Skip to content
Welcome guest. | Register | Login | Add
About | Wiki | Legacy

It's not the Internet anymore

It's important that we remember what makes the Internet so interesting and unique. There are two crucial characteristics:

  1. It's fundamentally decentralized, meaning you can cut out any part without affecting the rest,
  2. It allows freedom of access, meaning you have the same ability to access and write it as anyone else.

Because they permit extraordinary flexibility and rapid growth, both of these characteristics have brought the Internet way beyond any other network. Today, they are endangered. How come, and what can we do about it?

Ones and zeros are boring

Traditionally the cost involved with "Internet", for the end-user, was access. It was measured in kB/s and hours per month – the idea was that you paid a company to transport data (ones and zeros) from and to your computer.
Progressively though, all of these companies out there have understood that there is not much money to be made simply transferring data. Carrying ones and zeroes is mostly a matter of managing hardware, a domain where competition is tough and standing out is hard.

Providing content, however, is much more interesting and lucrative. Because there is some consumer decision involved, there is branding to be made - and therefore, room for differentiation and high profit margins. Voilà! Suddenly my Internet Access Provider becomes Internet Services Provider (ISP). Regular and mobile phone access, digital TV or pay-per-view movies, nation-wide wireless subscription, mobile phone ringtones, portable music player with legal music downloads... There is no media that they are not into, providing access, and, especially, content.

Controlling both sides of the cables

To keep control of the content, the key idea is to be at both ends of the cables. The result is currently sitting in my living-room. It's a box that replaces what was formerly merely a modem; now a sleek device, it stays on permanently and can manage ethernet, wi-fi, phone and TV.
It's also the reason my mobile phone only surfs "selected" websites, or why iPods won't allow to exporting music.

What happens inside these boxes? How do the operating systems inside manage the data I am sending through them? There is no way to know. Because we have no freedom over that software, we have no knowledge nor real control over it. Combine this with our ever increasing use of on-line services (something I have called the disappearance of the operating system), and you have created a network that is different depending on who you access it with.

Of course, if the information you can find and read depends on your access provider, there are chances that your thinking and writing also depends on them.

Do we want the Internet, or just some networking?

So all of this concurs to creating something that's not the Internet. Remember the two key characteristics above? When we ask ourselves whether a particular service contributes or detriments to the Internet, we have to answer two questions:

1. Does it create a closed private network?

Facebook, Linkedin and their peers aren't part of the Internet. Anything you do inside stays inside. If you cannot take your information home, then you shouldn't give somebody the legal and technical means to manage it – be it at zero price or for a fee.

Yahoo! Messenger, Skype and their siblings aren't part of the Internet. When you use a black-box program to chat, which connects to a black-box server, which only accepts people with a registered account within that black-box world, you are part of a closed, private network.

The incentive for a company to close up its network is great, because it binds users to their products. This isn't mandatory for them to generate revenue, however. For example, Facebook earns money by re-selling demographic consumer information (ex: how many campus students under 20 have watched this or that movie) and displaying targeted advertising. There is nothing wrong in that, if the users are free to leave.
But these private networks stay under the control of the entity that owns them. Only they can decide and influence their growth and evolution.

2. Does it create a non-removable block?

Imagine one single computer that would be necessary for the web to work. Imagine every single user in the world systematically turning to this unique giant directory, in order to access anything in the network. Imagine every creator/participant in the network having to register in the directory to exist inside the network.
Imagine that computer rating every page of the network with a 10-point grade. Imagine the company behind that computer earning its revenue from the lowest-rated pages. And imagine the grade-attribution algorithm being entirely secret.
You've got it! It's Google, of course.

As strange as it might sound, Google is certainly into networking – but it's not helping the Internet. There isn't an influential website which can afford not feeding Google with its content. There isn't a low-ranking commercial website that can afford not buying their advertising. Simply because we users have consecrated it as our do-everything Internet service.

Crumb: a hint of an answer

We can change the course of things. The two keys to success are open protocols and a decentralized system.


Vision: Instead of signing up into different services and logging in into different private networks, Joe user has all of his social Internet uses met by one application, Loaf. He can do either of the following:
1. Run a limited version of Loaf at no cost, through various Internet companies, similar to today's blog providers.
2. Run the full application with a web host of his choice (at the cost of a hosting fee).

Loaf is free software, and is based on an open set of protocols called Crumb (for the sake of finding a modest name). Anyone can develop/adapt an application to read a Crumb profile.

A decentralized social network

The key idea is to create a social network that does not rely on a private server. Joe creates his Crumb profile through Loaf, and interacts with other users who also have a Crumb profile. Any of them can host their profile where they choose – it is the working principle of the Jabber instant messaging, applied to social networking as a whole. Here are a few aspects:

An ID and a buddy list — The focus should be on private communication, regulated with identification similar to that of Jabber. Joe lets Janet access his profile by adding her Crumb ID ([email protected]) in his buddy list.

