Ask Dr. Vinton Cerf About the Internet

If anyone can claim to have "invented the Internet," (or at least to have co-invented it) it's Vint Cerf, who never makes this claim himself. But he's certainly had a hand in shaping most of what we call "the Internet" today, and is now working on taking the Internet or something like it to Mars and other planets. A Google Search for "Vint Cerf" brings up thousands of responses, so you should have no trouble coming up with a unique, interesting question for him. (As is usual with Slashdot interviews, we'll send 10 of the top-moderated questions to Dr. Cerf about 24 hours after this post, and publish his answers shortly after he gets them back to us.)

Comments

Early days of the net

[El_Smack]

What was it like working with Al Gore?

Re:Early days of the net

[Alomex]

What was it like working with Al Gore?

Actually Vint has publicly commented on this, and (seriously) said that Al Gore as a senator provided crucial support, allow me to quote: "The Internet would not be where it is in the United States without the strong support given to it and related research areas by the Vice President in his current role and in his earlier role as Senator."

No, Al did not invent the internet, but yes, he was a key player back then.

What do you think about Anonymnity?

[Planesdragon]

Although there's a certain moral argument to an individual's right to privacy, there's also a statistical argument that people simply act irresponsibly when given anonymnity.

What's your take on anonymnity in the internent? Is a good thing? A bad thing? Just a thing not worth talking about?

Re:What do you think about Anonymnity?

[Telastyn]

I'd argue that statement is a falicy. When anonymous, or more accurately, faced with the ability to do or say something with no recourse possible just means that people will act while being constrained by only their own moral principles.

The fact that most people are irresponsible, and generally assholes when constrained only by their own moral princples shouldn't be terribly suprising.

Re:What do you think about Anonymnity?

[Twylite]

Actually there is a significant amount of research into the phenomena which sees people lose or relax social and/or moral control when given anonymity.

The opportunity to vent destructive behaviour otherwise unacceptable in society is a coping mechanism employed by some people. It is often done in a (socially) harmless manner - yelling in your own back yard, hitting walls, going to gym - but occaionally people go "over the edge".

Anonymity brings the edge closer: it seems that people are just naturally more destructive when they cannot be held accountable.

At the core of the whole problem is that a society cannot exist with true anonymity. Society requires the ability to identify individuals who are acting against society.

Hindsight

[skywalker107]

Hindsight being 20/20. What is the #1 thing you would change about the internet if you could go back to the early days?

Dan Bricker

DRM?

[GreyWolf3000]

What is your perspective on DRM? Specifically, do you think that the Fritz chip, Palladium, and lobbying of the MPAA/RIAA, will change the Internet fundamentally? Can the Internet be tamed at this point? If so, do you find this DRM and such to infringe upon fair use? Is there legitamacy to the common fear that in the future, computers themselves, in order to gain access to the Internet, will have so many restrictions that the Internet itself will begin to suffer from it?

Better place?

[pubjames]

Do you think the Internet has changed the world? Is it now a better place?

Re:Better place?

[unicron]

Follow up question: Has fire made this a better world, and how do you feel about the wheel and agriculture?

Commercial Email's Early Days

[ekrout]

As vice president of MCI Digital Information Services from 1982-1986, you led the engineering of MCI Mail, the first commercial email service to be connected to the Internet.

As most engineers know, we have to make some sacrifices with every project and get rid of certain features that we had hoped would be there but cannot due to monetary constraints, etc.

Could you explain some of the more difficult decisions you had to make as the head of this particular project? Moreover, was there ever a point in the project where no one thought the final product was viable?

Thanks.

Do you use AOL Instant Messenger?

IP vs. IP?

[Lumpish Scholar]

What do you see happening over the next few years in the battle between the Internet Protocol community (computing/telecom hardware manufacturers, service providers, users) and the Intellectual Property industry (RIAA/MPAA/etc.)?

TCP/IP

[sdjunky]

considering your work with TCP/IP protocols what would you change now that you can look back retrospectively to how it has been used/misused. What would you incorporate into designs now that weren't even thought of at the time that TCP/IP was created?

Negatives of the 'Net

[Dirk Pitt]

Of all the Internet has evolved to be, in what aspect of it are you the most disappointed?

