Patrick from Popehat points to this old article by Paul Krugman that should be a profound embarrassment to him. In the column, he predicts that the internet is just a passing fad and won’t have much influence over the economy at all. For instance:
* The growth of the Internet will slow drastically, as the flaw in “Metcalfe’s law”–which states that the number of potential connections in a network is proportional to the square of the number of participants–becomes apparent: most people have nothing to say to each other! By 2005 or so, it will become clear that the Internet’s impact on the economy has been no greater than the fax machine’s.
* As the rate of technological change in computing slows, the number of jobs for IT specialists will decelerate, then actually turn down; ten years from now, the phrase information economy will sound silly.
Oops. The punchline: This appeared in a magazine called Red Herring.
37 comments
1 ping
Skip to comment form ↓
dean
July 16, 2012 at 12:14 pm (UTC -4) Link to this comment
In Mr. Krugman’s defense, he knew Al Gore had invented the internet (a true fact you hear all the time, even now) and so concluded that its creator doomed it to failure.
d cwilson
July 16, 2012 at 12:18 pm (UTC -4) Link to this comment
Krugman is a brilliant economist. Unfortunately, he stepped out of his area of expertise when he started making predictions about tech trends.
The Lorax
July 16, 2012 at 12:27 pm (UTC -4) Link to this comment
I wonder if anyone said the same thing about the telephone, or the television, or the computer itself.
Yet, he’s saying that about all three rolled into one.
*… clap … clap … clap …*
psweet
July 16, 2012 at 12:48 pm (UTC -4) Link to this comment
I remember reading a statement from Lee Iacocca once – apparently when he was running a company that made bras, he told the maker of Velcro that he didn’t see a use for it!
Nepenthe
July 16, 2012 at 12:52 pm (UTC -4) Link to this comment
So, was Red Herring a serious magazine? If so, that’s the dumbest name for a place for the exchange of ideas ever, perhaps excepting Conservapedia.
Jafafa Hots
July 16, 2012 at 12:54 pm (UTC -4) Link to this comment
Iacocca also didn’t see a use for a reinforcing structure between the rear panel and the tank in the infamous Ford Pinto, saying “Safety doesn’t sell.”
dingojack
July 16, 2012 at 12:57 pm (UTC -4) Link to this comment
d cwilson – Messieurs Dunning and Kruger the courtesy phone please…
:) Dingo
typecaster
July 16, 2012 at 12:58 pm (UTC -4) Link to this comment
Snark aside, Gore never claimed to have “invented the internet”. He did have a lot to do with authorizing and funding the research programs that led to its development. I’m pretty sure dean knows that, but just wanted to make it explicit.
But yes, Krugman put his foot in it fifteen years ago. Kinda like the classics, Bill Gates saying that 640K of RAM should be enough for anyone or the head of IBM saying that a dozen computers should handles the world’s needs. Predicting the secondary or tertiary effects of new technology is difficult. In 1997 the Web was a fairly static medium, used for putting up information that didn’t change frequently. The commercial possibilities of interactive web sites was just starting when this article was written, and embedded media was still in the future. It wasn’t clear even to those in the industry just where the technology would go, and Krugman wasn’t one of them.
As an illustrative exercise, think about emerging technologies today – quantum computing, neural network architectures, genetically engineered lifeforms, or something else of your choice. Predict the economic impact in the near and medium future. In fifteen years, pull these predictions out and mock those who got it wrong. Who wants to play?
tacitus
July 16, 2012 at 1:14 pm (UTC -4) Link to this comment
I once got into an argument with a colleague about the need to account for 64-bit systems in the user-interface code for the operating system we were working on (which we were converting from 16-bit to 32-bit at the time).
I was the one who said that it was unnecessary for us to make any special provisions for 64-bit code since it would be a long time before anyone would be running 64-bit operating systems. That was in 1995, seven years before Linux became the first native x86 64-bit OS.
