The Freenet Project

Freenet Frequently Asked Questions

Misc. questions

Philosophical answers

What is Freenet?
Freenet is free software designed to ensure true freedom of communication over the Internet. It allows anybody to publish and read information with complete anonymity. Nobody controls Freenet, not even its creators, meaning that the system is not vulnerable to manipulation or shutdown. Freenet is also very efficient in how it deals with information, adaptively replicating content in response to demand. For more information, please read What Is Freenet.

Who is behind Freenet?
Freenet grew out of a design for an anonymous publication system created by Ian Clarke while a student at the University of Edinburgh, Scotland. Since then many other people have contributed towards making Ian's proposal a reality.

If authors are anonymous how can you trust information?
Cryptographic signing of information allows people to prove authorship, this technique is frequently used to authenticate authorship of emails. Moreover, you could actually sign information while remaining anonymous, thus having an anonymous persona. You could prove that you wrote different pieces of information on Freenet, without revealing your identity. In this way you could build up an anonymous reputation for reliability.

Do I have to donate disk space and bandwidth?
You aren't really donating in the sense that you lose the disk space and the bandwidth; but you aren't really sharing either (at least not the same way as with filesharing programs). It is more like pitching in to the common Freenet resource pool.

I don't have to donate anything when using filesharing application X and I get to leech more.
Do you get to do that anonymously? Freenet is designed with anonymity in mind, performance comes second.

All my friends donate very little space and bandwidth. Why should I donate more?
If you are happy with what you are getting then no. But if you want more you should consider donating more and running your node as close to 24x7 as possible, and you should ask your friends to do the same.

If I donate a lot will my experience improve significantly?
Your experience will definitely get better, but for a really great improvement we need more people to start thinking like you. Bandwidth counts more than diskspace.

Is Freenet legal?
We don't currently know of any prosecutions for using merely using Freenet. Some people claim that the DADVSI makes Freenet illegal in France; others have told us about the German data retention law which allegedly requires anonymity systems to retain enough data to trace web requests, which may or may not apply to us. Europe-wide, the IPRED2 may have wide-ranging effects, including on Freenet, should it pass... or it may not. Many of these things are arguable either way (depending on how broadly the legislation is applied) and will have to be decided in caselaw. The law can be an ass sometimes. You can read the EFF's (US-centric) advice to peer to peer developers here. If you need legal advice, talk to a lawyer. Also read the next section especially if you are in China; blocking the protocol may suggest the authorities don't like us!

Is Freenet blocked by national firewalls?
The Chinese national firewall (Golden Shield) has blocked our website for many years, and was observed in 2005 to block the 0.5 protocol as well. This suggests China doesn't like us, so be careful if you run Freenet in China. Some other countries (e.g. France) are known to be hostile to peer to peer, and may eventually force ISPs to block peer to peer networks (but right now Freenet works fine in France and we have many French users!). Freenet 0.7 has some minimal defences against blocking; the protocol is relatively hard to identify (we will make this a lot harder in future), and it supports a darknet mode (i.e. only connecting to your friends) which makes automated harvesting and blocking of nodes very difficult.

Can I get trouble if I run a node?
This is related to "Is Freenet legal?". We have done everything we can to make it extremely difficult for any sane legal system to justify punishing someone for running a Freenet node, and there is little precedent for such action in today's developed countries. Many legal systems recognise the importance of freedom of speech, which is Freenet's core goal. Having said that, there is risk in doing anything that your government might not agree with; you should make an informed decision as to whether to take that risk. Furthermore, your ISP or hosting provider may have a problem with Freenet. At least one French hosting provider has been known to ban Freenet (along with Tor and others) from their servers; please read your terms and conditions to make sure you are allowed to run Freenet.

What about copyright?
There are some excellent thoughts on this subject on the Philosophy page. Specific copyright-related laws may be a problem, please read Is Freenet legal? and Is Freenet blocked by national firewalls?.

What about child porn, offensive content or terrorism?
While most people wish that child pornography and terrorism did not exist, humanity should not be deprived of their freedom to communicate just because of how a very small number of people might use that freedom.

