« African private safety nets | Main | Are we moving away from tipping? »
Did government invent the Internet?
Here is a useful corrective:
Back in the mid-1980s the Internet was the sole province of universities and government institutions. Private individuals who just wanted to send e-mail over the Internet would have had a hard time doing so.
But that doesn't mean there weren't vibrant computer networks. In fact there were tens of thousands of Bulletin Board Systems around the country that were relatively cheap to join and offered e-mail, files, discussion forums and a whole host of things that are now largely on the web; although some remnants of this BBS culture still exist.
The main problem with the BBS system was a lack of standards for interconnection. As the 1990s approached and computers became more powerful and modems supported more bandwidth there were several competing proposals for graphical interconnection standards, but those were wiped out by the Internet tsunami.
It is interesting, given [Barbara] Ehrenreich's view that the Internet was an innovation made possible by the government, that prior to the early 1990s almost nobody outside of governments and universities had home access to the Internet while several million had logged on to a BBS at one point or another. What caused the change? Something Ehrenreich and her left/liberal friends usually fight tooth and nail -- privatization. The floodgates of the Internet came open only after key resources became privatized and companies and individuals could operate on the Internet. For much of its existence, commercial activity on the Internet had been forbidden. The removal of that barrier is primarily responsible for the Internet we have today, where both anarchists and Abercrombie and Fitch use the web to broadcast their respective messages.
The Internet, in fact, reaffirms the basic free market critique of large government. Here for 30 years the government had an immensely useful protocol for transferring information, TCP/IP, but it languished with almost no added benefit other than to the military and academia. In less than a decade, private concerns have taken that protocol and created one of the most important technological revolutions of the millennia.
So can we interpret the history of the Internet in largely laissez-fiare terms? Well, uniform standards are useful, the U.S. had better telecommunications policy than most other countries (including a system of managed competition), and local phone calls were set at price zero. All of these helped the Internet, if you have other interventions in mind add them to the comments. It is nonetheless correct that private initiative made the Internet what it is.
Here is the link.
Posted by Tyler Cowen on August 30, 2005 at 06:05 AM in Web/Tech | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c66b253ef00e550978e068834
Listed below are links to weblogs that reference Did government invent the Internet?:
» Your Lunchtime Reading from The Agitator
1) Kerry Howley on how the oppressive state of Myanmar offers more medical and pharmacological freedom than we have here... [Read More]
Tracked on Aug 30, 2005 11:55:34 AM
» who took the initiative to create the internet? from Leviathan Slayer
Over at Marginal Revolution, Tyler Cowen points to an old article by Brian Carnell on who is really responsible for the age of communication [Read More]
Tracked on Aug 30, 2005 8:05:20 PM
» Did Goverment Invent the Internet? from Joe Grossberg
The Internet, with a capital "I", yes (along with goverment-funded universities). But the internet, in the sense of people remotely... [Read More]
Tracked on Sep 28, 2005 3:31:52 PM
» Buy MP3 Music Online - Digital Music Downloads from MP3 Music
MP3 Music Downloads - MP3.com offers links to legal digital music downloads from a wide variety of services. Buy MP3 music online from your favorite artists ... [Read More]
Tracked on May 2, 2006 11:54:48 AM
Comments
Isn't this fairly typical of developments in our mixed economy.
The governemnt makes many of the investments in the very early stages
when it is not clear what the payoff will be and how
long it will take to materalize.
Once the development gets past these early stages where the
the returns are unclear the private sector moves in and
takes the develoment to its final stages.
Posted by: spencer at Aug 30, 2005 8:30:32 AM
The most important development, apart from important "proof of concept" seed financing through federal subsidies was the establishment of url and html standards that were open source (microsoft, it should be recalled, wanted a proprietory solution, which would have significantly slowed the development of the internet). These open standards were the work of a publicly funded research physicist..
Posted by: Roland at Aug 30, 2005 9:12:20 AM
Isn't this kind of a chicken-egg thing? Without the government's initial investment in the research and creation of standards there would not have been a viable universal network in which private entities could invest, but without that investment the internet would be nothing more than a set of standards.
So it hardly follows that "[t]he Internet, in fact, reaffirms the basic free market critique of large government." After all hasn't it literally taken an act of congress to set HDTV standards?
Posted by: djlicious at Aug 30, 2005 9:48:48 AM
It's a good point that BBS infrastructure might have evolved fairly soon into an internet if "the internet" hadn't intervened. But if the internet had stayed in the government / education sector, TCP/IP might still have been picked up and used by the BBSs. The point about TCP/IP is that it was developed for powerful computers - the sort that might have been "the computer" for a university department. It wasn't until the mid 90s that personal computers became powerful enough to implement them effectively. The protocols that BBSs used were much simpler, and could be handled by the home computers of the mid-80s onwards.
