/tech/ - Technology★

/ipns/QmUkcA1EfgP6pWMDF4nDNNqXFKfaGgCfRM4vMLgsn3ZKVq

bunch of anime and good uptime (I don't switch off that computer so there is always at least one seeder)

Can you change the contents of a directory and keep the same link? I honestly think this would be one of the best features if possible.

>>916334

If you use IPNS like >>915976, it's possible to make it point to a different IPFS object. Old versions of the page are still available, but the IPNS link will point to the newest one.

>>916405

▓▓▓▓▓▓▓█▓▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒▓▓▓▓▓▓▓▓██░▓▓▓▓▓▓▒█▓▓▓▓▒░░░░░█░░░░░███▓▒██▓▓▓▓▓██▓▓▒░░░░░░░░░░░░░░░░

▓▓▓▓▓▓██▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒▓▓▓▓▓▓▓▓█▓░▓▓▓▓▓▓▓█▓▓▓▒▒░░░░░▓▒░░░░░██▓▓▒░█▓▓▓▓▓▓█▓▓▓▒░░░░░░░░░░░░░░░

▓▓▓█▓██▓▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒▓▓▓▓▓▓▓▓██░▓▓▓▓▓█░█▓▓▓▒▒▒░░░░░░█░░░░░░▓█▓▓░░▓█▓▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░░

▓▓▓▓█▓█▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒█▓▓▓▓▓▓██▒▒▓▓▓▓▓░█▓▓▓▒▒░░░░░░░░█░░░░░░░▒█▓▓░░░█▓▓▓▓▓▓█▓▓▓▒░░░░░░░░░░░░░░

▓▓█▓▓█▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒▓█▓▓▓▓▓██░░▓▓▓▓▒░▓▓▓▓▒▒░░░░░░░▒▓█▓▓▓▓▓▓▓▓▓██▒░░░░█▓▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░

▓▓▓▓▓█▒▒▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓███▒█▒█▓▓▓▓█░░░▓▓▓▓░░█▓▓▒▒░░░░░░░░░░█░░░░░░░░░░▒█▒░▒▒░░▒▓▓▓▓▓▓█▓▓▓▒░░░░░░░░░░░░░

▓█▓▓██░▒▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▓▓▓██▓█▒▒▓▓▓▓░░░░▒▓▓░░█▓▒▒░░░░░░░░░░░▒▓░░░░░░░░░░░░█▓░░░░░░░█▓▓▓▓▓▓▓▓▓▒▓░░░░░░░░░░░░

▓▓▓▓█▒░▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓▓▓▓▓██▓▓▓█▓▓██▓░░▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓▒░░░░░░░░▓▓▓█▓▓▓▓▓▓▓▒▓░░░░░░░░░░░

███▓█▒░▓▓▓▓▓▓▓▓█▓▓▓█▓▓▓▓▓▓▓▓██▓░▓▓█▓█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▓▓██▓▓▓▓▓▓▒░▓░░░░░░░░░░

▓██▓█▒▒▓▓▓▓▓▓▓██▓▓▓█▓▓▓▓▓▓▒▓█▓░░▓█▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓▓▓█▓▓▓▓▓▓▒░░▓░░░░░░░░░

▓██▓█▒▒▓▓▓▓▓▓▓██▓▓▓█▓▓▓▓▓▓░▓▓░░░▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓▓▒██▓▓▓▓▓▒░░░▓░░░░░░░░

▓███▓▒▒▓▓▓▓▓▓▓███▓▓█▓▓▓▒▓▓░▓░░░░░░░░░░░░░░░░░░▒▒▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▓░▒█▓▓▓▓▓▓▒▒░░░▓░░░░░░░

████▓▒▒▓▓▓▓▓▓█▓██▓▓█▒▓▓░▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓░░▓█▓▓▓▓▓▓▓▒░░░░▒░░░░░░

████▓▒▒▓▓▓▓▓▓█▓▓██▓█▒▓▓░░░░░░░▓██████▓▓▓▓▓████▓▓▓▒▒░░░░░░░░░░░░░░░░░░░░░░░░▒▒▓▓▓███▓▓▓▓███▓░░░░▓░░▓██▓▓▓▓▓▓▒░░░░▒░░░░░░