Largest Milestone

[Anonymous Coward]

Since the beginning the net has been ever-evolving by leaps and bounds. What single innovation/technology do you think has had the most profound effect on the net as a whole?

(i.e.: xml, php/asp, etc...)

Question

[CrazyDuke]

What do you think about big media corporations attempting to wrest control of the internet away from the rest of the world?

My question

[andyring]

Mr. Cerf, in light of the copyright battles, DMCA, legal battles, etc., surrounding organizations like RIAA, MPAA, etc., as well as the increasing popularity of broadband and wireless, what do you see the Internet as in five years?

The most surprising thing?

[zero1101]

Of all of the surprising uses that people have invented for the Internet, which surprised you the most (good or bad)?

Beyond Internet

[Ektanoor]

No matter the hype, the pros and cons, the rather primitive, raw and clumsy IP protocol proved its way. And the most fantastic is that its broadcast nature, what some people considered a drawback, proved to be one of its main advantages. We have seen it covering the whole world, proving its ideology on wars (well IP was a DoD protocol for a war situation wasn't it?) and even reaching Mars. However this same primitive, raw and clumsy nature keeps on... And we see lots of troubles on security, performance and reliability. It seems that even Mars is something harder for IP to reach.

Well, is IP protocol The Wheel? And is will this wheel be always a near-round polygon with several holes on it? Isn't any avenue of future for a better protocol? Will we see "ping Mars - timeout, timeout, timeout, timeout - 48 minutes - Mars pinged 80% lost packets" as a common reality?

Disappointment?

[Lumpish Scholar]

The Internet has moved from a research project to a part of mainstream life in less than a decade. Even the "Digital Divide" has turned out to be less of a problem than feared, with most schools and libraries (at least in the U.S.) providing access to anyone who wants it. Pretty impressive.

But what about the development of the Internet has disappointed you? Commercial dominance? Trivialization of the new resource? "Digital Divide"? Security problems? The Microsoft monoculture? The hype of the bubble circa 1999?

Answer

[Daniel Dvorkin]

Cerf has already answered this one. The last two lines are the most telling.

While it is not accurate to say that VP Gore invented Internet, he has played a powerful role in policy terms that has supported its continued growth and application, for which we should be thankful.

We're fortunate to have senior level members of Congress and the Administration who embrace new technology and have the vision to see how it can be put to work for national and global benefit.

It's worth noting that he wrote those words when Clinton was still President and Gore -- you know, the elected President of the United States -- was still VP. Makes me nostalgic for the days when we had an administration that wasn't living in the Dark Ages. [sigh]

IPv6?

[Lumpish Scholar]

IPv6 holds solutions to many of the problems the Internet faces today; but it's still almost exclusively an IPv4 world out there. The usual vicious cycle applies: no one wants to support it until it's widely used, and no one wants to use it until it's widely supported. How, and when, do you see this logjam being broken up?

Your most surprising personal use?

[Lumpish Scholar]

I never thought I'd be able to e-mail my mother. I never though I'd be able to access the public library's "card" catalog from home. I never thought there'd be a more compelling screen than my television set for wasting time.-)

How do you find yourself using the Internet, in ways that would have surprised you a decade ago?

Internet Governace

[cleetus]

The internet, in order to work even at the most basic technical level, needs some standards; some governace. What do you think is the proper scope of that governace/standard setting, who are the constituents, and what are the proper mechanisms for governing?

How do they differ from what we have to day? On the whole, are you optimistic or pessimistic about all this?

Internet vs. Interweb

[bsDaemon]

How do you feel about the proliferation of the "web" and how it has more or less overshadowed "the internet" for the vast majority of the "wired" portion of humanity? Has the amount of frivilous crap that has been allowed to flow over the wires benefitted or people or not, verses if the internet was still just for scientists and students and was restricted to services such as connecting computers for colaberative use and sharing of files that no one is going to get sued over?

filters?

[Triv]

Seeing how there's so much interesting information to be found on the net ('interesting' being good or bad, depending), what do you think about mandatory filtering on public (library, etc) computers? Whose responsability is it to decide what we can and can't see?

Triv

What's next?

