Trusting Telcos With Internet Is Like Trusting Fox With Henhouse

It is absolute madness from a policy perspective to trust internet rollout to telco companies. Telcos have a strategic special-interest incentive to delay, hinder, and obstruct any public-interest bandwidth improvements, as ubiquitous, accessible, and economy-boosting bandwidth will shatter their legacy investments.

Around the turn of the century, people in Sweden were laughing heartily at discussions in the United States about whether to connect to the internet using “ADSL or Cable”. It seemed most people were either connected through their phone company or their cable TV provider. To people in Sweden, this seemed mind-bogglingly odd: in the small Scandinavian country, private entrepreneurs had been fibering apartment blocks wholesale for years. I had fiber in my own apartment in 1999, and keep enjoying a 100 megabit-connection with several static, public IPs – from where you’re reading this article, as I run my server from home.

For the record, I should add that it is uncapped, unmetered, and bidirectional. This was a given in 1999, and it should have remained a given, but hasn’t, for reasons I’ll get to.

People laughed well at how rapidly things were changing. Placing a phonecall seemed downright anachronistic – why were you paying half a euro by the minute for a nine-kay-six connection that could only be used for a voice application, when you had 100 general-purpose megabits – more than 10,000 times the bandwidth of the 9k6 voicecall – uncapped and unmetered in the wall for a fixed monthly fee? The telcos were doomed to die, the telco business was so utterly doomed to die. The only reason we’d still be making phone calls over the telco network in the short term future was as part of an expected phase-out to something that used the internet instead.

Then, just after the turn of the century, disaster struck the Internet, at least from a public policy standpoint if you want to promote growth and innovation. All the small ISPs were bought out by giant telecoms companies. All of a sudden, if you wanted Internet connectivity, you could only buy it from the old pre-deregulation national telco monopolies: Telenor, TDC, TeliaSonera, KPN, Telefónica.

At that point, the growth in household bandwidth flatlined. Whereas all other IT capacities – storage, computing power, screen sizes – skyrocketed, the growth of household bandwidth just stopped dead and died.

About as predictable as a grandfather clock, really.

It made sense, after all: these businesses were about to die from a disruptive upstart, so they pulled a Red Flag maneuver, pretending to embrace the Internet, whereas in reality, they are trying to prevent its growth and public utility for as long as possible. The net will disintegrate the telco business (and the cable TV business), and they know it.

It’s not just concrete hinders such as capping, application-level filtering, and metering that the telcos are raising to prevent the net’s utility potential – more than anything, it is the lack of investments in ubiquitous, cheap bandwidth that hinders our economy’s growth; investments that the telcos will never make, proven by the decade-long bandwidth flatline in the graph linked above.

In 1999, Sweden was #3 in household bandwidth as measured by upload, only behind juggernauts Japan and South Korea. Sweden has now fallen to #18, eclipsed by Hong Kong, Singapore, Taiwan… but also by Lithuania, Latvia, Bulgaria, Ukraine, Romania, and Moldavia (!). Overall, the European countries formerly behind the Iron Curtain seem to be leapfrogging the telco-encumbered Western countries.

Creating policy on the pretext that telcos will invest and research in internet bandwidth improvements is policy madness. You could just as well have made policy in the car’s infancy with the assumption that buggy-whip makers would make investments and research on automobile engines. It just won’t happen. It goes against every bit of business sense and strategic interest. What will happen instead is that the obsolete industry tasked with researching their replacement will happily accept any policymaking favors and tax money, do nothing (or even be counterproductive), and then ask for more favors.

You could also look at the Wi-Max fiasco for a very sad story on the topic, if you like. Wi-Max was basically “Wi-Fi for the cities”. It was a long-range wi-fi network designed to outcompete 3G and wireless data for city-wide coverage in terms of cost-effectiveness, reach, and deployment speed. Then, disaster struck as the telcos managed to politically place the Wi-Max frequencies in telco-controlled spectrum space. Predictably, again like a grandfather clock, the standard and the hopes it brought died instantly. We could already have had city-wide unmetered, uncapped, loginless data coverage – but it is just not in the telco operators’ interest. Of course it isn’t. But it is in the public interest. It is in everybody else’s interest.

Techdirt highlights a recent article where the telco lobby is rising its voice in demanding control of the net, in an article titled “EU Telcos: Give us more taxpayer money, and no one’s internet gets hurt”.

