Technology Review - Published By MIT
Log in to My.TechnologyReview.com | Register
Advertisement
[1] 2 Next »

Wednesday, December 21, 2005

The Internet Is Broken -- Part 3

Researchers are working to make the Internet smarter -- but that could make it even slower, warn experts like Google's Vinton Cerf.

By David Talbot
smaller text tool iconmedium text tool iconlarger text tool icon

This article -- the cover story in Technology Review's December-January print issue -- was divided into three parts for presentation online. This is part 3; part 1 appeared on December 19 and part 2 on December 20.

In part 1, we argued (with the help of one of the Internet's "elder statesmen," MIT's David D. Clark) that the Internet has become a vast patchwork of firewalls, antispam programs, and software add-ons, with no overall security plan. Part 2 dealt with how we might design a far-reaching new Web architecture, with, for instance, software that detects and reports emerging problems and authenticates users. In this third part, we examine differing views on how to deal with weaknesses in the Internet, ranging from an effort at the National Science Foundation to launch a $300 million research program on future Internet architectures to concerns that "smarter" networks will be more complicated and therefore error-prone.

The Devil We Know
It's worth remembering that despite all of its flaws, all of its architectural kluginess and insecurity and the costs associated with patching it, the Internet still gets the job done. Any effort to implement a better version faces enormous practical problems: all Internet service providers would have to agree to change all their routers and software, and someone would have to foot the bill, which will likely come to many billions of dollars. But NSF isn't proposing to abandon the old network or to forcibly impose something new on the world. Rather, it essentially wants to build a better mousetrap, show that it's better, and allow a changeover to take place in response to user demand.

To that end, the NSF effort envisions the construction of a sprawling infrastructure that could cost approximately $300 million. It would include research labs across the United States and perhaps link with research efforts abroad, where new architectures can be given a full workout. With a high-speed optical backbone and smart routers, this test bed would be far more elaborate and representative than the smaller, more limited test beds in use today. The idea is that new architectures would be battle tested with real-world Internet traffic. "You hope that provides enough value added that people are slowly and selectively willing to switch, and maybe it gets enough traction that people will switch over," Parulkar says. But he acknowledges, "Ten years from now, how things play out is anyone's guess. It could be a parallel infrastructure that people could use for selective applications."

[Click here to view graphic representations of David D. Clark’s four goals for a new Internet architecture.]

Still, skeptics claim that a smarter network could be even more complicated and thus failure-prone than the original bare-bones Internet. Conventional wisdom holds that the network should remain dumb, but that the smart devices at its ends should become smarter. "I'm not happy with the current state of affairs. I'm not happy with spam; I'm not happy with the amount of vulnerability to various forms of attack," says Vinton Cerf, one of the inventors of the Internet's basic protocols, who recently joined Google with a job title created just for him: chief Internet evangelist. "I do want to distinguish that the primary vectors causing a lot of trouble are penetrating holes in operating systems. It's more like the operating systems don't protect themselves very well. An argument could be made, 'Why does the network have to do that?'"

According to Cerf, the more you ask the network to examine data -- to authenticate a person's identity, say, or search for viruses -- the less efficiently it will move the data around. "It's really hard to have a network-level thing do this stuff, which means you have to assemble the packets into something bigger and thus violate all the protocols," Cerf says. "That takes a heck of a lot of resources." Still, Cerf sees value in the new NSF initiative. "If Dave Clark...sees some notions and ideas that would be dramatically better than what we have, I think that's important and healthy," Cerf says. "I sort of wonder about something, though. The collapse of the Net, or a major security disaster, has been predicted for a decade now." And of course no such disaster has occurred -- at least not by the time this issue of Technology Review went to press.
[1] 2 Next »