Text and picture sharing — The core of the exchange will of course be text, in the form of blog-like entries, and picture sharing, with a system of comments. From basic functionality, we can later evolve into more advanced, Facebook-like interaction.

IM integration — Thanks to the maturity of Jabber/XMPP, it should not be difficult to integrate a Jabber account into the Crumb profile, and an XMPP server into Loaf. Instant messaging could be done directly on-line through the Loaf interface, or more traditionally with a PC client.

Room for plenty more — Once the features above are working, a set of plug-ins could be developed for Loaf. Some ideas include a full webmail, private file exchange (ex. allowing arbitrary people to access one large file), public blog, and streaming video hosting. In the long term, Loaf could evolve into a complete personal work and leisure space, provided by companies like Facebook, but which would give users the possibility to leave with their data at any time and host their profile elsewhere.

And does it all matter in the end?

Yes it does. More and more of our computing is networked, and the day might come when the only software used on your personal computers is a set of hardware drivers. The core of our activities and information will stay online — it's now up to us to decide whether this will be inside a few competing, closed private networks, or within a mesh with no center nor owner, otherwise known as the Internet.

It's not that difficult. The majority of what we need is here: RSS for dynamic content, XHTML for text, PHP-capable hosting, XMPP for instant messaging, high quality blogging software. We're only lacking a single, attractive platform to compete with the Facebooks out there. Perhaps Crumb exists already somewhere; if so, it's time we heard more about it. So everyone can contribute to the Internet again.

This article is licensed under CC-BY-SA terms.
Olivier Cleynen (personal page) is co-founder of the GNU/Linux Matters non-profit, focused on international Internet advocacy for free software.


Nice article! Not quite on


Nice article!

Not quite on topic, but you reminded me of it by mentioning google ads: I think it's interesting google gives websites who sell their own text link ads a lower pagerank, while google is in the business of selling text link ads itself. A lower pagerank means advertisers will want to pay less for the ads (because they think a high pagerank means lots of visitors who are willing to spend money), and so a website owner who previously took care of his own business might now feel forced to use google's services instead... and so google will get part of the advertising revenue.

To put it in a bit over the top way, google is saying "Nice website you have there. Would be a shame if someone made it look less popular." Smiling

This is an excellent article

This is an excellent article and a very important message that should reach, ideally, all internet users.

We're caught to believe that this is what internet should be like, for lack of imagination of better ways. Proprietary software attitude has done enough in having people believe that closed is not bad and now that the network becomes more and more important this attitude proceeds in actually making internet less of what it initially was and is supposed to be.

I'm thinking maybe an idea of a completely open search engine merits revisiting. We've been cooperating with for a while, but these days it doesn't seem like it really got anywhere. And perhaps we need a paradigm different and better from the one that and Google (and pretty much all other major search engines) use. Big centralized servers using bots to create their indexes to be accessed from a central point.

Even works this way. The only thing that makes it an "open search engine" is that it runs on Free Software and was initially supposed to listen to community input a lot (not sure how true that is today).

Maybe it's time to play with a decentralized search engine idea, the true internet way.

In fact, since you've put it so aptly in this article, it sparked a bit more clarity in the way I think of open standards - the way they actually make things not only accessible to more people and interoperable, but based on that ultimately simply easier to manage. It's easier because there are less barriers to cross. It essentially just works.

I feel that there could be a new sub-movement in motion.. one for restoring internet as it was, one that bans ICQ, AIM, MSN, Google,, Facebook and many other similar closed services as acceptable and instead promotes the use of completely functional (and even more powerful) free and open alternatives. ICQ/AIM/MSN -> Jabber, Google/ -> a yet to be developed decentralized search engine, Facebook -> Loaf/Crumb etc.

I want to participate in that movement. Smiling

This all sounds really


This all sounds really interesting, and I think I should link this article in my blog (the german one) in order to get a few more people to read it.
The problem, as with other closed software, is, as far as I see it, that most people really don't care. They use Facebook, Friendster, every day and are happy to be able to strip in public. Another thing that concers me here is that people on these services are willingly giving out a lot of information to the whole world. But ask them for that stuff on the street and they'll call the cops.
Where's people's sense of privacy today? Is it because "it's the internet"? Is it because "web 2.0 (I hate this expression) is so cool and everybody does it"? The latter one is another one of those sentences I like to reply to with "if your friends jump from the bridge then you also do it?". Really, there's better things everybody should do, instead of stripping in public by using those sites. For example using GPG/PGP, or encrypted IM (like with the OTR-plugin for Pidgin and Kopete). But since not everybody does it they would loose contact to their friends, etc...
That they could have some sort of "transition-time" where they use encrypted and unencrypted communication, until everybody finally got the point, doesn't seem to get to most people's minds.