So, if you shouldn’t base the internet rollout on telcos (and you absolutely shouldn’t!), what’s the alternative? Saying that something is bad is not enough, you also need something to replace the bad with.

I would argue that energy companies are a significantly better public partner for rolling out the Internet (to give a tangible example of a better option). They are built for decentralization and resilience, not for centralization and control; they are built for five-nines delivery quality, they are built as long-term investments to wire households, and most importantly, they don’t have any existing cash cows that will be killed by the net, so it is not against their strategic interest. This week, until September 24, such a paper modeling a future Internet on the Icelandic energy grid is out for comments. You can access the paper, Islands of Resilience, here.

Oh, and as a final note, my 100-megabit network in my apartment – the network through which you’re reading this very article – is indeed delivered by my local energy company. Unlike telcos and cable, they don’t have any existing cash cows that will be killed off by the internet, and therefore, have no strategic incentive to delay or hinder it. Tell me – do you know anybody connected through a telco company that is running their server from their home? No? None? Not one?

I wonder how many home offices and small entrepreneurs that never took off because they couldn’t start small-scale from their homes, and so, how much damage the telco buyout of the EU Internet has already done to the European economy as a whole?

Rick Falkvinge

Rick is the founder of the first Pirate Party and a low-altitude motorcycle pilot. He lives on Alexanderplatz in Berlin, Germany, roasts his own coffee, and as of right now (2019-2020) is taking a little break.


  1. harveyed

    Very good Rick. I think you are right on the spot on this one.

    Government subsidised building of internet infrastructure 10-15 years ago and today the same government is passing laws which cripple the possible services on this same new infrastructure to “protect” the old ways of doing business.

    That is like subsidizing building railroads/highways in the industrial times but then to pass strange ego-protectionist laws that prohibit people and companies from using those same new modern infrastructures to create services to compete with the old businesses…

  2. Björn Persson

    Tell me – do you know anybody connected through a telco company that is running their server from their home? No? None? Not one?

    Me. 🙂 The entire domain Rombobjö is physically located in my home, connected by ADSL. But I’m sure I’d need more bandwidth if I had a website as popular as yours.

    1. Caleb Lanik

      I’ve found the most difficult part is getting a static IP, I have a reasonably fast connection from my cable company, and wanted to run a small server to be able to ftp files without a third party service getting involved, but in my city the only way to get a static IP is to upgrade to ‘business class’ service, which starts at 500 USD per month. I need to pay more than five times as much as for residential service for a static IP. That’s insane, and the only reason for it is to prevent people running a server.

      1. DynamicMike

        Get yourself some dynamic dns. will redirect to your IP for free. Point your domain at your dyndns address and it’ll find the server on your dynamic IP.

      2. printersMate

        For private use, a DNS entry is not required, and the IP address assigned should not changes while the server stays connected. will give the IP address of the server, and the hosts file of the client can be used to map a suitable name to this address.
        Self created certificate can be used to set up for ssh, scp etc.
        The IP address should only change on server reboot.

        1. jima

          “The IP address should only change on server reboot.”

          In theory, yes, but there are DHCP implementations out there that force a new (not refreshed — completely different) lease daily. As such, I agree with the recommendation of dynamic DNS when statics aren’t available.

  3. jimbo

    the only hope we have is that those in governments that are ‘encouraged’ to back certain industries or award contracts to certain industries are found out before it happens. if that doesn’t happen before the damage is done, just like everything else that makes money for individuals who are supposed to ‘look after the public’s interests’, we will be screwed yet again!

  4. Emil Kirkegaard

    The situation seems to be similar in Denmark, even with some of the same companies. Except, that we are farther behind than Sweden since we didn’t get fiber roll-outs in 1999ish (to my knowledge). As for the numbers in the sources given, we are listed 17th inre. download speed and 15th inre. upload speed. The reason Sweden’s numbers are so low (well, high, but higher is worse here) is that it is a much larger country, and people who live far away from cities get poorer internet, and those in cities thus have even better internet to compensate for that (thus keeping the average up). Can someone confirm my scenario?
    Median, and other percentile, speeds are perhaps more useful. Variance in SD would also be useful. Given my analysis, it should be higher in Sweden than Denmark I think.

    The large formerly state-owned company, TDC, is really slowing things down. Not only by buying competitors to stop competition, but also for insisting on high prices for slow copper-based (copper?! in 2012?) connections. The argument for this is the familiar: People have no need for faster connections, which is not generally true, but there is some truth to it. However, it fails to take into account the fact that faster connections create the opportunities that will create the need for them.