And the web is a red herring. That is just one application of the internet, and other applications could have been used in similar ways if it hadn't been introduced (not that it wasn't a well-designed protocol). Gopher was already growing into something that could have performed the same kinds of function in 1992.
Microsoft were only interested in creating something like AOL - which also was less demanding on hardware, but didn't really flourish even without the internet for competition. If the open, extensible network had arrived from the BBS culture, as suggested, it would have eclipsed AOL / Compuserve / MSN in the same way the internet did.
Posted by: Andrew McGuinness at Aug 30, 2005 10:06:29 AM
You really need to look at the timelines of things even beyond the BBSs
AOL and Prodigy were increasing their user base before the Internet became commercial.
Posted by: Xmas at Aug 30, 2005 10:13:33 AM
End-to-end IP commoditized ISPs; incumbent carriers don't like having their products turned into commodities, so AOL, MSN, Compuserve, etc. naturally preferred proprietary protocols which preserved customer lockin and raised barriers to entry. The existence of the NSFnet backbone and NCSA Mosaic had a lot to do with the widespread adoption of IP by commercial BBSs in the early 90s. Without those carrots, what is the market force that would have overcome the incumbents' resistance to standard protocols?
It could as easily be argued that the Internet is a success story of government intervention--it did the research and built the infrastructure to the point where it was capable of supporing private development, and miracously had the sense to privatize it then.
Posted by: Dan at Aug 30, 2005 10:23:27 AM
It should be noted that the research oriented around the Internet wasn't at all done to make The Internet as we know it today.
It was a system brought on from the cold war so that large governmental data centers over the country would still be up even if an entire city was leveled.
So, while it's true that private enterprise helped to "mass produce" the Internet, it's the government-funded research that started it all. (And, today, Bush seems to not understand how important it is economically to fund high-risk research.)
Posted by: Macneil at Aug 30, 2005 10:31:55 AM
To echo Dan, it's one thing to acknowledge that markets often work, but it's another entirely to rewrite history. A good look at the IETF after the net became the Net shows that standards are well nigh impossible to develop once large corporate actors see them as a strategic tool for creating barriers of entry and lock-in. Moreover, to argue that the privatization just "happened" by market forces is also folly. From the NSF coordination of backbones to Postel managing the name system single-handedly, the shift to private ISP's was a managed process that demonstrates technology transfer at its best.
Posted by: allan at Aug 30, 2005 11:31:26 AM
As I understand it a lot of standards are developed without government intervention. Some of the comments here have stated or implied that it is impossible or very difficult for the market to come up with standards, that standards must be established by government or they will never be established at all. I think this is false.
Since IP happened to already be there, and simce some infrastructure happened to be there, there was good reason for private parties to build off of the IP internetwork. Similarly, if people build new roads, chances are they will build them to connect to existing roads. Suppose for a moment that people would have developed an inter-network with a common protocol regardless of what the government did. If that were the case, then given that the government had its IP and the physical equipment in place, then given that the IP was pretty good as it was, people had a good reason to glom onto that. Why reinvent the wheel?
That people use a particular standard is an indication that they need a standard, but is not much of an indication that they need that particular standard.
Posted by: Constant at Aug 30, 2005 11:40:05 AM
The story of adoption of the tech
is complicated by the way that
the private provision of Ma-Bell-ish
services was substantially
illegal. At Nortel some of my
co-workers --- not libertarian, just
cynical --- would say that the core
competence of our customers (i.e., telcos)
was lobbying. Meanwhile, within niches
which weren't covered by various telecom
franchise monopolies (e.g., the wide-area
networks within corporations big enough
to lease lines, or local networks even
in modest-sized companies) there was
substantial adoption of
Internet-ancestral technology.
The story of the development of the tech
is probably a bit complicated as well.
I'm no expert, but I'm skeptical of any
account which doesn't mention Xerox PARC
and the DEC and IBM networking
technologies. PARC comes up in histories
of the technology which don't seem to have
a rah-rah-capitalism axe to grind; when it
does not appear a history, I cynically
suspect rah-rah-planning axe-grinding.
And I don't know how to estimate
the sophistication and level of adoption of
DEC and IBM networking technologies, but
I note that they were prominent
enough that when I was setting up networks
in the mid-80s, bridging to them was an
highly-visible consideration in the
software and documentation.
Perhaps today it might be
technically possible for
a modern city to have many of
its transportation needs served
efficiently by a whizzy
system of jitneys dispatched by cell-phones
calling into sophisticated routing software.
