Receive the Bulldog Edition


Economic Blogosphere

Economic Journalists


economicprincipals.com banner

August 6, 2012
David Warsh, Proprietor


| contents |

Who /Really/ Invented the Internet?

“It’s an urban legend that the government launched the Internet.” That was former Wall Street Journal publisher L. Gordon Crovitz, starting an interesting discussion last month on the paper’s editorial pages. (He now writes the “Information Age” column for the paper – subscription required.)  He was, of course, riffing on President Obama’s “you didn’t build that” remark. In fact, he wrote, it was Xerox Corp. that came up with the idea of linking different computer networks together.

Crovitz buttressed his opinion with a quotation from blogger Brian Carnell, which he misattributed in US print editions to economist Tyler Cowen, of George Mason University (later quietly corrected in the WSJ digital archives).   ”The Internet… reaffirms the basic free market critique of big government,” Carnell wrote in 1999. “Here for thirty years the government has an immensely useful protocol for transferring information, TCP/IP, but it languished…  In less than a decade, private concerns have taken that protocol and created one of the most important technological revolutions of the decade.”

That brought a strong letter of dissent from Vinton Cerf and Stephen Wolff, each of whom made key contributions, while working for the US Defense Department in the early 1970s, to the architecture of the astonishing new technology. (Cerf was co-author, with Robert Kahn, of the Transmission Control and Internet Protocol suite (TCP/IP), the foundational standards that permit networks of computer networks to communicate with one another.) The Internet’s development had been a model of collaboration among government agencies, universities and the private sector, they wrote. “Focusing on one element to the exclusion or disparagement of other elements is simplistic, misleading and wrong.”

Crovitz chose to read that as a rebuke to the president. But Cerf and Wolf made it clear that the open architecture they devised was the government’s idea.

Many private sector corporations vigorously resisted the open TCP/IP protocols, preferring instead to support their own proprietary approaches.  Without open protocols, the private sector could not have possibly moved so quickly to commercialize and deploy networks on such a large scale.  It was US government policy that TCP/IP would be an open protocol.

But even after a second Crovitz column, headlined“WeHelpedBuildThat.com,” his readers had no better idea of how the Internet actually happened. Books such as The Dream Machine: J.C.R Licklider and the Revolution that Made Computing Personal, by M. Mitchell Waldrop; and Inventing the Internet, by Janet Abbate, make abundantly clear to anyone who takes the time to read them the centrality of government-funded research to the task of getting computers to communicate with one another. Computing: A Concise History, by Paul Ceruzzi, provides a fine overview of the first seventy- five years of the digital age in fewer than 200 pages.

But the organization that, more than any other, deserves credit for swiftly bringing the technological possibilities to commercial fruition, permitting millions of jobs to be created around the world (and others to be extinguished!), is the Internet Engineering Task Force, the planning body that developed various standards for the Internet as it evolved.

The ITEF hasn’t gotten the ink that it deserves.  Its website is a trove of information for technical readers, and proper narratives probably are on the way. But the only straightforward account I know in the lingua franca of book publishing is Scott Bradner’s essay in Open Sources: Voices from the Open Source Revolution.

The IETF started in 1986 as a quarterly meeting of US government-funded researchers, Bradner says (he was one of its founding members).  Non-government vendors were invited starting later that year.  After that, anybody who wanted to was able to attend. Barely 35 people came to the first few meetings.  By the late ’90s, more than 2,000 would show up (the meeting frequency dropped to three a year.) For the meeting in Vancouver last week, more than 1300 persons registered, organized into various working groups.

People, not organizations, are members; anyone who wants to can join: all its internal documents are freely available on the Web. The trick, since the beginning, has been to be on the cutting edge of developments. This is the organization that, back in the early 1990s, drew up plans for the commercialization of what to that point was a government- operated network.  The creation of the standards opening up the World Wide Web followed in a few short years. Early members note that never before in history has a new technology rolled out so quickly and exuberantly, with so little waste.

The IETF is remarkable for its bottom-up nature; most standards organizations are run from the top down. Take the British Banking Association, the organization that more than twenty years ago, created the standard known as the London Interbank Offer Rate (Libor).  All standards are “recipes of reality”: that is the subtitle that Lawrence Bush, of Michigan State and Lancaster universities, gave to Standards, his excellent (and timely) book on the ubiquity of these procedures, and their multi-faceted significance for our life.  From weights and measures to legal contracts and software protocols to the customs and expectations that form the ethical infrastructure of human communities, standards are “a means of partially ordering people and things so as to produce outcomes desired by someone else,” he says.