But although my friend still doesn’t let me live that comment down, it turns out that I was right, in the end. for the operating system in question was IBM’s OS/2, and all development stopped on it long before a 64-bit version was needed. Ah well.
Raging Bee
July 16, 2012 at 1:14 pm (UTC -4) Link to this comment
In fairness, depending on what part of the Internet Krugman saw, that could have sounded like a reasonable prediction. If you’re exposed to the useful bits, like on-line account management, news, and weather prediction, you’d be a fool to think that would go out of style and never improve (assuming, of course, that any of those things were feasible and reliable back when Krugman made that prediciton). But if all you saw of the Internet was LOLCats and porn, you’d be forgiven for saying there’s no future in it.
Predictions like this mostly sound stupid in hindsight. A similar prediction about eight-track tapes would have sounded spot-on.
dean
July 16, 2012 at 1:23 pm (UTC -4) Link to this comment
“Snark aside…”
Snark was the entire content of my comment.
timpayne
July 16, 2012 at 1:37 pm (UTC -4) Link to this comment
Speaking of myopia, I remember this boneheaded post on Citizens United by a reasonably savvy blogger: Yesterday’s Supreme Court Ruling: Much Ado About Little
whheydt
July 16, 2012 at 1:52 pm (UTC -4) Link to this comment
Re: #8, typecaster…
Never mind the real cutting edge stuff…I’ve been trying to predict how the Raspberry Pi (see http://www.raspberrypi.org) will develop over the next 2 to 3 years (512MB memory, 1GHz processor) and most commenters think I’m wrong by being too optimistic.
Strewth
July 16, 2012 at 1:56 pm (UTC -4) Link to this comment
Raging Bee, I don’t know about Krugman, but for me “They have found a way to use this technology to make, sell, and distribute pornography.” is a huge invest big in this thing sign. Pornography drives the development of communication media.
Not the only driver of media tech of course, but a big one.
Dennis N
July 16, 2012 at 2:11 pm (UTC -4) Link to this comment
Krugman blew the prediction on that one. But from everything I’ve seen from him, he is the type to admit he was wrong instead of equivocation and try to paper over his mistake, unlike ohhhh pretty much every other talking head you see show up on the tele-vision.
jws1
July 16, 2012 at 2:13 pm (UTC -4) Link to this comment
I’m with typecaster @ 8. This post is nothing more than “look, somebody slipped on a banana peel! Ha ha, very funny!”
dingojack
July 16, 2012 at 2:25 pm (UTC -4) Link to this comment
whheydt – you might find this interesting. Notwithstanding the ol’ ‘Reports of my death have been greatly exaggerated’.
Dingo
The Lorax
July 16, 2012 at 2:27 pm (UTC -4) Link to this comment
I’ll play.
Quantum Computing – Due to the technology still being developed, I predict it’s going to take another 5 – 10 years before they become mainstream, and about another 5 – 10 years before they measure up to standard computers (cost, performance, etc) but they will be hyped up due to the woo-factor of the word “quantum”. However, their potential is much higher, so I predict within our lifetimes they will out-perform the current tech by far.
Neural Networks – Not sure when, exactly, but I’m going to guess that most computer OS’s will have some form of basic AI to help predict what a user will want. Most programs will probably use it, too. Hell, once computing power reaches a certain point, I wouldn’t be surprised if basic AI comes with every program. Of course, if you’re talking about physical neural networks, I still can’t predict how long that’ll take to really nail down, but I’m sure eventually there will be android prototypes with advanced AI due to a physical network rather than a programmed one. Given the ethical thing, it’ll probably take years to become accepted (like cloning).
Genetically Engineered Life forms – Knowing what the genetic code is and knowing exactly what each modification will do are very different things, so customized genetically engineered pets are going to be a long way away, but I don’t doubt we’ll eventually be able to custom-order cats and dogs with specific details. And it’ll probably become necessary to produce genetically engineered food in order to feed an increasing population, unless 3D printing takes off (it will) in such a way that various chemical slurries can be used to produce “printed” food, which may mitigate the need to grow food.