[Remus Shepherd]

Not necessarily what you're working on next (although that would be interesting), but what do you think might be the next really big thing? What will be the next technological achievement to affect all of humanity? Are there any projects out there that are still small, like the internet was in the 70's and 80's, but which you believe may mushroom into a world-changing invention?

Advancing to the next phase

[jACL]

In a recent presentation with John Chambers of Cisco, he claimed that streaming media on demand, and therefore, digital rights protection was necessary to grow the Internet into the next phase. Many other people have the idea that the computer and the television should merge before the Internet will "advance."

Others take the Sony approach: the Internet will advance when we can use it as a facilitator -- such as being able to store photos or video from handheld cameras to servers, or access it from cell phones and PDAs for messaging and Bluetooth-type functionality.

Are there other approaches that you've seen (or considered!) for utilization of the Internet that don't head down these two widely-touted avenues?

OSI vs. TCP/IP

[bluestar]

TCP/IP was originally designed as interim solution until OSI could be finished. When do you expect that to happen?

Oh, heck, I'll ask a question.

[Latent IT]

I never ask a question. I want to ask a question.

When you were doing all the initial work, putting things together, and figuring out how things 'should' be, did you ever consider how easy it would become?

I mean, did you ever in your wildest dreams imagine AOL, or something like it? Instant Messaging, Plug and Play, and everything else? To me, back in the good old days (tm) the obfuscation of computer networking was a boon, even in the early '90's. Like Usenet before 1996. I'll admit to enjoying things maybe a bit more when everyone and their grandmother didn't contribute to discussions with one sided opinions in all caps.

So, I guess it's a to part question - did you ever imagine it becoming so easy, and do you wish it had stayed harder?

Distributed Computing

[Loki_1929]

What do you think about Distributed.net and other distributed computing projects that utilize the internet? At any point during your work before the mid-90's, did you ever invision such a concept as distributed computing over a worldwide inter-network being a viable alternative to expensive supercomputers?

Building on that last question, did you at any time consider the possibility of Distributed Denial of Service (DDoS) attacks against a single host on the inter-network, or against the inter-network as a whole? If so, what, if any safeguards did you consider implementing to protect against such problems?

Creators of the Internet

[cperciva]

Another person often dubbed "creator of the internet" was Jon Postel. How would you compare your role with his; and, if you can answer such a loaded question, if the internet had to be invented without one of you, which person (not being involved) would constitute a greater loss?

When the Digital Pearl Harbor happens...

[Rogerborg]

The Secret Service/NSA/FBI/CIA assure us that evil criminal masterminds and cyberterrorists are poised to take down the internet and cripple the global economy at any moment. Given the accuracy of their past predictions, this too will surely come to pass. When it does, the government will need a scapegoat, and fast. I think we know who that will be.

My question is: where do you plan to hide, what psueodonym will you adopt, and will you be travelling in company with Al Gore?

Don't worry, we won't tell them. This is just between you and us.

Which term do you hate more?

[gosand]

Which term do you hate more:

Information Superhighway

Cyberspace

Letter from John Gilmore

[Evro]

Did you ever respond to this message from John Gilmore, which asks why you sided against Karl Auerbach, who (to the best of my knowledge) sought to gain access to ICANN's financial documents? From what I can tell, ICANN's only motivation is to make ICANN more influential (i.e. for its directors to line their own pockets). Given that ICANN is technically a nonprofit organization, this doesn't seem very ethical. Anyhow, the email text is below:

Date: Tue, 19 Mar 2002 14:26:26 -0800
From: John Gilmore
Subject: Re: ICANN: Auerbach's Allegations Off Target
To: vcerf@mci.net, gnu@new.toad.com

> "Karl paints this as a dispute between him and ICANN management, but
> nothing could be further from the truth," noted Board chairman Vint Cerf.
> "ICANN management is merely carrying out its obligation to follow the
> wishes of the Board as a whole rather than follow the dictates of any
> single Director."

Hi, Vint.

I haven't wanted to disrupt our friendship, so I've held off a long time in telling you what I think about how you are leading ICANN. That's why this message is a little longer than it needs to be; I'm saying things that I've been bottling up for a while.

I don't want to be considered a friend of what you now stand for.