    Emil Kirkegaard,
    Pirate Party Denmark

  5. Troed

    Posted by an Anonymous Coward to a story about Google Fiber in Kansas – which has a “no servers” clause in the ToS:

    “Larry Page is really annoyed by the ‘no servers’ clause. In an internal weekly all-hands meeting he repeatedly needled Patrick Pichette about the limitation, and pointedly reminded him that the only reason Google was able to get off the ground was because Page and Brin could use Stanford’s high-speed Internet connection for free. Page wants to see great garage startups being enabled by cheap access to truly high-speed Internet. Pichette defended it saying they had no intention of trying to enforce it in general, but that it had to be there in case of serious abuse, like someone setting up a large-scale data center.”

    1. Caleb Lanik

      I’ve said it before, and I’ll say it again: If you can’t write your laws or contracts without adding clauses that only work as intended if enforced selectively you’re doing it wrong. Either tell me in the contract the cases under which I can run a server or don’t let anyone run a server, don’t leave it in a grey area so that I never know until after my website is shut down.

  6. Andrei

    As a person living in Bulgaria, I’d say you’ve hit the nail straight on the head. The people running BTC (the ex-state owned telecom) thought they had everyone by the balls because they owned the entire landline infrastructure. The bright ADSL future was before them… except they didn’t rush to put ADSL in place. We were stuck with modems or the occasional outrageously expensive ISDN line. This gave rise to small ISPs which bet their future on actual LANs with cables hanging between the buildings which only later evolved into MANs. If there ever was an ‘Internet for the People’ movement, that was it. Nowadays it’s not unusual to have fiber in one’s apartment though not exactly commonplace. Ironically, the present owner of BTC is one of the companies pushing cheap fiber… and ADSL too. As a sidenote, I am appalled at the prices people pay in the DSL-is-broadband-no-worries EU countries. Right now a 20Mbps uncapped connection with a static IP costs me €8 monthly. A similar package in the UK is 4 to 5 times as expensive considering they also pay ‘line rental’ money. I find £30 somewhat on the expensive side — 40GB cap, really, come on.

    Good Lord, people, do something about it!

  7. printersMate

    The early dial-up days of the Internet along with the limited power of personal computers made client server the preferred connection model. Cable service are also client server like,so this model was carried over into the permanent connections, by asymmetric connection speeds. The use of dynamic IP addresses, and often contract terms also make the use of home servers difficult.
    This is not a big problem for anyone who wishes to start a business, or offer a public facing web site, various hosting companies meet this need at a range of prices depending on bandwidth charges.
    The more interesting use for a home server is in private networks, where no domain name is needed. Connection to the server can be made dependent on knowing the actual IP address, and possession of a key issued by the server owner. This is the ideal route for family and close friends to carry out private conversations, share photos etc.
    This approach also extends to federated services, which includes relaying via mutual friends. This is the intention of the Freedom box project .
    A big advantage of this approach is that it separates public and private life at the Internet level. It would address the privacy issues of big companies watching every conversation on the Internet, and building profiles.
    The people who gain from no server clauses are big business, through advertising on services, and governments as it makes monitoring easier. They are the biggest barrier to the freedom box.

    1. Guest

      Hey, thanks for telling me about freedom box, I hadn’t heard about them before, but it sounds like a pretty cool project. Still in 0.1 for now, and largely developer only, but I’ll keep an eye on the project in the future. Projects like that can change the world with enough support.

      1. targ

        You should also check out Retroshare 🙂 It is an encrypted pseudanomymous sharing program.

  8. Andrei

    As a person living in Bulgaria, I’d say you’ve hit the nail straight on the head. The people running BTC (the ex-state owned telecom) thought they had everyone by the balls because they owned the entire landline infrastructure. The bright ADSL future was before them… except they didn’t rush to put ADSL in place. We were stuck with modems or the occasional outrageously expensive ISDN line. This gave rise to small ISPs which bet their future on actual LANs with cables hanging between the buildings which only later evolved into MANs. If there ever was an ‘Internet for the People’ movement, that was it. Nowadays it’s not unusual to have fiber in one’s apartment though not exactly commonplace. Ironically, the present owner of BTC is one of the companies pushing cheap fiber… and ADSL too. As a sidenote, I am appalled at the prices people pay in the DSL-is-broadband-no-worries EU countries. Right now a 20Mbps uncapped connection with a static IP costs me €8 monthly. A similar package in the UK is 4 to 5 times as expensive considering they also pay ‘line rental’ money. I find £30 somewhat on the expensive side — 40GB cap, really, come on.