Comments

  • Da!
    Guest (John Hammond) on 12/21/2005 at 8:47 AM
    Posts:
    1
    I do notknow anything I did not know last week SPAM is a problem.
    Rate this comment: 12345
  • Da!
    Guest (John Hammond) on 12/21/2005 at 8:47 AM
    Posts:
    1
    I do notknow anything I did not know last week SPAM is a problem.
    Rate this comment: 12345
  • slowing down of Internet
    Guest (H.M. Hubey) on 12/21/2005 at 10:56 AM
    Posts:
    1
    Long shift registers (multiple streams if needed for speed) at routers to catch worms, viruses, Trojan horses, etc will not slow down the Internet. The bits will be XORed as they speed along at their normal speed. The other end of the XOR will be registers that can be loaded with bit-images of unwanted pgms (e.g. viruses, etc). It will be a combination of HW and SW. Since it will be expensive, it will be best to implement at the routers. If the routers &quotsurrounding&quot a country known for spamming can catch these, it will be harder  for this kind of SW to spread all over the Internet. In effect, one can quarantine a country so that spam and viruses do not infect the rest of the Internet.
    Rate this comment: 12345
    • new internet?
      Guest (Erik Karl Sorgatz) on 12/21/2005 at 1:02 PM
      Posts:
      1
      If all the spam and cookies, virus and worm code were cut, wed have 50% more bandwidth! Then a little blacklist to keep the spammers from gaining access after a 3rd strike and we might find that the existing internet is fairly responsive. Tax it? Nah..regulate it? Yes, perhaps put all the porn garbage on its own backbone, with its own domain, and start fresh..it might be a good idea if the college kids were only allowed read-only access to USENET for the first six months too. The commercial interests should be blocking the known spam-friendly domains, and the pill-vendors could be held responsible for their commercial spams too - its a slippery slope, but the end user shouldnt be required to support the scum that perpetrate scams and spam.
      Rate this comment: 12345
    • new internet?
      Guest (Erik Karl Sorgatz) on 12/21/2005 at 1:02 PM
      Posts:
      1
      If all the spam and cookies, virus and worm code were cut, wed have 50% more bandwidth! Then a little blacklist to keep the spammers from gaining access after a 3rd strike and we might find that the existing internet is fairly responsive. Tax it? Nah..regulate it? Yes, perhaps put all the porn garbage on its own backbone, with its own domain, and start fresh..it might be a good idea if the college kids were only allowed read-only access to USENET for the first six months too. The commercial interests should be blocking the known spam-friendly domains, and the pill-vendors could be held responsible for their commercial spams too - its a slippery slope, but the end user shouldnt be required to support the scum that perpetrate scams and spam.
      Rate this comment: 12345
    • Long shift registers in routers
      Guest (Jesse) on 12/27/2005 at 5:53 PM
      Posts:
      1
      Will not work.

      1. You dont always have access to the contents. (encrypted)
      2. You dont always have access to the entire message (incomplete messages)
      3. You dont even necessarily have access to the entire packet (out of order fragmentation delivery)

      Check the Security Focus web site, and read the white paper on router hacking...

      You just CANNOT validate the contents at routers.
      Rate this comment: 12345
    • Long shift registers in routers
      Guest (Jesse) on 12/27/2005 at 5:53 PM
      Posts:
      1
      Will not work.

      1. You dont always have access to the contents. (encrypted)
      2. You dont always have access to the entire message (incomplete messages)
      3. You dont even necessarily have access to the entire packet (out of order fragmentation delivery)

      Check the Security Focus web site, and read the white paper on router hacking...

      You just CANNOT validate the contents at routers.
      Rate this comment: 12345
  • slowing down of Internet
    Guest (H.M. Hubey) on 12/21/2005 at 10:56 AM
    Posts:
    1
    Long shift registers (multiple streams if needed for speed) at routers to catch worms, viruses, Trojan horses, etc will not slow down the Internet. The bits will be XORed as they speed along at their normal speed. The other end of the XOR will be registers that can be loaded with bit-images of unwanted pgms (e.g. viruses, etc). It will be a combination of HW and SW. Since it will be expensive, it will be best to implement at the routers. If the routers &quotsurrounding&quot a country known for spamming can catch these, it will be harder  for this kind of SW to spread all over the Internet. In effect, one can quarantine a country so that spam and viruses do not infect the rest of the Internet.
    Rate this comment: 12345
  • The Internet is Broken
    Guest (Grant Callaghan) on 12/21/2005 at 11:09 AM
    Posts:
    1
    Its all software -- even the hardware -- and the only question seems to be, &quotWhere do we put the fixes?&quot  I think they belong at the end of the process rather than at the beginning or in the middle. 

    Charging a small amount per message would cut down on the spam, say a fraction of a penny, and it would generate enough money to police the system, free up bandwidth and catch bad hackers simply because the volume of traffic is so large.

    The only danger I see to this is that the government tends to want to feed its cash cows with ever larger increases in taxation of any kind.  If you let them start taxing the internet, there will be no end to it.
    Rate this comment: 12345
    • Encryption?
      Guest (Aaron) on 12/21/2005 at 12:52 PM
      Posts:
      1
      I think it is odd that an article about the future of the internet makes no mention of encryption. Public key encryption, the ability to know who is saying what, has existed for longer than I have been alive.