You are on the wrong side of this issue, as you have been on the wrong side of many issues regarding ICANN. If ICANN has secrets about who it is doing backdoor favors with, those *should* be made public. And you, as Chairman, as the most prominent and trusted board member, and as the architect of the openness that should still be in the Internet, should have been way ahead of Karl Auerbach in making them public.

Even if those secrets are never made public, or even if there are no terrible secrets inside ICANN, the activities of ICANN MUST be available to every person on the Board of Directors. Without restriction, without delay, without subversion. By law, and for good reasons.

You have been a rubber stamp for many corrupt ideas out of Network Solutions, Verisign and ICANN ever since your election. When I complained to you in the past, such as when the NSI contract was amended to give them a perpetual monopoly, you said that there was nothing else that you could do. I disagreed with that sentiment then, and I disagree with it now. You could have left the contract the way it was, rather than amend it. You don't even have to make things better to keep my respect; you could keep things from getting worse. But you continue to choose to make things worse. Now you are defending ICANN's lack of openness even with its own elected directors!

ICANN was created to promise openness, transparency, accountability, and competition. It has provided none of those, and actively works every month to reduce what little it has provided. You have worked with it to eliminate, rather than create, those promises.

Opening whatever squirming can of worms that is calling the shots at ICANN is what is needed. I can see that ICANN management is terrified that directors from outside the old-boy network might actually find out the details of what ICANN does day by day. They have eliminated any future threat of that, by eliminating outside directors after this term. And they are delaying the current directors' access to information, in the hope that they can permanently avoid outside scrutiny.

I've been a director of several California corporations. I've read that part of the law myself. I've invoked it in a couple of occasions. I contributed significant funding for Karl's lawsuit. Karl is right and you and the ICANN staff are wrong. And now I find you lying about it in a press release. "ICANN management is merely carrying out its obligation to follow the wishes of the Board as a whole..." ICANN *management* instigated those policies, the board didn't. The board has never even considered them.

Virtually everyone at EFF has been looking for ways that we could help to open ICANN and get it to do what it was chartered to do. I've had to hold them back for years, telling them that participation was a waste of our scarce time -- and that no matter how much time they put in, ICANN would have to get really bad before it would ever get better. I put two years of my own life into the domain-name issues, with CORE. It became clear that the strings were being pulled behind the scenes, because the right answers were relatively obvious, yet the wrong answers got approved, providing billions of dollars of benefit to certain parties with heavy ties to the US military. Rather than ICANN making open decisions and using transparent processes, whoever pulls those strings is still controlling what happens. But under ICANN, the process is even murkier and further hidden from public scrutiny. And you're helping.

All the way back at the start of ICANN, EFF and I proposed amendments that would provide a "Bill of Rights" and a "Sunshine Act" and a "Freedom of Information Act" in ICANN's Bylaws. These were all summarily rejected. ICANN does not give a damn about the fundamental rights of citizens or Internet users. It does not want to operate in. the sunshine. And it does not want information about what it's doing to be made available even to its own directors, let alone to the public. Give me one good reason why such an organization should get even a millisecond more of your support -- or anyone's.

The law gives directors an "absolute right" because directors exist to be INDEPENDENT OF and SUPERIOR TO the management. Each and every director has a separate duty to the company; each one carries it out in their own. way. The Board cannot prevent any board member from merely inquiring into the state of the company. The Board cannot condition any board member's inquiry on agreement to a set of arbitrary terms. Nor can the management. This is not only a good idea -- it's the law.

ICANN is going down, one way or another. Either it will go down like East Germany, with a peaceful transition to governance responsive to the public will, or it will go down like Japan, with big bombs dropped on it. ICANN has lost all semblance of credibility and merely seeks to entrench its unaccountable power.

I have absolutely no idea what you are doing leading that megalomaniac, unaccountable, unresponsive, anti-expression, anti-public-interest organization. Did they take your kids hostage? Did you sell your soul for a mess of pottage? What hold do they have over you?

I used to think much better of you than this, Vint. You can see that even now I'm grasping at straws rather than believe that YOU are one of the megalomaniacs. But the evidence continues to pile up, and I'm afraid it's true. I don't want to be the friend of such a person. I'll see you from the other side of the courtroom. Bye.

