, so some transition mechanism would be required.
My question is: What could have been done differently to ease the transition. In other words, if you could go back to the point where the design of the IPv6 protocol as it looks today was available, how would you have designed the transition plan? We all know that there was a transition plan, and nobody executed it fast enough to make it work out. How could the transition mechanisms have been done differently to ensure that the transitioning had happened fast enough to make IPv4 obsolete before the addresses ran out and we had to resort to hacky workarounds like NAT?
first trimester elegance abortion (Score:2)
Mod parent up. In ten years I've never said that. Old dog, new trick. Also, was the mistake thinking we needed to do this too quickly?
Absolutely, there are many questions here unworthy of extended discussion. One doesn't to consult Donald Knuth to master quicksort. There's a huge literature on overdesign, discounting through net-present-value, adoption risk, and the diff
Great expectations (Score:4, Informative)
Can you talk about any time when you felt that the direction of Internet development was not going in the way that you hoped it would?
A Simple Pogonological Question (Score:4, Funny)
Best and Worst of Communications Protocols? (Score:5, Insightful)
Re: (Score:2)
I'm wagering you've studied many communications protocols -- is there any protocols that you feel was terribly designed and implemented? Any modern day elegant/simple/innovative protocols that you've admired?
Extremely closely related to above question, Radia Perlman's Interconnections book, thumbs up or down, whats Vint Cerfs review?
Anything like Radia's book but a decade newer?
Have You Ever Met Al Gore ... (Score:1, Offtopic)
What would you like to see developed next? (Score:4, Insightful)
I'm curious what technologies you would like to see developed next, or what you think would be most important to develop next. In other words, what do you think researchers should work on now that would be most significant?
Oh, and thank you for changing my life!
Re: (Score:2)
He has recently between doing a lot of work on the interplanetary internet, you will find this is next big thing.
IPV6 (Score:4, Interesting)
Re: (Score:2)
Or, better question, why wasn't IPv4 ever designed to be extensible? Are we ever going to learn that upper bounds are problematic if they are hard coded? Things that seem improbable now, are likely to become reality later, from 640K, to Fat16/32 to NTFS's 2 TB boot drive limit to 3.64TB Ram to ... the impending doom of Unix epoch time in 25 years (or so).
Yet, we always seem to route around those issues, often with difficulty, when they arise. But wouldn't it be easier to have the solution to running into up
Re:IPV6 (Score:4, Insightful)
Engineering for an unpredictable future just means you waste a huge amount of time and often it means your idea won't get off the ground at all. The internet wasn't ever expected to get as big as it is, because it was essentially a research network. Perhaps analogously, our phone numbering system isn't designed to allow direct-dialing nearby galaxies.
2^32 was - and is - a huge number. 4 billion addresses was unthinkably high, when there were only a few thousand machines who could even use one. It was more than sufficient until a majority of the world needed their own address, or several.
And it wasn't an arbitrary number, either. It's 32 bits, or 4 bytes. Hardware at the time couldn't easily handle addresses larger than that, so if we'd started out with 128-bit addresses, nobody would ever use it because it would be impossible to implement, or far too slow. Hardware has gotten faster/cheaper/better and now it's no longer an issue. So now we're doing it.
Re: (Score:2)
THAT is not my question. And unpredictable my ass, it was huge at the time, but should have been easily seen. Hell, we make fun of 640K ought to be enough for anyone mythos, so why not 32 bits ought to be enough for anyone.
And unlike the "floating point" comment below, this isn't a floating point problem either. It is "wow, 32 bits is huge, but how do we manage going to 64 bits later? And 128 Bits after that?" Having a rough idea would have given time to figure out the larger problems, and provide transitio
Re: (Score:3)
You're missing the point.
Perhaps analogously, our phone numbering system isn't designed to allow direct-dialing nearby galaxies
"It should have been easily seen" is a hindsight argument. Were you there? Did you see it coming? Can you provide a compelling case that they had any suspicion that everybody and their dog in the middle of Africa would need an IP? You couldn't even fit a computer on an average desktop, let alone your pocket.
You also neglect the added hardship of managing the extra bits. Keeping with the analogy, imagine a 25 digit phone number - it's 'easy to see' that we might need one some day, i
Re: (Score:3)
You also neglect the added hardship of managing the extra bits. Keeping with the analogy, imagine a 25 digit phone number - it's 'easy to see' that we might need one some day, if intergalactic telephony takes off and we merge our phone system with the phone systems of a few alien species. Should we have done this back in the 60s when direct-dial came around, because it'll be a hassle to change when it's a problem?
No hassle, the direct dial phone system is infinitely flexible.
