These days, people report success with rather large files. Linux CD ISOs have been transferred around, though at present downloading a DVD ISO would present difficulties. Data rates are not awful, but they're not exactly fast either. Fetching older content may be slow or unreliable...
You use it because you're curious, or want to support free speech. Adding to the userbase and content available helps the network grow, and helps those who actually need it. There are plenty of people who need it or think they need it even though their government isn't out to get them -- for example, there's at least one freesite by a victim of abuse who doesn't appear to be particularly comfortable talking about it in other forums. There are also plenty of conspiracy theorists who seem to think they need it -- I think they're wrong, but who knows? Not for me to judge. I'm sure there are some people using it as a route for "normal" copyright infringement that's secure from the RIAA et al, though that usage is discouraged.
If you don't like darknets, don't turn them on. I think you're wrong, but I won't bother refuting that point here. Freenet 0.7 gives you the choice of darknet mode, opennet mode, or a mix. As a corollary, there aren't discrete "darknets" but rather one large network with a mix of darknet and opennet connections (for the most part; there may be a handful of small poorly-connected darknet subnets).
I do not recall any freenet developer talking about implementing any sort of blocking; nor have they done so. Unless you can back up that statement, I will be forced to conclude you are trolling. As you say, the ability to block anything, no matter how abhorrent, implies the ability to censor valid political speech and is therefore a bad thing for a network like freenet.
Also, I suggest you try out FMS as a replacement for Frost / Thaw; it is far more spam resistant for a variety of reasons.
I really don't understand this continued bashing of 0.7; now that it has implemented a proper opennet feature, with the ability to turn off the darknet option, what is the complaint?
Note that *because* it doesn't care what the content is that it's routing through your node, your should be protected by the various safe harbor laws in most jurisdictions.
It should reach usable performance quite quickly (a few minutes or a little more; if it's taking more than 15 or so, you may have something not working). Performance will continue to improve over the next several hours, though likely only somewhat.
Yes, it is visible, if your ISP can do enough logging. However, on an opennet (freenet in opennet mode, any other p2p app), you can run a program that spiders the network looking for nodes -- you don't even need to be able to sniff anyone's traffic! New nodes have to get inserted somehow, and that process is inherently abusable to find people running the application. Darknet isn't perfect, but it's a *vast* improvement. Some later release of Freenet (0.8.0 or maybe 1.0) should support steganographic transports -- eg making Freenet connections masquerade as Skype connections, or streaming RealMedia, or anything else that someone cares to implement. At that point, it becomes actually difficult to uncover; for now, the darknet mode is pretty good, though.
In most jurisdictions, being friends with bad people is not sufficient evidence in itself for a warrant; the authorities need some other evidence. I don't see why moving to Freenet instead of hanging out in coffee shops or whatever would change that.
There's an opennet. It's connected to the darknets, which is largely how they're connected to each other. In most cases, the approach should be to enable opennet, and then slowly replace the opennet connections with darknet ones as you find friends who run freenet (or convince them to). In locations where running freenet may be sufficient to cause you legal problems, life is harder, because it's relatively easy to find out that someone is running opennet.
I encourage you to download the installer and try it out. It should ask if you have any friends running freenet; tell it no, and that you'd like it to connect automatically, and everything should Just Work (tm). And if it doesn't, come join us on IRC...
No, you do host things requested by your peers as well. However, since the process is content-neutral, in most jurisdictions (including the US) you should have safe harbor laws working in your favor as long as you aren't the requester.
Also, as far as I can tell, as a regular freenet user, the pedophiles are a very small minority, and you won't normally encounter them in your freenet activities.
As in you think toad (yes, that's the same poster) should be developing alternate implementations or rewriting in C? Or you think that someone else, like you, should be? If the latter... Why aren't you? If the former, I respectfully disagree.