John

Re:Letter from John Gilmore

[Jim Tyre]

To my knowlege, he never responded; and as a friend of John and one of Karl's lawyers, it's likely that I would know if he had.

OTOH, ICANN tried to use the letter against Karl in the court case. Properly, the court ruled that John's letter could not be attributed to Karl - without regard to whether Karl agreed with what John said.

IP address shortage?

[thogard]

Is the IP address shortage a real technical problem or is it simply a managment issue thats hiding under the excuse that "routers can't cope with large route tables" combined with our current routing infastructure?

Evolutionary vs. Revolutionary Change

[north.coaster]

In the future, so you see the Internet envolving in a evolutionary fashion, or are revolutionary changes in store?

/Don

IPv6?

[Ransak]

We've heard the hype and the 'plans' to move to IPv6 for years now, but the USA seems fairly complacent at IPv4. Do you see IPv6 becoming a reality in the near future (2 to 3 years), and from a high perspective, what do you think (besides the obvious running out of addresses) could spur the movement? Or should we not move at all, and depend on network address translation more?

Voice?

[Lumpish Scholar]

There are a lot of efforts related to Voice over Internet Protocol (VoIP). How do you see them going? Will SIP rule the roost? Is the wireline "plain old telephone service" phone going to be obsolete any time soon, at home, at work, or both? Will VoIP look like part of the Internet to the consumer, or will it be part of the obscure infrastructure?

Al helped build the Intenet

[0x0d0a]

Al (and Clinton) shoveled massive amounts of federal dollars into producing the Internet. If not for their strong pushing of spreading the Internet all over (starting in educational institutions), we wouldn't be anywhere near where we are today.

Yes, Al misspoke. But he was also crucial to the Internet being what it is today, so he gets some points.

Re:Al helped build the Intenet

[richie2000]

Yes, Al misspoke. But he was also crucial to the Internet being what it is today, so he gets some points.

Al Gore's karma: Good (mostly affected by popular vote):

-1, Political Exaggeration
+2, Crucial Contribution

This senator is currently rated +1, Insightful but Boring.

Re:Al helped build the Intenet

[Zeinfeld]

I really don't think Al Gore or anybody in the federal government was all that crucial in "producing" or "building" the Internet after 1993 (start of the Clinton/Gore administration). The Internet was alive and thriving and widespread (in the U.S. anyway) back in the late 80's -- I used it all the time for email via my CompuServe account in 1990, for example.

That would have been when Gore was Senator for Tennessee and lead the committee that gave funding to the NSFnet at that time. Gore was involved with the Internet when it was still the ARPAnet.

Heck, Gore was involved when we were still having problems with AT&T trying to stop us sending packet data over the telephone system because they saw packet data as competition to circuit switching.

In 1990 the email you sent to an 'Internet' would most likely have travelled over the NSF supported backbone. In addition NSF picked up the tab to run the DNS system, IANA and a lot of other infrastructure we needed.

Today of course those services are all supported on a commercial basis but anyone involved in the transition process knew that Gore was calling the shots. The civil service view at the time was that the administration should simply wait for OSI networking to take off. Tom Kalil and Jock Gill spent a lot of time knocking heads together on that one.

Although the Web grew quickly in academia we did not make much impact in the commercial world outside the computing industry until after whitehouse.gov went online. Afterwards it was like someone had turned on a lightswitch.

To be fair there were also Republicans who were very helpful. Newt Gingrich made a lot of enemies setting up the Congressional Web site. However the people who smeared Gore were the same folk who did Newt's political career in.

So...

[0xdeadbeef]

What's up with that WorldCom thing? Did you personally get burned by any of this? Are you ashamed to have worked for those people? Do you think it has it damaged the credibility of the Internet?

And in your opinion, what is it about ICANN that causes people to hate it so vehemently? Is it justified?

third world access

[martin]

We seem to have a world of internet have and have nots.

The biggest set of have nots are still those who have not in respect of anything (the third world). We have the 'ring of fire' around Africa, but that's only really useful for the countries with a shoreline. Do you think your efforts for intra-planet internet-working would help to provide better satellite based access for making ISP's cheaper.