Private enterprise is not providing it.
Perhaps that is because capitalism is broken,
perhaps because entry into the market is
a lobbying problem not a technical
problem. If someday some innovative
city/county transit monopoly decides to start
providing it, I expect we will
hear plenty about how we should
be grateful to central planners for their
innovation. I expect, however, to hear
less about the cost of
the same planners outlawing innovative
competition; the usual comfortable
assumption seems to be implicitly
to approximate such costs as negligible
compared to the benefits of central
control.
Posted by: Bill Newman at Aug 30, 2005 11:56:49 AM
I thought Al Gore invented the internet. ;)
Posted by: Josh at Aug 30, 2005 1:32:38 PM
The general argument in the comments here against Tyler seems to be that the government did indeed manage invent the tcp/ip even if it was the private sector who created the useful applications on top of it. But tcp/ip is pretty useless without the useful applications. Tyler's point is that a tcp/ip type framework would have been invented by the private sector anyway, even if the government hadn't invented tcp/ip.
I don't understand those who argue that government needs to fund high risk investment. High risk means high losses. Its not any different than playing the lottery: high risk, high losses. Sure, every once in a while some one wins the jackpot -- and every once in a while the government invests in worthwhile research and comes up with the tcp/ip -- but for every winner, there is much more money lost on the losers. The costs are clearly more than the benefits. especially when, as Tyler has demonstrated, the market will get around to it soon enough anyway, and do it in a manner where benefits are clearly higher than the costs.
Posted by: eric at Aug 30, 2005 2:01:31 PM
The heart of the what makes the 'net as we know it possible, the Internet Protocol (IP), is a direct result of a government funding. Given that IP is the essence of a public good, (you can think of it as "law" if governning the interaction of computers rather than people if that makes you feel better) this is appropriate. Most of the applications associated with IP use when it was winning the standards battle (FTP, smtp, DNS, BIND, Apache's predecessor etc ) also came from the public sector, typically computer science departments.
Th "left" version of events is far closer to what actually took place in the tech trenches than what Tyler is pushing in the original post. Internet means: that which uses the IP protocol. All sorts of proprietary stuff throught the late 90s was still being pushed by various companies as an alternative to IP (eg Novell's proprietary standard). And it all lost out to the technically superior government-funded version.
The proprietary vendors' loss to the publicly funded research effort is hardly surprising: IP is all about getting previously incompatible systems to talk nice to each other, so the protocol that gets to evolve in the most heterogenous, open environment, with the least legal restrictions on testing and developemnt, has an immense adavantage. ATT and it's ilk could not begin to approach the the collective dieversity and talent of the nation's comp sci departments.
As for the BBSs -- they were the CULTURAL antecedent of today's blogosphere, but technically they lost. Pure and simple. When was the last time anyone used Kermit? How many sysops still use their serial port for networking?
This is not to say networking protocols must evolve in the public domain to be useful. A fine example of private setctor development is Ethernet, which started life as a memo from Bob Metcalfe to his boss at Xerox PARC in 1973, and continues to develop three decades later.
http://inventors.about.com/library/weekly/aa111598.htm
But the US Government, and DARPA in particular, deserves the credit for funding the development of the internet as we know and love.
Posted by: Tylerh at Aug 30, 2005 3:36:23 PM
I want to point out two different arguments which have been mentioned but not, I think, distinguished:
a) The government is needed to establish standards, and TCP/IP is that. The government's most important role is its special ability to establish a standard. Things like TCP/IP, UNIX, and Perl might be created by individuals in all walks of life, both privately and publicly employed, but will not be widely used without government backing them up all the way. The creation of TCP/IP could well have been done privately, but it took government backing to make it into a standard.
b) The government is needed for basic high-risk research, and the development of TCP/IP is that because of the hundreds of millions spent testing TCP/IP on monkeys and then on live human volunteers. The government's most important role is its ability to fund research like the research that produced TCP/IP. Inventions like TCP/IP, UNIX, and Perl cannot be developed in the first place without massive government assistance.
Posted by: Constant at Aug 30, 2005 3:44:44 PM
Isn't everyone forgetting the obvious?
Computers first became cheap in the mid 1990s.
We could have privatized the internet all we wanted in the 1970s and it wouldn't have made a whit of difference. Very few people could afford the hardware, and the technology was still so immature in the late 80s that it was a painful experience trying to do anything useful. Ever sit around hours waiting for a 300 baud modem to download a smallish program from a BBS only to find that the program didn't even work in the end? I did. It wasn't much fun.
Separate the influence of maturing technology from the effects of privation and you will have a much stronger argument.