What main index pages? All the default bookmarks have anti-CP policies. This is not even a result of editing by the freenet devs; it's a result of community standards -- all the well-maintained and usable indexes have such policies. The devs have explicitly taken a content-agnostic approach to the default bookmarks, and said that anything useful and regularly updated is a candidate. The result is a set of indexes free of child pornography.
Exactly. And if you *do* connect to the rest of the network in a few places, but not much, and none of you request CP -- then you can spread your message to the rest of Freenet, but routing won't take the long circuitous path that goes through your somewhat disconnected subnetwork when it comes time to route other people's requests. Or, looked at another way: the stuff on your node will be what you're requesting, to a lesser extent what your friends are requesting, to an even lesser extent what their friends are requesting... If your friends are requesting things you don't object to, you shouldn't be storing much if any objectionable content.
There are two issues here. One is that the network isn't as robust as would be ideal; there are legitimate concerns about buggy implementations causing problems. A lot of the work debugging freenet goes into things that are essentially emergent behavior, and the bugs get even harder to track down on a non-homogenous network.
The second is one of documentation. Yeah, it's practically nonexistant outside of the source code. But my impression from discussions (none recent) of alternate implementations was that the developers would be willing to support them by answering questions and such, and had no actual objections (concerns about buggy clients, yes, but not objections). There has been discussion of people creating alternate implementations, but so far no one has actually followed through. So, if you want to go write one, I suggest you start writing some code and posting questions to the mailing list or on IRC.
I speak here as a #freenet regular and a coder, though not a freenet developer.
Freenet 0.7 is vastly faster than 0.4, though not as fast as bittorrent (obviously). Currently, all the good index sites have anti-CP policies. They'll happily link photos from Tibet, though, or wikileaks mirrors (both present). The current crop of index sites also tends to do a good job indexing things. Also, much of the content is centered around FMS and the (less functional) Frost messaging systems (broadly similar to usenet; FMS even operates as an NNTP gateway, allowing you to use your favorite newsreader). You'll get content posted to boards you subscribe to, which tends to be at least somewhat relevant (ie, the signal to noise ratio is probably better than/.;) ). I'd encourage you to try it out again, if you're interested in privacy and an anonymous network, but not if all you're looking for is the next bittorrent (though you can find music, movies, etc on Freenet if you want).
Wikileaks has been mirrored to Freenet more than once. I don't know of an up to date link, or a single regularly updated source, but it's there.
A large number of photos from Tibet are available, and there is at least one highly active user posting them and keeping them up to date, with commentary.
Because, while I like a variety of forms of renewable energy and think they should be supported far more than they are, I realize that it will take a while before they can provide a substantial fraction of our energy needs. The same is not true of fossil fuels, and is true to a much lesser degree of nuclear plants. I didn't say to ignore alternative options; I was simply pointing out that coal is worse than nuclear in a wide variety of ways. The parent was comparing nuclear and coal, and I added a counterpoint to his argument. I did not feel a particular need write a treatise intricately comparing the pros and cons of all plausible energy sources.
Not everyone who fails to trumpet the virtues of renewable sources in every post is a troll, or even dislikes renewables.
Another tip: even when working with simple circuits, set interesting challenges for yourself. For example, I built the basic multivibrator circuit... and then decided to see how low I could get the power consumption. That makes for a surprisingly tough challenge, especially if you want to keep using a cheap multimeter as your basic instrument. Measuring current consumption when all your ammeter reads is zero will force you to start thinking carefully about your tools.
Huh? What bunker and satellite? Look at these pictures: we're talking about medium-sized RC planes, except with a camera on board and computers and such. Surely you don't think maintenence and such are free on a large plane?
Alternately, I could work on the code I know, you can work on the code you know, and the OpenOffice developers can work on the code they know. We all pay attention to user requests, and then we don't have to all go learn a new codebase every time we find a program that's missing a feature. Much more efficient that way, don't you think?