  9. printersMate
  10. Ian Farquhar

    You’re absolutely right on this.

    Unfiltered, widely available Internet is not some optional commercial product, but a critical national infrastructure for a robust democracy. Those national telcos were created with the concept that phone infrastructure was the same, as by and large, they actually created competent and robust infrastructure. They might have provided pretty poor customer service, but the reality is that they used government money to build extremely good nation-wide networks in most countries which would never have been built by profit-driven enterprises (they’d have cherry picked the profit centres and left everyone else without service).

    Then along came the 80’s and the delegitimization of the public commons happened, and these networks were fire sold to commercial interests who immediately gained powerful monopolies, which they immediately began exploiting.

    It’s time to realize that the public commons can legitimately own large infrastructure, that the inefficiencies and inflexibility seen in previous years were an issue of policy and management, not of fundamental structure, and regain these national critical infrastructures. This is not to say that private competition shouldn’t happen – it should – but government ownership is entirely valid and appropriate and, in the absence of the need to return a profit, can be cost efficient if properly managed. The concept that only for-profit companies can be efficient is dogma, not reality.

    It’s still a challenge. You do have to give full marks to Labor here in Australia for creating the National Broadband Network, largely as a band aid for the disaster which was the Howard government’s float of Telstra. That float should never have happened.

  11. Pedro

    They must be on something to even think of something like this *sarcasm*. More censorship attempts? This whole economic crisis is bad enough and many other things and now this too?

    * Leak shows plans for large-scale, undemocratic surveillance of all communications

    * EU Officials Propose Internet Cops On Patrol, No Anonymity & No Obscure Languages (Because Terrorism!)

    * Police to ‘patrol’ Facebook and Twitter for terrorists under EU plan

    * “Malmstrom – it looks messy” (translated from Swedish)

    This is just too much. >:(

    1. printersMate

      Apart from propaganda, and possibly code phrases in public forums, and these via third parties, no ‘senior’ terrorist or criminal will use electronic communication channels, as they are too easily monitored. Note, Bin Laden relied on thumb drives via sneaker net for his communications.
      Are these measures aimed at non violent terrorists; aka various protest groups and or anonymous and friends?
      Because of terrorism seems to be an excuse to build a police state.

      1. Caleb Lanik

        Sneakernet is pretty good for avoiding online snooper systems. Next best thing is TOR of course, if it can keep cops from locating pedos, it probably works just as well for terrorists. Hell, I’ve seen .onion sites with detailed instructions on bomb making and meth manufacturing. Admittedly, a few of these are probably honey pots for governments but certainly not all of them.

  12. […] rättsväsendet tar order från främmande makt, som i sin tur går storföretagens ärenden. När telekombolagen får ta kontrollen över den nya tekniken som kommer att förstöra deras affärsmodell att debitera per […]

  13. Anonymous

    I want to keep my regular telephone line for as long as possible. I know that there is devices that makes it possible to use your regular phones to make IP calls, but I want to keep the LINE, and I think everybody should.


    Because, as an electronics junkie I know that regular telephones are far more resilient in case of an emergency. The less technology needed, the better. It is only unfortunate that the telephone switches in Sweden are now AXE. If there’s a big solar eruption, a EMP attack, or any kind of powerful interference, the old analogue phone wins over computer fine-circuitry every time. By that I of course don’t say old phones and phonelines are invulnerable, but they are better off than more advanced circuits.

  14. Johan Tjäder

    I thing this is a tough case to sell. The statistics presented here are questionable in the sense that it isn’t really obvious what it really says. And the causality between the statistics and the conclusions that this is a conscious and coordinated strategic action from the telcos isn’t really obvious either.

    So, Sweden is no 18 on a list. A listing of the average upload speed in tests per country. But first of all, was launched in 2006. It couldn’t possibly be used for a comparison for Sweden’s position in 1999. And second, this list doesn’t tell you household bandwidth. it has a systematical error in that households without Internet aren’t counted.

    Yes, it’s true that almost all of the small ISPs failed, but why? Those were the days of the dial-up internet. Most ISP relied on an existing infrastructure – the plain old telephone service – to make their business. It’s obvious I’d say that most of them lacked the financial resources to handle the transition to fiber based access. It’s an enormous investment. Here in Stockholm, the city provided for the backbone fiber. It enabled a lot of independent operators to provide you with internet access independently of the telcos.