      It also seems that a lot of the original ideas that made the internet popular, decentralization and anonymous communication, are lost on its current inhabitants. My mother could care less that emails from me are signed, she just wants less spam in her mailbox.
      Rate this comment: 12345
    • Encryption?
      Guest (Aaron) on 12/21/2005 at 12:52 PM
      Posts:
      1
      I think it is odd that an article about the future of the internet makes no mention of encryption. Public key encryption, the ability to know who is saying what, has existed for longer than I have been alive.

      It also seems that a lot of the original ideas that made the internet popular, decentralization and anonymous communication, are lost on its current inhabitants. My mother could care less that emails from me are signed, she just wants less spam in her mailbox.
      Rate this comment: 12345
    • Interesting idea about access charges
      Guest (Dmitry Afanasiev) on 12/26/2005 at 6:34 AM
      Posts:
      1
      http://blog.tomevslin.com/2005/01/voip_spam_and_a.html

      Here access means access to user. Obviously, this needs sender authentication, automatic charging or balance verification, and probably some sort of rule-based message cost negotiation (e.g. I want to deliver this message, but only if this costs me less than $xy.z). But it makes a lot of sense since (thanks to Moores law) human time and attention are now the most scarce and expensive resources on the Net
      Rate this comment: 12345
    • Interesting idea about access charges
      Guest (Dmitry Afanasiev) on 12/26/2005 at 6:34 AM
      Posts:
      1
      http://blog.tomevslin.com/2005/01/voip_spam_and_a.html

      Here access means access to user. Obviously, this needs sender authentication, automatic charging or balance verification, and probably some sort of rule-based message cost negotiation (e.g. I want to deliver this message, but only if this costs me less than $xy.z). But it makes a lot of sense since (thanks to Moores law) human time and attention are now the most scarce and expensive resources on the Net
      Rate this comment: 12345
  • The Internet is Broken
    Guest (Grant Callaghan) on 12/21/2005 at 11:09 AM
    Posts:
    1
    Its all software -- even the hardware -- and the only question seems to be, &quotWhere do we put the fixes?&quot  I think they belong at the end of the process rather than at the beginning or in the middle. 

    Charging a small amount per message would cut down on the spam, say a fraction of a penny, and it would generate enough money to police the system, free up bandwidth and catch bad hackers simply because the volume of traffic is so large.

    The only danger I see to this is that the government tends to want to feed its cash cows with ever larger increases in taxation of any kind.  If you let them start taxing the internet, there will be no end to it.
    Rate this comment: 12345
  • Email postage, not so good
    Guest (B. Curtis) on 12/21/2005 at 1:04 PM
    Posts:
    1
    Although it seems simple, Mr. Calaghans concept of a small fee per email is no good in reality. It would equally penalize legitimate mass-email systems (newsletters, discussion lists, etc.) as well as spammers. E.g., there has been talk about sending tsunami warnings to peoples cell phones via email Id hardly want to charge the organization millions of dollars right when theyre trying to save my life. If the postage were optional (the recipient chooses if the sender pays), then youre talking about needing to positively identify both sender and receiver of an email, which amounts to SSL in every home. Some have posited using a difficult puzzle to extract a &quotcost&quot of sending emails even though no real money is involved, the same counter-arguments apply.

    No, postage on email is just one of those fun ideas that just wont work.
    Rate this comment: 12345
  • Email postage, not so good
    Guest (B. Curtis) on 12/21/2005 at 1:04 PM
    Posts:
    1
    Although it seems simple, Mr. Calaghans concept of a small fee per email is no good in reality. It would equally penalize legitimate mass-email systems (newsletters, discussion lists, etc.) as well as spammers. E.g., there has been talk about sending tsunami warnings to peoples cell phones via email Id hardly want to charge the organization millions of dollars right when theyre trying to save my life. If the postage were optional (the recipient chooses if the sender pays), then youre talking about needing to positively identify both sender and receiver of an email, which amounts to SSL in every home. Some have posited using a difficult puzzle to extract a &quotcost&quot of sending emails even though no real money is involved, the same counter-arguments apply.