I don't want my node to be used to harbor child porn, offensive content or terrorism. What can I do?
The true test of someone who claims to believe in Freedom of Speech is whether they tolerate speech which they disagree with, or even find disgusting. If this is not acceptable to you, you should not run a Freenet node. There is another thing you can do. Since content in Freenet is available as long as its popular, you can help limit the popularity of whatever information you do not like. For example, if you do not want a file to spread you should not request it and tell everyone you know not to request that specific key. However, keep in mind that freenet is not designed so as to only allow communication between people if a sufficient number of people agree with the communication. Freenet is designed to make communication possible even if there's just one publisher and one reader, and this is already reasonably feasible on the current freenet.

How about encryption export restrictions?
The Freenet Project has notified the US authorities (since the files are hosted on SourceForge, which is on US soil) that it will be exporting crypto. As long as your country doesn't prohibit the use of encryption you are fine.

I have nothing to hide and don't need anonymity. Is there anything else Freenet can offer?
Yes, in fact even without the anonymity feature Freenet is very useful because of its unique way it handles content distribution and information load. In simple terms that means you can publish a website without worrying about how big the site will be and without having to put someone elses ad banners on it. While it is unlikely that freenet sites will ever load faster than regular websites, it does adapt to sudden surges of visitors (which will often occur when relatively unknown sites get linked to from a big site) better, and high download speeds for big files are feasible too. Just don't expect very low latency.

Technical answers

How do I use this software? I downloaded it, but when I run it there's no GUI.
Fred (the Freenet REference Daemon) runs as a daemon, or service, in the background. You normally talk to it through a Freenet client. One built-in client is fproxy, which lets you talk to Freenet with a web browser. Freenet should have installed a Browse Freenet shortcut on the desktop and/or the start menu. This should launch a customised firefox profile optimised for browsing Freenet (with a black theme so you can see clearly that you're not browsing the web). Failing that, point your web browser to http://127.0.0.1:8888/ for the gateway page. Try clicking the various links in the "Bookmarks" panel to reach some of the popular Freenet index sites.

If you're looking for applications that run on top of Freenet and provide a different interface or functionality, please see the Tools page.

Why is Freenet so slow?
When you first install Freenet, it will be slow, and you may see Data Not Found or Route Not Found errors for freesites. This is normal, and Freenet will speed up significantly over time. For best performance you should try to run Freenet as close to 24 hours a day as possible. This is why we install Freenet as a service.

Please bear in mind that Freenet is inherently high latency: it can take a while to (for example) load a page for the first time, even if it is capable of reasonable speeds (as anonymous systems go!) for large popular files. The Browse Freenet icon/menu item should launch a customised version of Firefox which will use many connections to maximise performance especially when you open many freesites in tabs. If that doesn't work then see here. And of course, protecting your anonymity does cost a certain amount of performance.

Is Freenet searchable?
Yes, there are a few different search mechanisms. To search the freenet web, go to the plugins page, load the XMLLibrarian plugin if it isn't already there, and use that. Alternatively, Frost and Thaw also provide searching for messages and files. Note that searching on Freenet is a good deal more difficult than on other networks because of Freenet's different architecture and design goals.

How do I get freenet working with a Firewall/NAT?
Mostly, Freenet should just work with a NAT. However, you should forward the ports manually if you can. Click on the Friends or Strangers page. At the bottom you will see a list of ports used by the node. You should forward (for UDP) the Darknet FNP and Opennet FNP ports. You may need to look up your router's documentation to figure out how to do this. Freenet should have forwarded them itself through Universal Plug and Play, but this doesn't always work (and it never works if you don't have the UPnP plugin loaded, or have one router behind another).

If you have a dyndns address or other domain name pointing to the computer you run your Freenet node on, tell the node about it. Go to the config page, and find the option "IP address override". Put your domain name in that box, and apply the settings.

Do I need a permanent connection to run a node?
No, but it is preferred. You can run the software and test it from a "transient" connection (such as provided by typical modem/ISP setups), but for the network as a whole to be most useful, we will need as many permanent nodes as possible (most cable modem or DSL setups are sufficiently "permanent" for this). A later version of Freenet may take better advantage of transient nodes.