    It’s undeniable so that other IT-capacities have skyrocketed in comparison. But investing in a bigger hard drive or a larger screen or a factory for manufacturing these devices, is hardly the same thing as building a new high speed network infrastructure. That takes some doing, and a lot of money – and the bulk of the investment these last ten years have been in mobile internet build-up. Back in 1999 when Rick had his 100 M wired access, everybody’s wireless access was this 9.6 k GSM dial-up. Not exactly what you’d run youtube on.

    The telcos have been aware for quite some time I’d say that their old wired services were over the hill. Interestingly I’d say that ADSL and similar technologies actually prolonged the old copper wire network lifespan with at least a decade or so. But the real fear (and hope?) for the telcos has always been the wireless technologies.

    So while I cannot prove that there is no conspiracy, I think there are a lot of reasons of why things got where they are.

  15. Lars

    …and today Telia announces that they, in their infinite goodness, have decided not to ban IP telephony in their mobile network. Instead, they are going to raise prices on data transfers in general so their own old-style telephony becomes more attractive in comparison.

    You really couldn’t ask for a better confirmation of the point of this blog entry.

    1. Johan Tjäder

      I’d say that’s an entirely different point. Telia has realized (well actually it’s bosses, because the technicians would have realized this long ago): you cannot, by any realistic means, filter out real time voice application traffic from any other in an IP-network.

      But what all customers will have to realize is that the voice traffic has long subsidized the data traffic, because customers have been paying a premium for those bits and bytes that has been carried for voice. And in the 4G networks this separation is no more. Telia voice and Skype and any other IP-telephony application will truly be on the same competing basis, and they can’t really stand that apart when it comes to charging.

      This means that if we want to have a roll out for faster 4G networks, data customers will have to fill their share of the investment. The telephony service may be a mature market, but mobile data networks sure isn’t. It is to be expected that any operator would want to make a substantial profit, if you want to be a part of it. Or you can stay on 3G.

  16. Filino Rupro

    Hello From Portugal. Excuse my bad english!
    I think the way to go, it’s create a free structure of comunication. How? I have some ideas but I don’t know electronics to make it. But it’s easy for someone familiar with radio. The basic idea it’s a kind of radiomodem. Emission on a random frequency, and reception by scan on all radio spectrum. This works, the french military have AM radios with similar idea.
    The way it works, means more users, more bandwith! This is the only way of an independent and free internet in the hands of citizens.
    If any electronic hacker it’s interested go for it PLEASE!

  17. […] to kill the net instead, the enabler of their successor industries. The copyright industry and the telco industry […]

  18. […] result of the Internet. But if they can assert themselves as masters of the Internet, they get the desired ability to kill the potential of the net to replace them and their current […]

  19. […] como consecuencia de la Internet. Pero si pueden hacerse valer como maestros de Internet, tienen la capacidad deseada para matar el potencial de la red para reemplazarlos por sus actuales rentables productos. En otras […]

  20. […] A indústria de telecomunicações e a indústria do direito autoral não querem mesmo saltar, obviamente. Portanto, eles estão, compreensivelmente, tentando matar a rede em vez disso, quem ativa as indústrias que irão sucedê-los. Ambos, a indústria do direito autoral e a indústriade telecomunicações. […]

  21. […] But the report also highlighted something else: the Internet provides the same services as the telco industry, at a cost five order of magnitudes less. In other words, the telco industry is currently overcharging for voice service by five orders of magnitude. (This ties well in with our previous observations that the future sales value of voice and storage is exactly zero, but the OECD is arguably a much heavier voice than this site.) Their ability to do this – to prevent the net’s utility, the public interest, and economic growth – is entirely due to a gatekeeper position that comes from having strategically bought all the small ISPs in the infancy of the net’s commoditization. It is now completely against the telco industry’s interest to roll out internet connectivity at the pace of the public interest, so it doesn’t happen. […]

  22. […] Article original de Falkvinge, en anglais. […]

  23. […] incumbent telco industries to prevent this economic growth (and their replacement) by giving them any kind of control of the Internet, like with the upcoming coup attempt at the ITU’s meeting in […]

  24. […] telemonopolen fått ta kontrollen över utbyggnaden av bredband, och utnyttjar det till att fördröja bredbandsutbyggnaden och överdebitera på absurda […]


    Timely article . I was fascinated by the info , Does anyone know where my company could obtain a fillable 2012 TAR-2003 copy to edit ?

Comments are closed.