    No, postage on email is just one of those fun ideas that just wont work.
    Rate this comment: 12345
  • Tariffing email
    Guest (Jim Hayes) on 12/21/2005 at 1:54 PM
    Posts:
    1
    B. Curtis seems to not be aware that postage of about $0.20 per letter and who knows what per catalog does not keep my mailbox at the end of my driveway from getting filled with junk mail on paper, especially in the last month.
    Legit emaillers would gladly pay a penny per email to interested recipients while spammers sending out tens of millions of messages a day to random addresses - many of whom seem to illegally use some of my email addresses as return addresses by the way - would be put to rest.
    By law, 911 calls are toll-free.
    The issue of billing is easy - include 1000 emails per month in an account from an ISP, so only the excess is billed, so few users will even need to be billed.
    BTW, I do know companies who have limited access to the Internet for employees because of overloads of viruses and spam, as well as abuses in downloading inappropriate material - I fired an employee myself for storing his downloaded porn on a company computer.
    Rate this comment: 12345
    • Parallel Internet
      Guest (Khushnood Naqvi) on 12/28/2005 at 3:27 AM
      Posts:
      1
      The idea of having a parallel Internet is good. The parallel Internet can be implemented on the next generation of protocols - all with authentication (through digital certificates) and the like. And will have no spam. Commercial sites would perhaps like to have a presence on the more secure  Internet. Users also wont mind to connect to a different Internet for things like Banking, or any business transaction for that matter. Even if users have to pay a slightly higher amount for that one it will be a success.

      But the only problem, I see with that one is that the Internet in the current form will be abondoned and so become more hazardous for people who continue to rely on this one.
      Rate this comment: 12345
    • Parallel Internet
      Guest (Khushnood Naqvi) on 12/28/2005 at 3:27 AM
      Posts:
      1
      The idea of having a parallel Internet is good. The parallel Internet can be implemented on the next generation of protocols - all with authentication (through digital certificates) and the like. And will have no spam. Commercial sites would perhaps like to have a presence on the more secure  Internet. Users also wont mind to connect to a different Internet for things like Banking, or any business transaction for that matter. Even if users have to pay a slightly higher amount for that one it will be a success.

      But the only problem, I see with that one is that the Internet in the current form will be abondoned and so become more hazardous for people who continue to rely on this one.
      Rate this comment: 12345
  • Tariffing email
    Guest (Jim Hayes) on 12/21/2005 at 1:54 PM
    Posts:
    1
    B. Curtis seems to not be aware that postage of about $0.20 per letter and who knows what per catalog does not keep my mailbox at the end of my driveway from getting filled with junk mail on paper, especially in the last month.
    Legit emaillers would gladly pay a penny per email to interested recipients while spammers sending out tens of millions of messages a day to random addresses - many of whom seem to illegally use some of my email addresses as return addresses by the way - would be put to rest.
    By law, 911 calls are toll-free.
    The issue of billing is easy - include 1000 emails per month in an account from an ISP, so only the excess is billed, so few users will even need to be billed.
    BTW, I do know companies who have limited access to the Internet for employees because of overloads of viruses and spam, as well as abuses in downloading inappropriate material - I fired an employee myself for storing his downloaded porn on a company computer.
    Rate this comment: 12345
  • Press Re-Start button
    Guest (666) on 12/21/2005 at 3:04 PM
    Posts:
    1
    The core problem is that Internet, like its underlying software is becoming legacy and is an institution.

    The problem with all software is that underlying software is hard and unmaintenable instead being soft and flexible.

    This will be rectified by my chosen acronym.
    Rate this comment: 12345
  • Press Re-Start button
    Guest (666) on 12/21/2005 at 3:04 PM
    Posts:
    1
    The core problem is that Internet, like its underlying software is becoming legacy and is an institution.

    The problem with all software is that underlying software is hard and unmaintenable instead being soft and flexible.