Why does Freenet only download 1 or 2 files at a time?
Many browsers limit the number of simultaneous connections to something far too low for efficiently browsing Freenet (since Freenet pages often have much higher latency than web pages). This can usually be reconfigured. For example, for Mozilla Firefox, type about:config in the address field of the browser and replace the value of the following settings to the one stated. Filter on "connections" to get only the relevant settings:


network.http.max-connections 200

network.http.max-connections-per-server 200

network.http.max-persistent-connections-per-proxy 200

network.http.max-persistent-connections-per-server 200

Note that these settings will cause mozilla to use more connections for all your browsing, which may not be desirable from a network congestion point of view; volunteers to make mozilla allow this sort of settings to be set per host would be welcome...

Note also that if you have installed Freenet recently, it should have created a Firefox profile for browsing Freenet with similar settings, with a black theme, which you can access through the Browse Freenet icon.

Why can't Freenet store data permanently?
Because we can't find a way to do this without compromising Freenet's other goals. For example, people often suggest that someone's node could just never drop data they want to cache permanently. This, however, won't work because even if the data is still available on their node, there is no way to ensure that requests for that data will be routed to that node. We have considered many other ways that Freenet could store data permanently, but they either won't work, or compromise Freenet's core goals of anonymity, and scalability. However, if you simply reinsert your data regularly, or if it is regularly downloaded (not by the same person as it would be cached on their node and the path to their node), it should remain available indefinitely.

Why is Freenet implemented in Java?
Opinions differ about the choice of java for the reference implementation of freenet (even among the core developers). Ian Clarke and several other developers are java proponents and the choice for java was made. Even if everybody could be convinced to switch to a different language reimplementing the current freenet protocol would be quite a big task, and take up a significant amount of time, while there is only a limited amount of developer-time available. Flame wars on the development list about the language choice aren't welcome, people willing to implement freenet in other languages however are very much encouraged to try. Don't underestimate the amount of work however.

How do I allow connections to FProxy from other computers?
If you want everyone to be able to use your node you have the following options:

Go to fproxy's configuration page and enable advanced mode in the FPROXY section
Stop your node and edit freenet.ini manually

In both cases change the following parameters:

If you want everyone to be able to access your node:


fproxy.bindTo=0.0.0.0

fproxy.allowedHosts=*

Of course, this leaves your node wide open, unless you control access with a firewall of some sort. If you'd prefer to use access controls within Freenet, then you can use lines like this:


fproxy.bindTo=0.0.0.0

fproxy.allowedHosts=127.0.0.1,192.168.1.0/24

Or even (find your IP address from ipconfig/ifconfig/winipcfg and substitute it for 192.168.1.1):


fproxy.bindTo=127.0.0.1,192.168.1.1

fproxy.allowedHosts=127.0.0.1,192.168.1.0/24

And if you want to grant full access (i.e. change config settings, restart, etc) to the node (WARNING: Be very careful who you give full fproxy access to!): fproxy.allowedHostsFullAccess=127.0.0.1,192.168.1.0/24

Problems installing with OpenJDK

Some versions of OpenJDK, particularly the one included with Ubuntu 8.04, have some problems with SSL which cause the installer to fail. Please install the Sun JRE, at least version 5. On Ubuntu or Debian, open a root terminal and type:

apt-get install sun-java6-jre
update-java-alternatives -s java-6-sun

What is FEC?
FEC stands for Forward Error Correction. When large files are inserted into Freenet, they are split into many small blocks -- this is called a splitfile. FEC adds redundant check blocks to a splitfile, so that if some of the blocks fall out of the network or can't be found, you might still be able to retrieve enough of the file to reconstruct it.

What is healing?
When you download a splitfile that uses FEC (see previous question), some of the blocks may have become unavailable. Healing lets you repair the file by recalculating the data for the missing blocks, and then reinserting them into Freenet.

What is a Heisenbug?
The name comes from the Heisenberg Uncertainty Principle of physics, which states that when measuring both the momentum and the position of a particle, the product of the uncertainties is a constant. In Freenet, the Heisenbug refers to a problem which plagued the developers for a long time and was extremely difficult to reproduce.

The current theory is that the Heisenbug is actually a bug in certain Java virtual machines, particularly IBM's, which causes certain calculations to return the wrong results; but these results match a certain pattern. Whenever your node sees this pattern, it reports that the Heisenbug may have occurred, and writes verbose information to the logs. Since this pattern can also occur under normal circumstances (about once in every 65,000 times), a few Heisenbug reports may be expected. But if you're getting lots of them, something's wrong.