If you're idly curious, I'd recommend the corresponding Wikipedia articles. They're pretty good. If, on the other hand, you're the poster or someone else interested in really learning the stuff, I recommend you turn to the textbook that forms another major piece of your educational material (the final major piece being actual hardware that you experiment with and occasionally let the magic smoke out of). The circuit without the theoretical basis won't teach you much, but it is a tremendously important addition. Seeing how current mirrors are put to use in a practical circuit will give you a far better understanding of them than just seeing them in the textbook.
The 741 (even though it's a somewhat odd design by modern standards) makes an interesting first non-trivial transistor circuit -- but you should certainly have a basic understanding of how a 1-transistor amplifier works before moving on to any nontrivial transistor circuit. I suppose I wasn't entirely clear, but my post was intended as a second step in the learning process; other people have offered excellent suggestions for introductory texts.
If you want to learn the stuff, but not really "study" it, I suggested some parts to get. Buy them, along with a solderless breadboard, some 22-gauge hookup wire, a voltmeter, some jumper clips, and go build yourself a circuit or three. Looking for something simple? Start by building a multivibrator that blinks a pair of LEDs. Or try a single-transistor microphone amplifier (a cheap microphone intended for eg walkie-talkie use will cost perhaps $1-2 at radio shack; less at Digikey. The 2N3904 NPN becomes your gain stage, with some coupling capacitors and biasing and gain-setting resistors, and it should easily drive headphones or a small speaker (though be careful with headphones: letting the magic smoke out with them connected would probably be loud). Heck, you could build both of those circuits combined for $10 in parts at radio shack plus a breadboard (<$10 for a small one) and a multimeter ($20 or less for a cheap but versatile digital model), and power them off a 9V battery. (Note that at least the radio shack near me stocks all the part numbers I listed, but you'll find it cheaper to go to Digikey if you're planning to buy more than a couple random transistors.)
Slashdot Mirror
These days, people report success with rather large files. Linux CD ISOs have been transferred around, though at present downloading a DVD ISO would present difficulties. Data rates are not awful, but they're not exactly fast either. Fetching older content may be slow or unreliable...
You use it because you're curious, or want to support free speech. Adding to the userbase and content available helps the network grow, and helps those who actually need it. There are plenty of people who need it or think they need it even though their government isn't out to get them -- for example, there's at least one freesite by a victim of abuse who doesn't appear to be particularly comfortable talking about it in other forums. There are also plenty of conspiracy theorists who seem to think they need it -- I think they're wrong, but who knows? Not for me to judge. I'm sure there are some people using it as a route for "normal" copyright infringement that's secure from the RIAA et al, though that usage is discouraged.
If you don't like darknets, don't turn them on. I think you're wrong, but I won't bother refuting that point here. Freenet 0.7 gives you the choice of darknet mode, opennet mode, or a mix. As a corollary, there aren't discrete "darknets" but rather one large network with a mix of darknet and opennet connections (for the most part; there may be a handful of small poorly-connected darknet subnets).
I do not recall any freenet developer talking about implementing any sort of blocking; nor have they done so. Unless you can back up that statement, I will be forced to conclude you are trolling. As you say, the ability to block anything, no matter how abhorrent, implies the ability to censor valid political speech and is therefore a bad thing for a network like freenet.
Also, I suggest you try out FMS as a replacement for Frost / Thaw; it is far more spam resistant for a variety of reasons.
I really don't understand this continued bashing of 0.7; now that it has implemented a proper opennet feature, with the ability to turn off the darknet option, what is the complaint?
Note that *because* it doesn't care what the content is that it's routing through your node, your should be protected by the various safe harbor laws in most jurisdictions.
It should reach usable performance quite quickly (a few minutes or a little more; if it's taking more than 15 or so, you may have something not working). Performance will continue to improve over the next several hours, though likely only somewhat.