From the UK
I can dial my local takeaway with "654433"
I can dial my a takeway on the other side of the UK with "01768 654433"
I can dial a takeaway in Libya by dialing "00 218 21 654 433"
If we take an unused country code (say 990) for
I can dial a takeaway on Mars, in the country of "New UK", by dialing "00 990 40 44 1768 654433"
I can dial a takeaway in Andromeda by dialing "00 990 9 939 483 343 342 459 1768 654433"
00 (international access code)
990
Re: (Score:2)
No it isn't. Everytime we have hard coded limits in computers, from single bit registers to two digit year codes to whatever it has bit us in the ass in broken systems.
That isn't to say that we can't make intelligent decisions along the way, and admit that we are far too short sighted far too often.
Re: (Score:2)
I believe Vint Cerf's answer to this is that 32-bit or 128-bit were both options, so he pressed ahead with 32-bit for the sake of the experiment with the intention of changing things for the "final" version of TCP/IP, but TCP/IP just slipped into usage in its experimental form and it became too late!
Re:IPV6 (Score:4, Interesting)
Or, better question, why wasn't IPv4 ever designed to be extensible? Are we ever going to learn that upper bounds are problematic if they are hard coded? Things that seem improbable now, are likely to become reality later, from 640K, to Fat16/32 to NTFS's 2 TB boot drive limit to 3.64TB Ram to ... the impending doom of Unix epoch time in 25 years (or so). ... But wouldn't it be easier to have the solution to running into upper bounds built into the spec before we implement it in the first place?
Awesome. "Mr Vint Cerf my question to you is, are you aware of any routing protocols implemented using floating point endpoint addressing, and if so how are rounding errors dealt with?"
Re: (Score:2)
Don't be ridiculous. Having variable-length addresses != floating-point.
Re: (Score:2)
Re: (Score:2)
A reasonable question. The problem is that if you leave unused bits in your protocol that can be used to flag extended capabilities, various digital geniuses will use them for all sorts of things. Then, years later, when you need to use them for a real purpose like flagging an extended address field, you will find routers and devices dropping like flies. You'll have analogous problems with any other expansion technique you try to use.
Re:IPV6 and a related question (Score:2)
Interplanetary Internet (Score:5, Interesting)
Re: (Score:2)
The Interplanetary Internet has been adapted for more general purposes as DTN (Delay-Tolerant Networks). There are plenty of applications on besides space, such as acoustic networks - you can basically send out bits with sonar pulses. Used a lot in e.g. underwater sensor networks in oceans (monitoring currents and whatnot). They send data to buoys on the surface as sound and the buoys then send it onwards with radio. However, since the power is provided by solar panels, there's only enough power for the tra
Was college worth it? (Score:2, Interesting)
What impact did your college experience have on you? Do you feel it set the foundation for your future or not?
Privacy and the Internet (Score:2, Insightful)
What is you opinion regarding Internet privacy?
Do you think anything should be changed?
how do we get to 1000Gps in the home? (Score:3)
Future of the Internet (Score:4, Interesting)
Do you think governments and corporations world-wide will be able to kill the Internet as we know it?
Re: (Score:2)
Do you think governments and corporations world-wide will be able to kill the Internet as we know it?
This gets at DNS - TCP/IP was supposedly designed to withstand a nuclear attack. DNS can't even survive a court order.
Do things like P2PDNS and Namecoin have an inevitable future?
What's next? (Score:2, Interesting)
Could you see a protocol ever supplanting TCP/IP?
Postel and Crocker (Score:5, Interesting)
So you went to high school with Postel and Crocker according to wikipedia? Did you guys hang out all along or meet up decades later?
Neutrality (Score:2)
ARIN ip address exhaustion (Score:5, Interesting)
So you're on the board at ARIN. Anything public you want to say about how ARIN is handling ip address exhaustion other than the "company line"?
ipv6 NAT (Score:2)
So are you keeping current with IPv6 and if so what is your opinion of IPv6 NAT? Good / evil / other?
Future of the Internet (Score:5, Interesting)
My question: Do you feel the security concerns over collected information will trump the leveraging of information in future Internet technologies? Will there be a separate "opt-in" or "opt-out" web to cater to each preference?
Context: There have been many controversies recently regarding the collection of data and the privacy of individual information. As we move forward, I've heard a mixed set of messages regarding the direction we should expect to see.