Update: It's turned out that the heisenbug was actually a bug in the fred code, there was a bug in our implementation of Yarrow, a piece of code should have been synchronized (a java method of locking) but wasn't. If you're still seeing heisenbugs being reported in your logfile and are running a current version of freenet, please report them to the mailinglist.

What's new? Is there a changelog?
On every new build, a brief summary of all the main changes is posted to the support and devl lists and the freenet-announce board on Frost. Alternatively for a much more detailed view, check the SVN logs.

Why are there so many messages in my logfile with a backtrace attached?
Fred (and freenet in general) are still very much in development, and if something goes wrong it's worthwhile to know exactly what went wrong.

How can I change from stable to unstable?
There is no unstable network anymore.

Publisher answers

If I publish something in Freenet, how will people find it? Don't they have to know the key I used?
Yes, people will have to know what key you used to publish your information. This means you will have to announce your key in some way.

The most common way to do this is to send a message, containing your key and brief description of your information, to the author of one of the existing Freenet sites. Most of the "portal" sites which are linked from the Freenet web interface (fproxy) have a submission form which lets you send them messages anonymously. You could also send your key to people by using the Freenet mailing lists, in the IRC channel (irc.freenode.net #freenet), by private e-mail, or by advertising your Freenet site on your World Wide Web site. If you're feeling extravagant, you could even try skywriting it. (Graffiti is not recommended, for legal reasons.)

How do I publish a Content Hash Key (CHK)?
A Content Hash Key is based on the actual content contained within it - and as such, the key will only be known after it has been inserted into Freenet. To insert a CHK, simply insert it as "CHK@", Freenet will tell you what the actual CHK is once the insertion completes.

Can Freenet documents be updated / deleted?
Currently, a document posted to Freenet with the same name as one already present may actually serve to propagate the existing document. There is also currently no means of deleting a document from Freenet. Documents that are never requested are eventually removed through disuse.

However, you can use an Updatable Subspace Key (USK) to provide a form of updatable freesite: your node will automatically look for later editions of the site (after you visit it, or always if you bookmark it), and show you the latest version.

Contribution answers

I have this great idea....
Good! First step: read the mailing list archives. Odds are good that someone else had the same idea and discussed it with the group. Either a flaw was found in the idea, or perhaps it was decided to postpone implementing the idea until later. Some examples of ideas already discussed are storing information by content hash, key redirection, signed keys/data, use of UDP, server discovery, URLs, document versioning, and others. If you don't see the idea discussed in the archives, by all means bring it up in the appropriate mailing list.

Can I contribute to the Freenet Project?
Absolutely. Even if you don't have the time or skills to become a co-developer of the project, you can contribute in other ways:

Help test Freenet by installing and configuring the server software on your machine.
Install the client software on your machine to test retrieving information and publishing your own.
Work on the Freenet web site (including the FAQ).
Contribute your ideas to the discussion lists.
Translate the user interface into your local language.

If you are a developer, you can also help by designing and developing other applications to run on Freenet.

People that want to contribute to Freenet in any way, by contributing artwork, new ideas, or even correcting spelling/grammar mistakes or ideas for adding new pages/paragraphs on this site, are also welcome to email.

How can I access the code and website?
We are using a subversion server : Freenet's subversion repository.

What tools do I need to help develop?
To build and deploy the Freenet server, you will need Java tools compatible with Sun's JDK 1.5 or later. To retrieve and update the code at http://freenet.googlecode.com/svn/trunk/freenet/using subversion, (This is not necessary if you only want to download, compile, and run the server without contributing to its code). Further instructions for building and deploying the server are included with the code itself. Generally speaking, joining our IRC channel is a good idea: #freenet on irc.freenode.net

Is there a Help Site that goes deeper into the questions newbies may have about Freenet, and where people can contribute too?
You could have a look at our wiki. There is also an older, largely unmaintained Freenet Help Site wiki. You can contribute to both, since they are wiki's so most pages are editable.

Where can I report bugs?
You can use our bug tracking system or send a mail to our development mailing list.