I also think neural interfaces will take off soon; devices will respond to your thoughts, rather than your commands. Not implants, but brain waves. It’ll take a while, much like voice recognition, and it will probably never be without bugs (until that AI software I mentioned comes around), but I’m sure it will become mainstream in due time.
As for the future of the Internet… all I can say is…
_o| Connect ALL the things!
dingojack
July 16, 2012 at 2:32 pm (UTC -4) Link to this comment
The Lorax – “Connect ALL the things!”
Resistance is futile!
:( Dingo
Marcus Ranum
July 16, 2012 at 2:55 pm (UTC -4) Link to this comment
Quantum Computing – Due to the technology still being developed, I predict it’s going to take another 5 – 10 years before they become mainstream, and about another 5 – 10 years before they measure up to standard computers (cost, performance, etc) but they will be hyped up due to the woo-factor of the word “quantum”.
The question is: Will Deepak Chopra will get a job as a marketing spokesman for Intel.
lofgren
July 16, 2012 at 2:57 pm (UTC -4) Link to this comment
If this was pre-Napster I think it can be forgiven. I don’t remember the state of Napster in 1997 but I know by 1999 it was huge and no economist should have missed the implications of delivery content directly to consumers rather than going through a middleman of a physical product.
Even if we generously allow that by “internet” he meant “world wide web” or AOL, somebody who makes a fair portion of his living saying things ought to have seen that “people have nothing to say” is just curmudgeonly crankery. Even if the web remained a text and image only medium it already had the potential to displace all other forms of written communication. Although they had not been built yet, the Kindle, iPod, and smartphones were already more than a gleam in somebody’s eye by that time. Just a smidgen of forward-thinking would have at least allowed for the possibility that a portable web would be a game changer, and enough players were still in that game that it was way too early to be declaring such a thing an impossibility.
Also I disagree with the comparisons to CB radio or 8 track. Disregarding the potential of the internet is more like saying that the underlying concept of the 8-track (privately owned recordings the user can access at-will) or CB radio (the ability to broadcast sound over long distances to be decoded at the destination) were unlikely to succeed. You would have to deny not just the impact of 8-track but also of CDs, cassettes, VHS, etc.
Krugman appears to have honed in on one use of the internet (possibly not even the most typical use in 1997 when my mom was already emailing her coworkers in the office next door rather than walk 20 ft.), socialization, and decided that there is no future in that and therefore no future in the entire technology.
While I can sort of forgive him for not seeing the potential of the internet as a tool for socialization (this was pre-Meetup, pre-Facebook, pre-MySpace and I think even pre-whatever that one that everybody was using before MySpace), I have to say anybody who thinks that having nothing to say might be a barrier to people saying things nevertheless is not a very astute student of human nature. While nobody ever figured out how to make money off of AOL, it made it clear that the only reason bulletin boards and other online lurkspaces weren’t more popular outside of the geek tech crowd was a simple lack of access, not a lack of interest. Even if nothing besides 4chan and reddit ever evolved out of AOL and the increased access to the internet that coincided with their CD blitz, it would still be pretty damn relevant.
lofgren
July 16, 2012 at 3:19 pm (UTC -4) Link to this comment
Also the comparisons to quantum computing is a little unfair as well. It would be more accurate if quantum computers were already in use by millions of people and getting more popular everyday, if every university provided every student access to quantum computers, and if quantum computers had been available to hobbyists and profit-seeking corporations for almost 20 years already.
Ed Brayton
July 16, 2012 at 4:04 pm (UTC -4) Link to this comment
timpayne-
What I wrote in that post was and remains accurate. Did you see even a single corporation taking out ads endorsing candidates in 2010, or in this year’s election? Nope. They did what they always did, give money to third party groups to put out “issues ads” against candidates. The unleashing of the SuperPACs did not happen because of Citizens United, it happened because of a 4th Circuit ruling that was not appealed to the Supreme Court.