Consumerism is indeed driving innovation and everything is going mobile these days (there's an app for that I think). One example I heard recently of the benefit of the convergence of information and mobility: a consumer can point their mobile phone at a shelf of groceries, get an active "overlay" of information regarding the products and determine which best suits the customer needs. On the flip side, sensors that track customer behavior are installed at the grocery shelf and based on detected behavior (like stopping for a moment to reminisce about Coco-Puffs even though you know they are bad for you) initiates a coupon for whatever the vendor may feel would provide enough motivation to purchase their product -- in the example a $1 off coupon to the mobile phone of a shopper.
Will this become reality in the future? I think there are benefits to be had, but also am fiercely protective of my personal information and preferences.
SMTP, DNS, US Customs (Score:5, Interesting)
It seems that it is getting more and more difficult to successfully run your own SMTP server. See, for example, this post [slashdot.org] responding to the idea that a user was going to move off gmail to their own server. Are there any prospects for meaningful SMTP reform that would lower the barrier to entry for legitimate emailers?
DNS has been often criticized as a centralized single point of failure / censorship. Have you been following the development of namecoin and P2P DNS? Are these systems viable in your estimation? How would you improve them or encourage their adoption?
The US Customs department recently created headlines in seizing domains. These seizures appear to be extra-legal (not founded in law), but ICANN has gone along with them. Are those fair statements? Should ICANN's trustworthiness be suspect as a result of this process?
Thanks and cheers.
-molo
Re: (Score:2)
It seems that it is getting more and more difficult to successfully run your own SMTP server. See, for example, this post [slashdot.org] responding to the idea that a user was going to move off gmail to their own server.
Note that due to spammers running an outbound is indeed a freaking nightmare. Inbound is a breeze. Just smarthost your outgoing thru your local ISP's relay. Its not that hard.
I like running my own inbound, because I have fetchmail all tuned up to funnel multiple pop/imap/whatever accounts into my "real" account. Also they all get the same spamassassin treatment, etc.
Re: (Score:2)
There are a few problems with using your ISP as your outbound relay: You are now dependent on their SMTP and DNS service. Both for performance and availability. Part of the appeal of running your own SMTP server is that you can operate independently.
For inbound, port 25 has to be unblocked, which eliminates many residential connections. This is less of a big issue, but not ideal.
-molo
32 bit as numbers (Score:4, Informative)
Two extremely closely related questions:
1) Conversion from 16 bit to 32 bit BGP AS numbers half a decade ago or so: Went smoother or rougher than you personally expected? Or just right?
2) How does the answer to #1 above modify your view of whats likely to happen with the ipv4 to ipv6 transition?
IPv6 once again... (Score:4, Interesting)
Re: (Score:2)
Should IPv6 have been designed to be interoperable with IPv4?
Isn't it already?
Net Neutrality (Score:2)
hardware accelerated ipv6 (Score:2)
Hardware accelerated ipv4 routing/switching was out there, I donno, at least a decade ago, or more. Your expectations on the rollout of hardware accelerated ipv6 switching?
Was that supposed to be you in the Matrix? (Score:2)
Sorry if this brings back traumatic memories but was "The Architect" in "The Matrix Reloaded" (the second Matrix film) supposed to be you? Did the producers get your permission? Did you like your portrayal?
(Or maybe that was "The Colonel" from KFC after he went on a diet.)
Would you have come up with TCP/IP today. (Score:4, Interesting)
The key advantage of TCP/IP is how it handles for Loss packets, going across an unknown network and far more failure prone hardware. However today as the internet is now running on much more reliable hardware and the path goes threw some well maintained backbone. Would you have come with TCP/IP today if you had access to modern technology/infrastructure?
Or do you think you would have a different design all together?
Interplanetary TCP/IP (Score:1)
Dr. Cerf,
I had the pleasure of working with you years ago at MCI. I recall that at the time, you spoke frequently of the need for an "Interplanetary TCP/IP" standard leading to development of an "Interplanetary Internet." With the recent budget cuts for U.S. manned missions, do you see this effort becoming more or less imperative? On a related topic, do you believe humanity will need to rely more on autonomous robots for space exploration? Or will Americans have Chinese and Russian space-chauffeurs for th
Joining our moon to terrestrial internet (Score:2)
In the future, hopefully in the not too distant future, we will begin to utilise the moon.
Do you think that it will be possible to extend the terrestrial internet to encompass the moon? Would you envisage if that happened that it would be using an extended version of TCP? Or do you think that there would have to be two separate internets one on each body?
Slow start is not slow! (Score:1)
Can you elaborate how TCP "slow start" got its name.
Since the congestion window is exponentially opened, it's not exactly slow.
I mean, congestion avoidance, with additive increase, that's what I would call slow in comparison.
Encryption (Score:2)
Re: (Score:2)
It is already available.