Security answers

Can I use my regular browser to browse Freenet?
Freenet has a web interface: much of the content on Freenet is in the form of "freesites", downloads, configuration and friend connections can be managed from the web interface. However, because of weaknesses in current browsers, we strongly recommend that you use a separate browser for Freenet. Specifically, browser history stealing, in all its forms, is a major threat if you share a browser between Freenet and the WWW at large: malicious web pages will be able to probe which freesites you have visited, and report this information to their owners.

Won't attack X break Freenet's anonymity?

Short answer: Probably yes.

Long answer:

Freenet has a different threat model to Tor and the Mixmaster remailers. Freenet is designed to resist censorship: The network must therefore be robust, and content must be distributed without requiring a central server, whether anonymous or not. Anonymity is important for requesters and especially for those who upload content in the first place. The typical example is a corporate or government whistleblower. Generally to find the originator of some content, the attacker must be able to predict the data in advance, must be able to move across the network relatively quickly, and must be able to perform the attack while the data is being inserted; after that, it is distributed across the network and is much harder to trace, and the originator may have left the network. However, if by chance or by overwhelming force the attacker is connected to the whistleblower (or e.g. seizes the computers of everyone on the network), he may be able to identify this much more quickly. All of this is vastly more difficult on a darknet, where everyone connects only to their friends, where it is very hard for an attacker to find nodes, and where to connect to a given node he must social engineer its operator! Freenet does support opennet mode (plug and play), but darknet is far more secure, and additionally is far more difficult to block on a national firewall.

Tor on the other hand is designed to anonymise real-time data streams, on the assumption that the list of nodes can be public, that there is a free world where nodes can be operated safely, that the authors of controversial content will be able to either host (hidden) web servers themselves or upload it to other (hidden, but usually centralised) storage systems, and so on. And Tor has a concept of a "client", which is somebody who uses the service without providing any value to it; on Freenet, every node relays data for its neighbours. Hence the attacks on Freenet are completely different to the attacks on Tor. Both compromise to some degree to enable more or less real-time performance.

If you can use the darknet, trust your friends, don't reinsert files or insert easily predictable data (this requirement will be relaxed in 0.9), and change your anonymous identity after some volume of inserts, you should be relatively safe using Freenet. If you can connect, build up some trust in your anonymous persona, insert your controversial content, and then disappear, again, you are better off with Freenet, especially if the content is a website. In some other cases, Tor is better.

In Freenet 0.9, we will add a form of cryptographic tunnels, somewhat similar to Tor's onion routing; this should greatly reduce the impact of many of the below attacks.

Major known attacks:

In the interests of giving would-be users as much information as possible, and on the assumption that any serious attacker would do their homework, here are the major classes of attack on Freenet we are presently aware of:

Harvesting: Simply by running some powerful Freenet nodes, an attacker can identify most of the opennet (Strangers network) relatively easily. These nodes can then be attacked one by one (subject to resources), their traffic analysed, or simply be blocked on a national firewall. Connecting only to friends largely solves this problem. It is still possible for ISPs to identify nodes with traffic flow analysis. Obviously a large network will make this harder. And this attack is trivial on Tor, although they provide a means to work around it.
Datastore attacks: Everything any node requests through your node is cached locally. Everything you fetch is also cached locally (we may make this configurable at some point, but turning it off enables an easier attack). If the attacker can obtain your hard disk, or connect to your node and probe your datastore by requesting keys and timing how long they take, he may be able to determine what you have been downloading from/uploading to Freenet. Right now the best way to prevent this is to only connect to your trusted friends, as above, and encrypt your hard disk. Once we have tunnels, we will not need to cache locally requested keys in the datastore, which will solve this attack.
Correlation attacks: If you are connected to a node, and can recognise the keys being requested (probably because it was posted publicly), you can show statistically that the node in question probably requested it, based on the proportion of the keys requested from that node, the locations of nearby nodes, the HTL on the requests and so on. This will again be largely eliminated by tunnels.
Adaptive search: If you want to find the author of some content, and you can predict the content or the keys to be inserted, you can listen out for their inserts. Each request gives you a data point suggesting roughly where the originator might be (on the location space), and on each insert you intercept you can move slightly closer to the originator, by obtaining connections to nodes which are closer to your current guesstimate of the target's location. If you are right these new connections will yield more of the target's requests/inserts, and you can rapidly close in on the originator. The best defense right now is darknet (connecting to friends only): Darknet makes it very difficult for an attacker to obtain new connections closer to his guess of where the target is. The tunnels proposal will likely make this attack considerably more difficult. Also, if the attacker cannot guess the content to be inserted in advance, this is very much more difficult (hence reinserting big files is bad); we will randomize the insertion keys in 0.9 to give additional protection for inserters. And of course this is of no use for tracing the author if the author has finished the insert and left the network: you have to intercept the insert while it is in progress.
Traffic analysis: Freenet provides minimal protection against global traffic analysis (basic message padding etc); if the attacker also has nodes on the network, the extra data will likely be helpful. However on Tor-style networks, global traffic analysis will defeat the network completely: all that is needed is to observe both the entry and exit points.
Swapping attacks: It is possible to attack the location swapping algorithm, and thereby disrupt routing. This has been demonstrated by the authors of the Pitch Black paper. We are working on a solution, but this only affects the darknet (Friends network), which presently is a small proportion of the overall Freenet.