    This will be rectified by my chosen acronym.
    Rate this comment: 12345
  • Security vs privacy
    Guest (Jose I. Icaza) on 12/23/2005 at 9:40 PM
    Posts:
    1
    Can we trust a government (NSF) initiative to design a more secure internet that nevertheless makes government and corporate tracking of individual users and their data at least as difficult as the present internet?
    Rate this comment: 12345
  • Security vs privacy
    Guest (Jose I. Icaza) on 12/23/2005 at 9:40 PM
    Posts:
    1
    Can we trust a government (NSF) initiative to design a more secure internet that nevertheless makes government and corporate tracking of individual users and their data at least as difficult as the present internet?
    Rate this comment: 12345
  • A better mousetrap?
    Guest (Bob Benitez) on 01/12/2006 at 12:00 AM
    Posts:
    1
    Build a better mousetrap and someone will grow a better mouse. How much of the assault on the internet comes from countries hostile to the free market system, for example, China? Perhaps if we took a harsher approach and assaulted the perpetrators, electronically of course, we might send a message. This is part of a war we had best recognize or we will lose.
    Rate this comment: 12345
  • To protect one Big but faulty player ..
    Guest (Cornelio Hopmann) on 01/12/2006 at 12:00 AM
    Posts:
    1
    harm all others?
    That's the key issue behind the issued V. Cerf raises.
    The basic concepts of the architecture of Windows-XX are faulty as they mix user-level and system-level functionalities (last example the wmf-hack). The Software-engineering community and Microsoft itself do know it for decades. Yet as the business-model of MS relies on bundling Operating Software and Application-Software -like Office etc.- this flaw is intentionally a part of the business-model. Hence before starting to repair at Internet-level what was misdone by Microsoft -at the expense of public and customer money- Microsoft should be obliged by law and courts    to produce consumer-safe products - as any other producer of technology for mass-consumption like car, freezers or air-conditioners.
    Rate this comment: 12345
  • NetAlter - Alternative of current internet/web
    Guest (NetAlter_Fan_1001) on 08/04/2006 at 12:00 AM
    Posts:
    1
    www.netalter.com
    Google keywords : netalter
    Rate this comment: 12345
  • HORRIFIC VIRUS
    Ko on 03/04/2007 at 8:06 AM
    Posts:
    2
    crippling a major canadian media conglomerate.

    This company needs IT saviours. I can't go into details but I'm trying to put together a team of specialists to help the company get rid of this horrific virus (they say it comes from Romania). CNN was infected with it too.

    Check HERALD TRIBUNE at: http://www.heraldtribune.com/apps/pbcs.dll/article?AID=/20070302/BREAKING/70302006&start=1

    "March 02. 2007 12:08PM - Computer virus hits Herald-Tribune
    By TODD RUGER
    todd.ruger@heraldtribune.com

    SARASOTA — A computer virus spreading through business systems this week has hindered several media outlets, including the Sarasota Herald-Tribune.

    The newspaper’s production equipment was affected Thursday night, forcing the newspaper to print Friday’s editions without several of its local news, sports and editorial pages. The technical problems also caused papers to be delivered late.

    Media reports from across the country show similar problems occurred at a dozen media outlets, including Turner Broadcasting, owner of cable news channel CNN, McClatchy-owned newspapers and the Minneapolis Star-Tribune.

    The culprit is the latest version of an old computer virus, RINBOT.L. It is not specifically targeted to attack media outlets, but it cripples Internet and e-mail communications."

    Contrary to the journalist, I think it does target mainstream media outlets. It looks like a group of hackers are terrorising the MEDIA by trying to shut it down. We don't see that kind of thing on You Tube.

    Anyway, this major media conglomerate is PARALYSED by the virus. IT'S A HUGE PROBLEM.

    Please reply if you know who the best IT virus repairperson is. Just for the record, the media conglomerate has hired the best of the best IT specialists. And it’s still CRIPPLED.

    If the media co wants me on board to help out, it will only be as a go between and to help create a team. I’m not involved in IT, I’m a tv producer. I will also try to get in touch with David D. Clark but am not sure he's the right person for this problem. We sort of need a special team, sort of a swat team, a guerrilla team made of hackers themselves, engineers, internet architects, internet researchers. etc.

    Someone from the company mentioned changing the OS from windows to Mac but somehow I have a feeling it wouldn't solve our problems. Would Linux be a valuable solution?

    Anyway, this is a long comment, I’ll give more details and be less confidential if someone replies to me.

    Thank you,

    Ko (koandco@gmail.com)
    From Montreal and Toronto
    PS: I’ll forward people’s name to IT Director and it’s between you and them. The company is Canadian.
    Rate this comment: 12345
Add New Thread

Related Articles
Advertisement

Current Issue

Technology Review September/October 2008
How Obama Really Did It
Social technology helped bring him to the brink of the presidency.
•  Subscribe
Save 41%
•  Table of Contents
•  MIT News

Magazine Services

•  Gift Subscription
•  Digital Subscription
•  Reprints, Back Issues, Customer Service

Career Resources

Visit the Job Board and Resource Center to move your career to the next level.

MIT Technology Insider

Stories and breaking news from inside MIT about the latest research, innovations, and startups--in a convenient monthly e-newsletter. Subscribe today

Follow us on Twitter

Twitter

Get Technology Review updates via the web, cellphone, or Instant Messager – Follow techreview on Twitter!
Advertisement

More Technology News from Forbes

Wire Stories

Advertisement
Advertisement
Advertisement
TECHNOLOGY RESOURCES