I for example have started signing every single mail I send out. People who know what that is will probably appreciate that I do such thing. Others either won't care (neutral-bad), be scared away by the attached signature-file (bad) or ask me what the hell I'm doing there (good).

Somehow I really don't understand people today. They complain that the government wants the ISPs to keep the connection-records, but they post all their information for everybody else to see on the internet, sometimes including phone-number and/or postal-address.
I don't care that people know my real name and that I live in HK. So if anybody has a problem with me they are invited to come over and beat the shit out of me, but usually I don't write stuff that would piss people off (that much), so I don't think there's anything I need to be afraid of. People don't get my mail-address, they don't get my phone-number (I get enough spam-calls where I don't understand a word already) and they usually don't get my email-addresses (which also receive enough spam already).
But all those people freely give out at least their mail-address, and then complain that they get tons of spam. Often enough I have read phone-numbers too, which makes me shudder, really.

And then add to this lack of sense of privacy on the internet (as said, in their personal live it's different, just try asking a random girl on the street for her phone-number) these closed environments, which conserve data and do with it whatever they like. Can anybody be sure that they won't sell out your information? Can you be sure it's not all just a scheme of a bunch of terrorists in order to steal identities big-time?
Of course, there's the fine-print, which usually says if they would do something like that. But who reads that crap anyway?

And Taco, the last sentence you said so nicely, would then be a clear case of blackmail. The scary thing only seems to be that this really seems to be the direction their steadily drifting.

Well, enough confusing stuff for now...

While I don't exactly feel

While I don't exactly feel compelled to defend Google, I have to be fair. Google's text link ads are wrapped in javascript and can't be followed by bots, so they don't pass on page rank to sites it advertises. It is merely there for clicks, nothing else.

However, text link ads sold by a site owner are being followed by bots and do pass on page rank (this is how page rank works after all) so Google deems it abusing for people to buy text links to their sites and hence buying a portion of their page rank instead of getting them naturally.

And of course since Google controls the Page Rank algorithm and nobody else has access to it (not even to read and understand it), they can change it to fit their stand point on the issue they see as unfair.

You'd be correct, however, in at least one thing, that hitting on the text link buyers actually helps Google sell more adwords themselves, which is probably for them a more expensive way of gaining traffic, long term.

Personally, I see how the issue of text link selling and buying can be seen as manipulation, but I am yet to be compellingly convinced that it is a major enough hit on the relevance and usefulness of Google's results to warrant trying to kill the TLA industry. I've explained why this is so in a topic I recently wrote on digital point.

So that considered then indeed it is easy to start noticing how coming down on TLA industry helps Google's own business. If the other arguments don't carry as much weight, that's really all that's left. And that greatly goes in favour of points raised in Olivier's article.

I agree with you here,


I agree with you here, except that I do not have any clue how we should tackle the idea of a decentralized search engine. I clearly lack computing knowledge here to be able to propose an alternative - I merely hope that one day the business model of Google (having the poorly-ranking websites pay for ads) might turn against them, by lowering the quality of their search results.
For all things outside of search, I am sure we have all the possible answers close at hand. I merely wanted to suggest that, and perhaps crystallize things with a project name and blueprint. In a sense, I wish that less of the free software community would work on building one of 400 GNU/Linux distros, and more would participate on building projects such as Crumble/Loaf.

reptiler wrote: The

reptiler wrote:

The problem, as with other closed software, is, as far as I see it, that most people really don't care.

I'm afraid you'll find this is a problem with just about everything in life ;-) We just cannot all be concerned about every single cause. Simply talking/writing about it is a start, and things change progressively.

For info, I submitted the article to reddit =)


a bit of exaggeration


You are exaggerating a bit on how much people give out. Facebook actually does not make so much available. The only people that can see your profile (ever, not just by default) are your friends and network(s) (school, workplace, geographical region). You can restrict it further from there, such as making your phone number only available to your friends and not just anyone in your network (not sure what the default on that is).

Anyone working for Facebook could still look at everything. I trust Facebook with this. However, that does not mean that in the future they will not change, like another company *coughGooglecough*, and we should not take precautions by working on this decentralization.

a_thing wrote: However,

a_thing wrote:

However, that does not mean that in the future they will not change, like another company *coughGooglecough*, and we should not take precautions by working on this decentralization.

We shouldn't?

I'd say we should, and maybe that's what you actually meant. If not, I'm interested in an elaboration.



Whoops that was not very clear. I meant the last part of the sentence to be a double negative. We should take precautions by working on this decentralization.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.