D. C. Sessions
July 16, 2012 at 5:01 pm (UTC -4) Link to this comment
Off by a bit, there. I was running 64-bit Linux on a DEC Alpha system in 1997.
escuerd
July 16, 2012 at 5:21 pm (UTC -4) Link to this comment
This part makes some sense in context. He was writing this at a time when the economic potential of the internet was overhyped leading to the “dot com” bubble (a name which does sound kind of silly now).
And though I’m not sure precisely what he means by saying the growth of the Internet will “slow drastically”, it does make sense to take any predictions based on Metcalfe’s Law with a grain of salt for pretty much the reason Krugman says (most people aren’t talking directly to one another, and there are limits to how many people any human can interact with in a substantive way).
This isn’t defending the statement as a whole, of course.
steve oberski
July 16, 2012 at 5:55 pm (UTC -4) Link to this comment
He got that part right, not that it stops them (it’s never stopped me).
steve oberski
July 16, 2012 at 6:09 pm (UTC -4) Link to this comment
@D. C. Sessions I was running 64-bit Linux on a DEC Alpha system in 1997.
Your only alternative was 64 bit Windows NT (the horror). I don’t think Digital ever ported any of their in house OSes to the Alpha and after running Unix on PDP 11s and VAXes in the 1980s I learned to truly despise Digital’s strong arm attempts to force you to use their OSes.
I ran 64-bit Linux on the Alpha (21264 EV6) as well, it was only for seriously deranged early adopters or if you really, really, really needed the larger address space. That sucker used more power than the Intel Itanium which is saying a lot.
caseloweraz
July 16, 2012 at 6:59 pm (UTC -4) Link to this comment
So Dr. Krugman was wrong. I think he’s entitled to be wrong a time or two — especially about the acceptance of technology. This article was written in June 1998, and IIRC even Bob Metcalfe was predicting some sort of collapse of the Internet back then (although in his case it was a technical failure: the collapse of the “big pipes.”)
The Internet was neither so easy to use nor so useful then. I was on it mostly for Usenet; I browsed the Web with Netscape or NetCruiser, but that was mainly recreational.
I’ve got to gripe about one thing: I remember Jakob Nielsen’s criterion that a Web site had to load fully in 3 seconds or less. Sure wish that were still the gold standard.
regexp
July 16, 2012 at 7:08 pm (UTC -4) Link to this comment
@typecaster #8
Gore put his foot in his mouth but what’s humorous is that you state that Gates made the 640k comment – and there is zero evidence he said that.
re: red herring and predicting the future
Red Herring is a tech mag that has been around for a while. There is no date on the article so I’m not seeing how someone can make fun of Krugman over this (especially since there are so many other things to make fun of him about).
But this is a perfectly plausible opinion if it came out before or during the invention of an easy to use web browser and means of encryption for transactions (i.e. SSLv1). Those events changed the Internet from being a tool for academics and researchers into a real platform for everyday people to use. A lot of other smart people got it wrong at the time as well.
tacitus
July 16, 2012 at 7:39 pm (UTC -4) Link to this comment
Not quite — I did say “x86 64-bit” — i.e. Intel-based Linux. OS/2 was an x86-only operating system, so in context, I was right. (Well, I was out by one year, but that’s because I can’t do basic math.)
I’m in the process of writing some short stories around the subject of “First Contact” (with aliens) and one of the themes running through them is the gaping chasm between our technology (in its infancy, really) and the technology of a civilization that’s been around for, say, half-a-million years. Given how tough it is to predict technology trends over a single decade, just imagine what a thousand, ten thousand, or a million years would bring (assuming we get our act together and survive that long, of course).
lcaution
July 16, 2012 at 9:32 pm (UTC -4) Link to this comment
First, re the infamous Gore quotation. He mispoke, but it was his legislation which converted Darpanet, a military/government/university collaboration, into the commercial internet we know today.
I found this link which, I think, accurately reports what happened:
http://tomliberman.wordpress.com/2012/03/27/did-al-gore-invent-the-internet/
As for Krugman: He did show in that piece a remarkable lack of insight into how much people like to talk. He had obviously never been on any of the bulletin boards that proliferated in the 1980s. But in that same piece, he made a right-on prediction re the U.S. economy.
