i2p

i2p thread! Requirements >firefox >foxyproxy >latest version of i2p http://www.i2p2.de/en/ i2p websites to start off: collection of websites http://i2pwiki.i2p/index.php?title=Eepsite/Services simple help forum http://zzz.i2p/ booru http://hiddenbooru.i2p/ If any of you know of more websites, please let me know! I am specifically interested in finding forums/imageboards.

I tried I2P in 2015 and could not get it to work. I reinstalled it last week, it took up 98% of my CPU usage, and it still did not work. Wtf is this soykaf

NNTPchan is pretty cool, you should check it out if you can.

>>15585 Well, I don't really know what it is that you could've done wrong. I remember I tried using it with Opera the first time (because Opera is my main browser) and it didn't work. The second time I tried it, I installed firefox, since that's the browser everyone seemed to use, and it worked. Either way, here's some video tutorials to hopefully help you! https://www.youtube.com/user/i2ptutorials >>15586 I couldn't find it... I tried typing nntpchan.i2p but it couldn't recognize it. Jump service didn't work either. Could you post the link?

>>15586 woah is nntpchan still a thing? if so that's dope. i really liked the concept of it i2p is good but i just wish i had more reasons to use it

>>15585 Try i2pd, it is a C++ implementation that uses less resources and should perform better than the original java one.

>>15592 I don't why people seem so fond of a c++ version? It'll have lots more bugs than the original even faster.

Okay sounds

>>15595 It means we'll be able to bring I2P nodes into the world with less powerful hardware needs. Like small single board computers, or old computers. I got it running on a iBook G4 with Debian 8.

>>15595 Because Java is a thousand times worse. i2p people really should drop it and focus their efforts on a sane implementation people can actually use.

What about the fact the huge number of cves are only possible in c/c++ domain? can't find the numbers right away, but believe it was something like 1/4th of all vulnerabilities - that to me sounds like a pretty solid reason to chose another language. Especially for a security related piece of software.

>>15656 How many higher level languages are written in C C++ ?

>>15657 At some point almost all of them are. Does it make program in those higher level languages inherit all the properties of the lower level language? Some pieces (bootstrapping devices, runtimes, vms, etc) effectively are c/c++ programms and do get to keep all the props, but the program in the higher lang do not have to. That's how manual memory handling errors aren't a thing say in java. You still have to tend jvm, but the java apps would get a much better set of guarantees. I don't want to turn this into the general language flame war debate. I have no foot in java, just find the idea of rewriting stuff in c/c++ for perf reasons alone questionable. My point was that other languages can make for a safer programs, which might or might not be important to you.

>>15656 just because they're possible doesn't mean they'll actually happen. Programmer incompetence is the direct cause of 95% of bugs.

>>15656 Good reason to rewrite it in a language that's neither those nor fuarrrking Java.

Just starting to use I2p here, and it's awfully slow for me. Sites might open about half the time after a minute or so, and then when I click a link I might make it there or it might tell me it couldn't find the page. Is it always this slow? What can I look into to speed it up?

>>15905 Change your bandwidth share settings to a lower number and be a hog about things. You can also reduce the number of hops your stuff does from the control panel, but I wouldn't change these settings unless you're running a seedbox from somewhere that doesn't give a soykaf so you can upload/download faster. The outproxy though, is slow as soykaf forever, and always will be, use tor if you want out of i2p or foxy proxy into tor from i2p and be super anonymous, then go into a vpn for extra latency and to be allowed access to wherever you want to go.

>>15649 If I remember, they basically got stuck with Java. It has to do with the protocol, which they said at the time was too complicated to re-implement in another language. That said, I didn't know there was a C++ version available, so clearly there's been some work done.

>>16223 I've tried the c++ implementation and its crap. The one that receives updates and attention is the java one, and unless you want to patch up the c++ one I wouldn't do it. The java one is much simpler to use and its the only one I can recommend using until there is development on the other implementation or they move it to another language.

>i2p dead networks, doesn't provide security, resource hog literally why use this

>>16259 Tracking down i2p websites is even harder than onion websites. That's the only reason I could think of. Basically, its even deeper deep web.

>>16260 the marginal anonymity improvement is small and the network is microscopic in comparison. I think tor could use some improvements to be a little more i2p-like(garlic routing is a really good idea) but i2p itself, is just not that great.

>>16236 >I've tried the c++ implementation and its crap Absolutely, stay away from the C++ version. I agree with >>15675 just because C++ *can* be dangerous doesn't mean you shouldn't use it. Unfortunately the guys writing the C++ version of i2p simply haven't got a clue what they're doing. Just look at this soykaf, straight up buffer overflows all over the place https://github.com/PurpleI2P/i2pd/pull/1016/commits/e1b58039026a661ff6c39eeca4f2a04c9a2b3bbf https://github.com/PurpleI2P/i2pd/commit/cb91891f2209aa66afa72e055b2efd5aef496eb4 https://github.com/PurpleI2P/i2pd/commit/0aa3aa1b8d27d40b6203efb5b4c47269835b21cd The irony is that retard-tier languages like Java were invented for exactly the kinds of people trying to rewrite i2p in C++. >>16259 >literally why use this It's an decentralized anonymity network so use it for the things decentralized anonymity networks are used for. You realize your mommy is not forcing you to pick one or the other, you are allowed to use both. The main thing i2p has over Tor is that it's decentralized. Tor security and operation is entirely dependent on a handful of servers called directory authorities. Without them it's over. Most people don't realize that the only reason Tor continues to exist is because a small number of governments are *letting* them. i2p is peer to peer so can continue to operate in a world where Tor can't. Hopefully it's a world we'll never see but it's important to have something standing by. The other thing is that Tor is tightly coupled to TCP as a transport. i2p is more flexible, and can be used with protocols and applications that don't really work over Tor like bittorrent or meshnets.

>>16272 to be fair: i2p also has "bootstrap nodes" which bootstrap the DHT in a very similar way, it's not all that different.

>>16273 Stop spilling soykaf, please. It's like telling Internet and meshnets are the same - they are both made of computers and links. Directory authority nodes are essential to Tor, while bootstrap nodes are essential to bootstrapping network connectivity in I2P. Ones are used to run the thing, the others - to initialize another thing.

>>16271 >>16259 Which hidden networks do you use? What advantages do they have over i2p?

>>16275 the actual risks involved are the same though. When you bootstrap, you're trusting that the bootstrap nodes you use are giving you correct information, just as you are trusting that the directory authorities are giving you correct information. The exact same attacks would work: compromise the bootstrap nodes or directory authorities, compromise most of the network.

>>16283 >When you bootstrap, you're trusting that the bootstrap nodes you use are giving you correct information, just as you are trusting that the directory authorities are giving you correct information. Ok, but the point of >>16272 is that knocking down like 4 or 5 direct authority servers (either legally or illegally) would disable the whole Tor network for everyone until someone changes the Tor code and distributes it somehow. Taking down the hardcoded i2p bootstrap nodes would make it harder for new people to join the network but it would still continue to exist.

What are some things you can do to make an eepsite more secure? I know they have a note about Tahoe-Lafs for deepsites but I haven't seen much written about it.

I really hope that Kovri takes off, it seems to have decent potential but apparently the lead dev decided that he was going to work on some other project... can't remember the name of it off hand.

What are main imageboards in i2p?

There's a chinese one, and some russian ones. They're all in the main addressbook anyway.