Posted by: Tim at Aug 30, 2005 5:13:46 PM
Tim,
I have to disagree with your argument:
"I don't understand those who argue that government needs to fund high risk investment. High risk means high losses. Its not any different than playing the lottery: high risk, high losses. Sure, every once in a while some one wins the jackpot -- and every once in a while the government invests in worthwhile research and comes up with the tcp/ip -- but for every winner, there is much more money lost on the losers. The costs are clearly more than the benefits. especially when, as Tyler has demonstrated, the market will get around to it soon enough anyway, and do it in a manner where benefits are clearly higher than the costs."
Computers, the internet and our interstate highway system are all large govt projects. Their present value is difficult to calculate, but we can be sure it is in the range of many tens of trillions of dollars. Yep, we fund lots of losers. These three are so overwhelmingly positive that they more than justify our overall social investment. I haven't included the entire medical industry, where lots of govt. research is basically given to private firms.
We are essentially long call options with all of the research the govt funds. Lots of losers expected, but with unlimited upside potential on all of them. Very occasionally, we get a civilization changing winner that improves our standard of living far, far beyond what the private sector could ever dream. The private sector is unwilling to fund lots of losers, and who can blame them. I don't. Its great when the govt does something for 30 years at a loss(computers, internet) and then it gets co-opted by the private sector. We get other small stuff like the bird in Alaska that repells mosquitos with something similar to DEET. I am guessing that SC Johnson didn't fund that research. But they will use it and we're all going to benefit.
I don't know this for sure, I haven't looked back in the press. I can however, imagine some staunch free marketers slamming computers back in the 40s and 50s. They were expensive, huge and unable to do anything well. Moths would fly into them and get stuck and short out the system. What part of the private sector would have spent the hundreds of billions to do all of this essential basic research? What is the NPV of the value added by computers? Its impossible to calculate, as we are in the infancy of this technology.
I don't want to say 'all govt research is worthy'. Of course it needs to be monitored and all that. But to think the private sector would have made the investments necessary to create these incredible positives is not realistic thinking.
Posted by: Mike at Aug 30, 2005 6:36:48 PM
It seems to me that the most accurate one paragraph history of this issue is: Once reasonably inexpensive PC's became sufficiently common and powerful enough to handle the demands of networking, and enough people became sufficiently sophisticated to want to network their computers, a networking standard sufficiently robust and convenient for time and the forseseeable future emerged. TCP/IP had the great benefit of being in the public domain and well tested, so it ended up winninng. Were it not there, something else would have emerged, albeit almost assuredly at greater cost and more time.
You can put what spin you want on that story, but it seems highly implausible to me to think that what we now refer to as the Internet wouldn't largely exist even had TCP/IP not been available as a public domain gift from the Defense Department. At the same time, it also seems highly implausible that we would have had as speedy a path to it.
FWIW, it also seems plausible that we would have a superior mechanism by now had we not, in the 90's, chosen to adopt a 30 year old standard, but I would be hard pressed to attack TCP/IP, or the path of Internet development, on that score.
Posted by: Scott Wood at Aug 30, 2005 9:25:39 PM
For me, the internet became useful and valuable when Mosaic arrived.
Posted by: Chris Adams at Aug 30, 2005 10:00:19 PM
The Internet exists, in part, because of five government policies: (1) the decision to build its predecessors (ARPANET et al., started with Ike); (2) the decision to let you own your own terminal (personal computer) and connect it to the phone network (FCC decisions 1968-1975 over Ma Bell's opposition); (3) open entry into long distance (FCC decisions in the 1970s over Ma Bell's opposition made all that backbone possible); (4) allowing ISPs to use phone lines (FCC Computer decisions in the 1970s over Ma Bell's opposition); and (5) FCC prohibiting Bell from charging 8 cents per minute for Internet use of her phone lines (1995-1996).
Posted by: John Berresford at Aug 30, 2005 10:05:40 PM
Al Gore would have a bone to pick with this argument. Everyone knows he invented the internet. :P
I do research on broadband internet, and in Canada, the government pays for the infrastructure due to highly convex cost structures and irregular population densities. The private sector provides internet access, but the government provides the infrastructure.
The analysis is obscuring the difference between provision and delivery. A government can provide 'health care' but this can be delivered privately by the market.