Some standards are better-constructed than others. The standards that brought the Internet into existence were exemplary.  The standard that is Libor, written into trillions of dollars of contracts around the world, is something else again.  A fair amount of the squabbling going forward over who was best served by the overall rate-setting procedure and whatever possibilities existed for its abuse: borrowers? lenders?  or, perhaps, the banks themselves?

These are very complicated issues, but they are precisely the ones that standards organizations exist to debate and resolve. Eric Von Hippel, of the Massachusetts Institute of Technology, has studied more closely than anyone else I know the means by which industries, firms, and their customers band together to improve the standards by which products are produced. But improving products is one thing.  Creating whole new industries is another.

There are certain tasks so inherently risky that only government can be expected to undertake them. Getting the computer industry started was one:  governments built the first computers during World War II; and, in the early days of the Cold War, the US Air Force financed IBM Corp.’s entry into the business. Next came the first ventures in time-sharing and packet-switching – the technologies on which the Internet is based.

Meanwhile the National Institutes of Health were financing several different revolutions in health care: molecular biology, imaging techniques, genomics.  Today government is funding research in gas, solar and wind technologies that will be necessary in order to cope with climate change.  That’s why it’s so important to understand the beginnings of the Internet. Mighty oaks from little government acorns sometimes grow.

There’s even a moral here for corporate statesmen. As WSJ  publisher, Crovitz built a successful pay-wall around the paper’s website, and pointed the way to successful newspapering in the future. But if he (and other Dow Jones executives) had paid more attention to where the Internet came from, and how it really worked, they might still be able to call that newspaper their own.  Instead they are working for Rupert Murdoch.

| contents |

8 Comments

  1. Dale Worley wrote:

    Another example of government-sponsored innovation was postal air mail. Air mail was sold at a subsidized price in order to generate an above-market volume of demand for aviation services. Interestingly, the air carriage of mail was contracted out to private airlines, more or less preventing aviation from becoming a noncompetitive bureaucracy. But because the nascent aviation industry saw much higher demand than its cost structure justified, the technologies were developed faster than they would be in a pure market situation. There’s a brief introduction in http://en.wikipedia.org/wiki/Air_mail#The_United_States

    Monday, August 6, 2012 at 2:37 am | Permalink

  2. Dale Worley wrote:

    Being a “member” of the IETF, I have to say that its sociology is unusual, and may well have contributed to its remarkable success. People participate as individuals, and its structure is rather anarchic, so in order to influence the technical outcomes, one must gain the respect of other members. But there is an overall ethos that the Internet must be “open” — particularly, not under control of any government (for political reasons) or company (to extract profit). The result is that the benefits of Internet growth and usage tend to accrue to the end-users rather than governments, equipment vendors, or communications carriers.

    I suspect that the ethos initially developed because the members were young academic researchers, with a certain countercultural spirit of the liberating possibilities of technology. But the ethos is still rigid; the fastest way to “not be listened to” is to be perceived to be directly representing the commercial interests of your employer (who is paying for you to attend).

    Oddly, despite that the US government funded initial TCP/IP development as a research project, during the period 1990-1995, it mandated that federal government networking product acquisitions would include (exclusively or not) the “OSI” protocols. This was met with extreme derision from the Internet technical community, and of course, ultimately failed. (http://en.wikipedia.org/wiki/GOSIP)

    BTW, the Vancouver IETF attendee list link given above seems to be incorrect — the right one is https://www.ietf.org/registration/ietf84/attendance.py

    I hope that someone writes a proper history and sociology of the IETF, but since we’re geeks, it would probably have to be done by an outsider.

    Monday, August 6, 2012 at 2:59 am | Permalink

  3. mulp wrote:

    What does it matter who created the Internet if the prohibition of commercial use of the Internet remained because the Internet was a research project funded by government that was to promote proprietary products from the free market?

    And lots of corporations were building proprietary products using the knowledge reflected in the Internet research network:
    * IBM
    * Digital
    * Microsoft
    * AOL
    * Novell
    * AT&T
    * Sprint
    and many more

    Each had a plan to takeover the world with their proprietary network solution. Bill Gates was backing the proprietary MSN network to beat AOL in 1995 when MSN first rolled out.