As an example you can do opportunistic encryption (IPSEC) with DNSSEC.
what are you working on?? (Score:2)
T-Shirt (Score:1)
Freedom (Score:2)
The Internet has provided an excellent medium for freely expressing different viewpoints. But governments and businesses are increasingly threatened by such freedoms, and doing a lot to suppress them. How do you see this playing out, and how can we ensure that we keep such freedoms?
.here TLD (Score:4, Interesting)
Do you think there should be a .here TLD, reserved officially for local use in an analogous way to the way that the RFC1918 IP addresses are reserved officially for private use?
Currently many are coming up with their own adhoc TLDs for local use. In my opinion this is suboptimal. Having a standard official TLD would allow more interesting things to "organically grow" on it.
See also: http://tools.ietf.org/html/draft-yeoh-tldhere-01 [ietf.org]
Could IPv6 have been designed better? (Score:2)
Could IPv6 have been designed to be more like IPv4 and thus easier to adopt? In retrospect, would it have been better to restrict design changes to the minimum required to support larger addresses?
What can we do to get ISPs to switch on IPv6? (Score:3)
6lowpan: the ip-domestics/embedded standard? (Score:1)
Statutory Internet (Score:3)
Hello Vint
I've enjoyed since com-priv days, the rise of the Internet - Thank you. 20/20 hindsight shows the Internet grew without statutory jurispudence. Is that statutory blackhole in which the Internet now lives responsible for holding back growth and development of Internet economies, digital cash and cyber laws that are truly global w/o jurisdictional boundary?
If a global medium is not able to support global economy, however does the world escape market constraints bounded by juridictional rules, regs, statutes and accords to enable freedom of exchange, trade, commerce and rising World-wide economic tide?
RexRiley
Why the colon in IPv6? (Score:2)
What's it like... (Score:2)
Has the Internet become too centralised? (Score:1)
Censorship? (Score:1)
Dress (Score:3)
Few human's can attributed to be both communications protocol inventers and well.. dapper.
Can you speak to the importance of being well dressed, groomed, etc. when interacting with non-technical people? Do you attribute your stylistic dress to your overall success in anyway?
Ooh! Settle An Argument For Me! (Score:3)
Moreover, every C standard library that I have ever tried is able to resolve this bigger number to the correct address. If I ping a 10 digit number in that address range, the C standard library will figure it out. It is my position that this is a feature and not a bug.
It seems that the OSX Firefox Guys [mozilla.org] don't agree with me. Admittedly they do have an RFC on the subject, but their browser breaks a known behavior that every other TCP/IP client program on the planet exhibits, including other operating system versions of Firefox!
Would you kindly bludgeon one of us into submission? I don't really care which side of the argument you come down on, but one of us has to be able to say "Because Vint Cerf said so!"
Oh, and while I've got you, I'm sick of writing stateless http applications. May I have your permission to go back to writing plain old socket servers on other ports, providing data based on whatever query format I feel like implementing? It kind of looks like REST, I suppose, except that I don't have to load 14 layers of frameworks to get to that point.
Beginning of the internet (Score:1)
Concurrent Multi-path and Multi-streaming (Score:1)
TCP port 443 is the new waist of the Internet, and it doesn't look like that's going to change with the transition to IPv6 either. Should we just forget about concurrent multi-path and multi-streaming at the transport layer and do it all at the application layer? Or do you think there might still be room for fixing these problems at the transport layer?
Interplanetary Communication (Score:1)
The IP of TCP/IP (Score:1)
Smart Grid (Score:3)
You're currently on the Governing Board [nist.gov] of the NIST Smart Grid Interoperability Panel. What is the state of standards development, and how big an impact does it have to move national infrastructure communications into the public IP arena so far as our ability to strengthen and expand our infrastructure? Conversely, how big are the threats in this new world?
No more "peace and love" in software designs (Score:1)
I take it that the "route around failures" and other original design features of TCP/IP and the Internet as a whole relied upon trusting others always having good intentions and cooperating. Those designs were necessary at the time and the reason the internet exists today.
Nowadays distrust, firewalls, and coding defensively is the norm (or it should be). In that light, the internet's design seems creaky and vulnerable.
Do you have any thoughts or feelings on how software has changed and seemingly become so
How can we bring trust back to the internet? (Score:3)
One of the secrets of the internet's massive success is the lack of controls over it; if there had been strict security and processes in place it would likely not have come about. One of the downsides is that all our security measures are tacked-on, there is no built-in security to the protocols used on the internet and as a result security is a massive problem. How do we go from the wild west to having at least a reasonable level of trusted computing?