Yes, it is visible, if your ISP can do enough logging. However, on an opennet (freenet in opennet mode, any other p2p app), you can run a program that spiders the network looking for nodes -- you don't even need to be able to sniff anyone's traffic! New nodes have to get inserted somehow, and that process is inherently abusable to find people running the application. Darknet isn't perfect, but it's a *vast* improvement. Some later release of Freenet (0.8.0 or maybe 1.0) should support steganographic transports -- eg making Freenet connections masquerade as Skype connections, or streaming RealMedia, or anything else that someone cares to implement. At that point, it becomes actually difficult to uncover; for now, the darknet mode is pretty good, though.
Yes there are. I won't go looking for them for you, but google will find them. I assume most are hosted outside the US.
In most jurisdictions, being friends with bad people is not sufficient evidence in itself for a warrant; the authorities need some other evidence. I don't see why moving to Freenet instead of hanging out in coffee shops or whatever would change that.
There's an opennet. It's connected to the darknets, which is largely how they're connected to each other. In most cases, the approach should be to enable opennet, and then slowly replace the opennet connections with darknet ones as you find friends who run freenet (or convince them to). In locations where running freenet may be sufficient to cause you legal problems, life is harder, because it's relatively easy to find out that someone is running opennet.
I encourage you to download the installer and try it out. It should ask if you have any friends running freenet; tell it no, and that you'd like it to connect automatically, and everything should Just Work (tm). And if it doesn't, come join us on IRC...
No, you do host things requested by your peers as well. However, since the process is content-neutral, in most jurisdictions (including the US) you should have safe harbor laws working in your favor as long as you aren't the requester.
Also, as far as I can tell, as a regular freenet user, the pedophiles are a very small minority, and you won't normally encounter them in your freenet activities.
As in you think toad (yes, that's the same poster) should be developing alternate implementations or rewriting in C? Or you think that someone else, like you, should be? If the latter... Why aren't you? If the former, I respectfully disagree.
What main index pages? All the default bookmarks have anti-CP policies. This is not even a result of editing by the freenet devs; it's a result of community standards -- all the well-maintained and usable indexes have such policies. The devs have explicitly taken a content-agnostic approach to the default bookmarks, and said that anything useful and regularly updated is a candidate. The result is a set of indexes free of child pornography.
Exactly. And if you *do* connect to the rest of the network in a few places, but not much, and none of you request CP -- then you can spread your message to the rest of Freenet, but routing won't take the long circuitous path that goes through your somewhat disconnected subnetwork when it comes time to route other people's requests. Or, looked at another way: the stuff on your node will be what you're requesting, to a lesser extent what your friends are requesting, to an even lesser extent what their friends are requesting... If your friends are requesting things you don't object to, you shouldn't be storing much if any objectionable content.
There are two issues here. One is that the network isn't as robust as would be ideal; there are legitimate concerns about buggy implementations causing problems. A lot of the work debugging freenet goes into things that are essentially emergent behavior, and the bugs get even harder to track down on a non-homogenous network.
The second is one of documentation. Yeah, it's practically nonexistant outside of the source code. But my impression from discussions (none recent) of alternate implementations was that the developers would be willing to support them by answering questions and such, and had no actual objections (concerns about buggy clients, yes, but not objections). There has been discussion of people creating alternate implementations, but so far no one has actually followed through. So, if you want to go write one, I suggest you start writing some code and posting questions to the mailing list or on IRC.
I speak here as a #freenet regular and a coder, though not a freenet developer.
Freenet 0.7 is vastly faster than 0.4, though not as fast as bittorrent (obviously). Currently, all the good index sites have anti-CP policies. They'll happily link photos from Tibet, though, or wikileaks mirrors (both present). The current crop of index sites also tends to do a good job indexing things. Also, much of the content is centered around FMS and the (less functional) Frost messaging systems (broadly similar to usenet; FMS even operates as an NNTP gateway, allowing you to use your favorite newsreader). You'll get content posted to boards you subscribe to, which tends to be at least somewhat relevant (ie, the signal to noise ratio is probably better than /. ;) ). I'd encourage you to try it out again, if you're interested in privacy and an anonymous network, but not if all you're looking for is the next bittorrent (though you can find music, movies, etc on Freenet if you want).