It’s lots of fun, but somewhat unfair, to criticize inaccurate predictions. Surely we could at least use ratios (percent correct to percent incorrect) before we start to skewer folk.
As for the Internet: Mosaic was launched in 1994 or 1995 and was, a think, the kickstart software for the public’s use of the internet. The earlier search engines were hard to use and very limited.
By 1997, the internet was both a key part of my job and my leisure time. But I could not have predicted just how much the net would grow or how central to our lives. That took a combination of hardware advances (faster computers), new communication structures (you cannot understood the turtle unless you’ve used a 300-baud modem), serious public encryption, and companies willing to risk building a new sales system.
I never liked WordPerfect – but my standard word processor remains a Windows-based product that went out of production at least a decade ago. Krugman was definitely right about bloatware.
robertrichter
July 16, 2012 at 11:23 pm (UTC -4) Link to this comment
Did anyone actually check to see whether his predictions were accurate? All I see is a blithe dismissal.
I mean, yes, they sound wrong, but are they?
What was the economic impact of the fax machine? What was the economic impact of the Internet? Until you’ve quantified and compared them, you can’t reasonably make the claim that “the Internet has had no more economic impact than the fax machine” is wrong.
How fast was the Internet growing in 1998? How fast is it growing now? At any point between 1998 and the present, did that rate of growth decline substantially? What did Krugman even mean by “the growth of the Internet?” Has anyone asked him?
There are a lot of ways to measure the Internet’s size (and thus to derive its growth,) one of the most interesting to a professional economist is surely the tech business share of GDP. How did that fare between, say, 1998 and 2002? At any point in the last fifteen years, did IT employment ever decline?
Perhaps more importantly, why bring this up?
Benjamin "Derp" Geiger
July 17, 2012 at 1:28 am (UTC -4) Link to this comment
Krugman is far from the first to make this mistake. Vannevar Bush made the eponymous prediction.
Azkyroth, Former Growing Toaster Oven
July 17, 2012 at 1:43 am (UTC -4) Link to this comment
Seven years IS a long time.
But those are the parts that exploded and drove it!
Azkyroth, Former Growing Toaster Oven
July 17, 2012 at 1:44 am (UTC -4) Link to this comment
This is the most terrifying thing I’ve read for weeks. O.O
lofgren
July 17, 2012 at 4:22 am (UTC -4) Link to this comment
Well, it didn’t revolutionize the movie and music distribution apparatus or kill weekly magazines almost outright. It didn’t lead to a massive ongoing infrastructure overhaul by some of the most powerful communications companies in the country. It didn’t spawn TV shows or launch the careers of dozens or hundreds of stars who got their start with hilarious faxes on youFax. It hasn’t ruined the reputations and careers of many celebrities by leaking salacious letters they wrote to millions of faxers.
Whatever the economic impact of the fax machine, it’s safe to say that it was less than the internet. I’ll put $20 down that says that the economic impact of the entire life of the fax machine – including every dollar spent on r&d, every fax machine ever purchased, every mechanic ever trained to repair them, and every hour saved by faxing instead of wiring or phoning in some information, or even mailing it – is completely dwarfed by the amount of money put into the internet on a single day in 2011.
'Tis Himself
July 17, 2012 at 9:42 am (UTC -4) Link to this comment
So a human being made a comment outside his field of expertise and that comment turned out to be wrong. Mark me unimpressed.
Incidentally, Krugman is a highly regarded economist. While I have my disagreements with some of his economic comments, I do pay attention to what he says. When he’s talking about a non-economic subject I give him as much attention as I give every other non-expert.
Tomorrow is more like today than you think « The Words on What…
July 16, 2012 at 4:09 pm (UTC -4) Link to this comment
[...] came across this post at Dispatches from the Culture wars. It got me thinking, and eventually my mind thought about [...]