An internet of the people, or for the people?...

[tekrat]

Back when the internet (as we now it) was being developed, it was a government military project.

However, after the internet revolution (of the early 90's) freed it from being Arpa-Net, we had a "golden age" where anyone could connect, and anyone with enough technical know-how could run a server and become a permanent part of the system.

But now we see a day looming in the future where large media conglomerates control it all through draconian service agreements that dis-allow private individuals to run servers in their homes, as well as "linking lawsuits", and patents of obvious business methods, all resulting in an internet where the vast majority of the people can only passively view information rather than interactively take part in providing information.

Do you think it's a "good thing" for everyone to run servers (an internet of the people), or do you believe that it's better for the government and corporations to control the flow of information to citizens (an internet for the people).

While it seems an obvious choice, remember that the situation we have now, where the internet is the "wild west" and mailboxes are littered with spam, and internet rumours become accidental news stories, is a direct result of an internet "of the people".

So there are pros and cons either way. Basically the question boils down to "do you prefer the wild west" versus "do you prefer a controlled, moderated internet?"

IETF and ICANN

[rich_salz]

The IETF is an amazingly transparent organization that has consistently "delivered the goods" with almost no back-room politics. ICANN is its exact opposite, perhaps reaching a nadir when one of its own board members had to sue to see the financial records. Why doesn't ICANN operate in a completely transparent manner? Do you feel the slightest bit uncomfortable with its policies and procedures? Given your background, Welch's comments in the McCarthy Army hearings come to mind.

The internet and spam

[MrIcee]

I'm sure that when you helped to shape the early internet that you didn't anticipate current use, or even the explosion of use.

And, I'm sure you find some issues troubling. I would be interested in your views of SPAM. Did you anticipate it? What do you think about it? And do you have any ideas on how it can be managed or controlled (or, better yet, stopped)?

Biggest promise?

[Jeppe Salvesen]

Mr Cerf,

What do you see as the largest promise of improvement of the Internet? Specifically, what would you like the Internet to be in 20 years?

best regards,

Jeppe

What about NAT?

[swb]

Most people that I run into in the corporate IT world all know/love/use NAT (network address translation). However, as much as NAT conserves IP addresses and provides a measure of inbound-connection security, I've also seen it be the cause many problems because too many sites that have to interconnect are running overlapping IP space. This isn't even counting the number of tools or protocols that have been broken by NAT (even if they're "fixed" in smarter versions of NAT that know layer 3 or 4 protocols; eg traceroute, ftp).

Since the IP protocols were originally built around the idea of unique addresses, I'm wondering if you think NAT has been a beneficial kludge or a curse. Do you think IP should have been had a built-in NAT mechanism allowing for a more protocol-friendly NAT?

Will the (eventual) adoption of the larger address space of IPv6 lead to the elimination of NAT? Should it?

Taming the Spam

[Hanno]

you led the engineering of MCI Mail, the first commercial email service to be connected to the Internet.

On a related note...

Spam is growing out of control and many
administrators now consider SMTP/email to
be broken by design.

Did the problem of unsolicited email, forged
addresses and falsified mail headers ever occur
in the early design of SMTP/email?

What was the opinion on internet abuse and
forgery back in the early days?

Do you think there is a possibility to replace
SMTP with a new design?

Internet censorship

[autopr0n]

How do you feel about internet censorship in places like China, and Saudi Arabia? Recently the Chinese government began knocking (for a short time) people off the internet who did google searches for politically sensitive terms. Do you feel this is morally wrong? Do you think that it has any chance of succeeding?

Workstations and Timeshare Systems

[okl]

In the early days of the Internet (say, pre-1975), when it consisted of the Arpanet and a few linked networks, the goal of ARPA's funding was to tie together users of timeshare computers. In the last quarter of the twentieth century the Internet has morphed into a vast network of LANs, servers, and personal workstations. Timesharing computers and their users have dropped out of the picture.

You and Kahn were doing your early work on TCP in the same years that the first workstations (at Xerox Parc, for example) were being developed. I'd like to know, if you can remember, when you first began to appreciate the magnitude of this change in the internet user base, and whether this change had any affect on your TCP/IP design work in the late 70s.