Posted by: Aaron at Aug 31, 2005 4:19:36 AM
As Tyler Cowen implies this is a "usefull corrective" on the view that without government we would not have an internet. The important thing here is that the internet only took off when it was privatized. This is not too surprising: the market really is best to find out what users and consumers want. No government policy can improve on that. Look for instance at Minitel in France. It only became popular when the French started to use it for other purposes then those for which it was build for by the government. And it could not stand up against the onslaught from the privatized internet. The other question is: who can be more innovative? And here matters are not so clear. One can argue that the real succes of the internet began with the world wide web. And the world wide web was not invented by a private company under pressure to compete but at a public institution. Indeed, Tim Berners-Lee invented the World Wide Web while working at CERN, the European Particle Physics Laboratory. So a usefull corrective yes, but not the definitive answer.
Posted by: ivan at Aug 31, 2005 4:24:56 AM
What makes the internet the Internet is the end-to-end principle--that is what enables you to connect to amazon.com from anywhere, not just from AOL or MSN. Before the Internet, there were many networks running different proprietary protocol stacks interconnected by gateways, and those gateways were enormous obstacles to the deployment of new services. Proprietary protocols also created lockin and the associated barriers to entries, so the incumbents liked that situation. From a consumer perspective, the problem was too *many* internetworking protocols.
However, no government standardized tcp/ip--it was a research project funded by the US DARPA and then NSF, but the body that standardized it (the IETF) is not chartered by anyone. OSI was the closest thing to a government-blessed protocol stack, and it failed miserably in the marketplace due to protocol bloat and a lack of interoperability guarantees.
IPv4 was (and is) seriously under-engineered as the Internet protocol, but its widespread adoption was still an enormous benefit, because it lowered barriers to entry and commiditized ISPs. That opened up the services layer to competition, with direct benefits to consumers.
I would not call this a triumph of government planning--most of the Internet's evolution was not planned. The essential "planning" step was to get out of the way (by rapidly privatizing the NSFnet backbone) once the commercial potential became clear (via the www and NCSA Mosaic).
Carnell, in the piece Tyler quoted, writes that the "floodgates of the Internet came open" only after it was privatized--but why was that even necessary? The existence of the explicitly non-commercial ARPA/NSFnet wasn't inhibiting development of a privately developed Internet--so why hadn't that happened independently? Isn't Carnell's argument an implicit admission that government intervention at the very least hastened the development of the Internet?
There's lots more I could write on this, but this is too long aready. I'll close with the comment that I don't believe the history of the Internet offers unqualified support for or against government intervention, but it does offer useful lessons about intervention.
Posted by: Dan at Aug 31, 2005 9:23:03 AM
Constant: "Things like ... Perl ... will not be widely used without government backing them up all the way. ...... Inventions like ... Perl cannot be developed in the first place without massive government assistance."
From whom, the US Department of Larry Wall?
Wait, I just got to the line about monkeys and human volunteers... Okay, you had me there for a moment, Constant. That was good.
ivan: "who can be more innovative? ... One can argue that the real succes of the internet began with the WWW. And the WWW was not invented by a private company under pressure to compete but at a public institution. Indeed, Tim Berners-Lee invented the WWW while working at CERN..."
As an inventor I always hate saying this, but invention's not all it's cracked up to be...
True, TBL invented the WWW at CERN, but MarcA brought it out of the lab and into the mainstream with Mosaic/Netscape. Have you ever used one of the early web browsers? No, as much as I admire TBL's vision, I think that you have to credit the growth of the web as we know it not to him but to his offspring. (And that's no small honor, either.)
Posted by: botboy at Sep 1, 2005 12:10:07 AM
The pill levitra which is primarily used to treat erectile dysfunction is now revealed as a suitable cure for problematic heart condition also. This new revelation on cheap levitra is indeed a boon for all impotency afflicted people across the world, especially for the heart patients. Heart is doubtlessly the most essential part of the body and so if this FDA approved medication cures you of heart problems along with erectile dysfunction then what is the point in delaying to order levitra? From your neighborhood pharmacy also you can buy levitra but the fact is that you will be able to gain access to cheap levitra only if you buy levitra online. Order Levitra through Levitra and move on with your war against erectile dysfunction.
Posted by: levitra at Dec 22, 2006 12:23:26 AM
When things comes about obesity and weight loss, Adipex diet pills are one of the first prescribed medicine that is used to control obesity. By suppressing people's appetite, cheap adipex has achieved people quick and effective weight loss. However, appropriate result of adipex diet pill heavily encouraged by proper diet, and exercise. Today adipex is in demand because of simple and safe appetite suppressing methodology to achieve quick weight loss. For the ease of people, online adipex has made the purchase of cheap adipex more comfortable. Moreover, online adipex is cheaper than available at physical stores. Why to wait any more if you are one of the depressed obese people. Just buy adipex and enjoy a beautiful shape and healthy life.
Posted by: adipex at Jan 25, 2007 12:13:04 AM