    But it was Congress that picked the winner in a one line amendment to a technician training program bill that gave NSF, which had been given the authority over the Internet, the authority to promote the commercial use of the Internet.

    Ironically, this was opposed by a couple of Internet companies who were trying to build a parallel Internet to the government run Internet that had been built by engineers over the previous two decades using government funding.

    The Congress picked the Internet to be the winner, probably without intending to, with one line of law inserted into a bill, that gave the authority to the engineers running the Internet to open the gates wide to the likes of Amazon to make money on the Internet.

    Congress picked the Internet as winner, warts and all, and the Internet still has the big wart of limited address space that was known in 1985 and that made OSI the superior technical standard. Unfortunately, the superior technology did not have government funded users and infrastructure, so once Congress picked the Internet as the commercial winner, the Internet was the winner because of the big head start the government gave it.

    Monday, August 6, 2012 at 3:53 pm | Permalink

  4. This is and economic progress in timeline which is appreciated. As a former student and now a amateur blogger it enhances information which is not always available to extract and present. As a learning tool for networking and a follower of popular technological advancements, I have to admit that my interest was with the OSI Layers and the how it was introduced by our goverment.

    Monday, August 6, 2012 at 4:39 pm | Permalink

  5. admin wrote:

    You mean that Crovitz couldn’t erven get the quotation right?

    Monday, August 6, 2012 at 10:29 pm | Permalink

  6. Jeanjour wrote:

    The author needs to be corrected. All standards organizations I am aware of in the technology area are bottom up: IEEE, ANSI, ISO, ITU, etc. The IETF is unique in being organized so that a small elite can control the agenda. While most standards organizations have the usual bi-cameral organization of a “lower house” of technical experts and an “upper house” of over all management where all stakeholders are represented in both houses. The IETF chose to organize along the lines of a mass subgroups of technical experts (the IETF meeting itself) coordinated by a select group of area coordinators (the IESG). It is more like a Party Congress and a Politburo, than an Western representative democracy. There is nothing in the IETF rules that prevents a large company from stacking a meeting (and they do).
    The success of a standard has nothing to do with technical excellence or good engineering, but only about adoption. (See Bluetooth, probably the worst design of the 1990s). That adoption is purely a decision of self-interest by the companies building product. A successful standard is one that is widely adopted.

    One of the comments above says the government funded development of TCP/IP during the period 1990-1995. This is grossly incorrect. TCP/IP was developed by the government between 1974-1983, when it was deployed in the Internet and with patches made continually ever since. Most notably adding a patch for congestion control in 1986. It is still unclear why it took 9 years to be mature enough to deploy. Few other protocols have taken that long.

    Tuesday, August 7, 2012 at 4:09 am | Permalink

  7. Tyler Cowen wrote:

    I would ask that you correct the text, would that be possible? Thanks…

    Tuesday, August 7, 2012 at 8:56 am | Permalink

  8. admin wrote:

    Done.

    Tuesday, August 7, 2012 at 11:14 am | Permalink

4 Trackbacks/Pingbacks

  1. Links for 08-06-2012 | FavStocks on Monday, August 6, 2012 at 7:24 am

    [...] Who /Really/ Invented the Internet? – David Warsh [...]

  2. A misquotation episode on Monday, August 6, 2012 at 10:22 pm

    [...] DeLong and David Warsh criticize a quotation from me on the origins of the internet.  But in fact the quotation is not [...]

  3. [...] Economic Principals » Blog Archive » Who /Really/ Invented the Internet?: “It’s an urban legend that the government launched the Internet.” That was former Wall Street Journal publisher L. Gordon Crovitz…. In fact, he wrote, it was Xerox Corp. that came up with the idea of linking different computer networks together. Crovitz buttressed his opinion with a quotation from economist Tyler Cowen, of George Mason University: [...]

  4. Saturday links: getting hacked | Abnormal Returns on Saturday, August 11, 2012 at 11:29 am

    [...] More on just who invented the Internet.  (Economic Principals) [...]

Post a Comment

Your email is never published nor shared.
Required fields are marked *

*
*


Skim past columns here.


Support Economic Principals by subscribing to its bulldog edition—receive the weekly via email a day before it is posted on the Web, and, as well, a quarterly Report to Subscribers.

To reach the proprietor, ask a question about the website or report a problem email warsh@economicprincipals.com.