More information on the current practical state of Freenet security is available here.

Is Freenet vulnerable to flooding attacks?
Short answer: no.

Long answer:

We don't think so. Aside from protecting freedom of speech, Freenet is also designed to be an efficient dynamic caching system. If information is requested a lot from a limited number of nodes, the nodes that the requests pass through will cache the information, lowering the load on the network. If information is inserted on a limited set of nodes and then subsequently requested a lot from a separate set of nodes, with repetition, the sets will close in on one another in the network topology until they are "neighbors" and only the originally targeted nodes are suffering from the attack.

In other words, in order to harm Freenet with a flood you need to consistently change your point of entry into the network and continually insert and request new data, and you will still only increase the workload for the network that is linear to your own. Given an immense will and capacity greater than the total of the entire network, it is possible to cripple any public network (including the Internet itself) with floods, but it is our intention to always keep Freenet as resistant to this as theoretically possible.

Curiously enough, the above analysis only applies to Opennet. On Darknet, you might have a little more success, although it would be much harder to change your entry point in any significant way. Nonetheless, you have a reasonably low bandwidth multiplier (the total number of nodes visited, around 20 on average), and you are severely limited by the number of nodes you can connect to, which will be low on a darknet.

Why hash keys and encrypt data when a node operator could identify them (the data) anyway if he tried?
Hashing the key and encrypting the data is not meant a method to keep Freenet Node operators from being able to figure out what type of information is in their nodes if they really want to (after all, they can just find the key in the same way as someone who requests the information would) but rather to keep operators from having to know what information is in their nodes if they don't want to. This distinction is more a legal one than a technical one. It is not realistic to expect a node operator to try to continually collect and/ or guess possible keys and then check them against the information in his node (even if such an attack is viable from a security perspective), so a sane society is less likely to hold an operator liable for such information on the network.

What about hostile "cancer" nodes within the network?
The existence of malicious nodes within the network is the most difficult problem that a distributed network must face, and has been the bane of many previous ideas. Many systems (such as multiplayer gaming networks) try to avoid malicious nodes by keeping the protocol and code closed, but we have yet to see an example of that working in the long run. And anyway it is opposed to Freenet's philosophy.

Freenet is based on a balance of positive and negative feedback loops that bring requests for information to a node when it is functioning well, and keep requests away from it when it is not. The key to avoiding "cancers" is (as in the body) to make sure these loops can correctly identify even the most carefully designed malicious node and not keep sending requests to it. This issue is not fully dealt with by the current test code, but you can rest assured that a number of possible solutions have been on the table and discussed for some time now. Several have been implemented (enforcing hashes or signatures on content, per node failure tables, backing off from a node that causes timeouts ...)

What about attack Y?
Freenet is still in testing and there are bound to be attacks found that we have not dealt with yet. So if you do manage to figure out a truly new kind of attack, we are interested in hearing about it. Please keep in mind what Freenet is and what it is not, however. No single network can offer everybody everything, and there are security issues that Freenet, by it's nature, may not deal with to extent you might wish. If this upsets you, all of our code is freely available, so you are free to take as much of it as you like and write your own distributed network that suits your desires.

Where can I get a high quality copy of the rabbit icon?
Here.

Freenet Frequently Asked Questions

Philosophical questions

Technical questions

Publisher questions