Wikileaks has been mirrored to Freenet more than once. I don't know of an up to date link, or a single regularly updated source, but it's there.
A large number of photos from Tibet are available, and there is at least one highly active user posting them and keeping them up to date, with commentary.
Because, while I like a variety of forms of renewable energy and think they should be supported far more than they are, I realize that it will take a while before they can provide a substantial fraction of our energy needs. The same is not true of fossil fuels, and is true to a much lesser degree of nuclear plants. I didn't say to ignore alternative options; I was simply pointing out that coal is worse than nuclear in a wide variety of ways. The parent was comparing nuclear and coal, and I added a counterpoint to his argument. I did not feel a particular need write a treatise intricately comparing the pros and cons of all plausible energy sources.
Not everyone who fails to trumpet the virtues of renewable sources in every post is a troll, or even dislikes renewables.
So, have you looked at how much radioactive waste comes from burning coal? Hint: it's not zero.
Another tip: even when working with simple circuits, set interesting challenges for yourself. For example, I built the basic multivibrator circuit... and then decided to see how low I could get the power consumption. That makes for a surprisingly tough challenge, especially if you want to keep using a cheap multimeter as your basic instrument. Measuring current consumption when all your ammeter reads is zero will force you to start thinking carefully about your tools.
Blue? You haven't gone to UV yet? I hear it's worth at least an extra 0.4FPS.
Huh? What bunker and satellite? Look at these pictures: we're talking about medium-sized RC planes, except with a camera on board and computers and such. Surely you don't think maintenence and such are free on a large plane?
They're cheaper to operate, they can stay aloft longer, and they pollute less. What's not to like?
Alternately, I could work on the code I know, you can work on the code you know, and the OpenOffice developers can work on the code they know. We all pay attention to user requests, and then we don't have to all go learn a new codebase every time we find a program that's missing a feature. Much more efficient that way, don't you think?
If you're idly curious, I'd recommend the corresponding Wikipedia articles. They're pretty good. If, on the other hand, you're the poster or someone else interested in really learning the stuff, I recommend you turn to the textbook that forms another major piece of your educational material (the final major piece being actual hardware that you experiment with and occasionally let the magic smoke out of). The circuit without the theoretical basis won't teach you much, but it is a tremendously important addition. Seeing how current mirrors are put to use in a practical circuit will give you a far better understanding of them than just seeing them in the textbook.
The 741 (even though it's a somewhat odd design by modern standards) makes an interesting first non-trivial transistor circuit -- but you should certainly have a basic understanding of how a 1-transistor amplifier works before moving on to any nontrivial transistor circuit. I suppose I wasn't entirely clear, but my post was intended as a second step in the learning process; other people have offered excellent suggestions for introductory texts.
If you want to learn the stuff, but not really "study" it, I suggested some parts to get. Buy them, along with a solderless breadboard, some 22-gauge hookup wire, a voltmeter, some jumper clips, and go build yourself a circuit or three. Looking for something simple? Start by building a multivibrator that blinks a pair of LEDs. Or try a single-transistor microphone amplifier (a cheap microphone intended for eg walkie-talkie use will cost perhaps $1-2 at radio shack; less at Digikey. The 2N3904 NPN becomes your gain stage, with some coupling capacitors and biasing and gain-setting resistors, and it should easily drive headphones or a small speaker (though be careful with headphones: letting the magic smoke out with them connected would probably be loud). Heck, you could build both of those circuits combined for $10 in parts at radio shack plus a breadboard (<$10 for a small one) and a multimeter ($20 or less for a cheap but versatile digital model), and power them off a 9V battery. (Note that at least the radio shack near me stocks all the part numbers I listed, but you'll find it cheaper to go to Digikey if you're planning to buy more than a couple random transistors.)
See, that's the thing. If they're being sincere, they're not really trolling. Of course, that just makes it all the scarier.