████▓▒▒▓▓▓▓▓▓▓█▓▓█▓█▒▓█░░░░█████████████████▓▓██▓▓░▒░░░░░░░░░░░░░░░░░░░░░░▒░▓▓████████████████░░░░███▓▓▓▓▓▓▒░░░░░▓░░░░░

████▓▒▒▓▓▓▓▓▓▓▓▓▓███▒░▓░▒█████▒░░░██████████████░▒▓░░░░░░░░░░░░░░░░░░░░░░░▒▓█████████▓▓▒░░▒███▓░░░██▓▓▓▓▓▓▓▒░░░░░░▓░░░░

████▓▒▒▓▓▓▓▓▓▓▓▓▓▓██▓░░░░▓██░░░░░░██▓███▓▓███▓██░░░░░░░░░░░░░░░░░░░░░░░░░░░░██▓██▓▓██▓██░░░░██▓░░░██▓▓▓▓▓▓▓▒▒░░░░░▒░░░░

████▓▒▒▓▓▓▓▓▓▓▓▓▓▓▓██░░░░░▓█░░░░░░█▓▓▓█▓▓▓▓█▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░█▓▓▓█▓▓▓█▓▓▓░░░░█▓░░░▒██▓▓▓▓▓▓▓▒▒░░░░░░▓░░░

████▒▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓██░░░░░▓▒░░░░░▓▓▓▒▒▒▒▒▒▒▓▓█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▓▓▒▒▒▓▓▒▓▓░░░░░▓░░░░▓█▓▓█▓▓▓▓▓▒▒░░░░░░░▓░░

███░▒▒▒▓▓▓▓▓▓██▓▓▓▓▓▓▓█░░░░░░▒░░░░░▒░░▒▒▒▒▒▒▒▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▒▒▒▒▒▓░░░░░▒░░░░░██▓▓▓▓▓▓▓▓▒▒▓░░░░░░▒░░

██▓▒▒▒▒▓▓▓▓▓▓▓▓█▓▓▓▓▓▓▓▓▓░░░░░░░░▒░░░░░▒▒▒▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▒▒▒░░▒▒░▒░░░░░▒██▓▓▓▓█▓▓▓▒▒▓░░░░░░░▓░

██▒▒▒▒▒▓▓▓▓▓▓▓▓░▓██▓▓▓▓▓▓█▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░███▓▓▓█▓▓▓▓▒▒█▓░░░░░░▒░

██▓░▒▒▒▓▓▓▓▓▓▓▓░░░░▒██▓▓▓▓▓█▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██▓██▓▓▓▓▓▓▒▒█▓░░░░░░░▓

███░▒▒▒▓▓▓▓▓▓▓▓▒░░░░░░░▒▓██▓▓▓█▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓█████▓▓▓▓▓▓▒▒▓█▓░░░░░░░

████▒▒▒▓▓▓▓▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████▓▓▓▓▓▓▒▒▓██▓░░░░░░

████▓▒▒▓▓▓▓▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓██████▓▓▓▓▓▓▓▒▒██▓▓░░░░░

████▓▒▒▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░████▓██▓▓▓▓▓▓▓▒▒███▓▒░░░░

████▓▒▒▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█████▓███▓▓▓▓▓▓▒▒███▓▓░░░░

████▓▒▓▓▓▓▓▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████▓███▓▓▓▓▓▓▒▒▓███▓▓░░░

████▓▒▓▓▓▓▓▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓██████▓███▓▓▓▓▓▓▒▒▒███▓▓▓░░

████▓▒▓▓▓▓▓▓▓▓▓▓▓▓▓█▓▓▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▒▒▒▒▒░░░░░▒▒▒▒░░░░░░░░░░░░░░░░░░░▓███████▓███▓▓▓▓▓▓▒▒▒▒███▒▓▓░

████▓▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█████████████▓▓▓▓▓▓▒▒▒▒▓██▓░▓▓

████▓▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████████▓▓▓▓▓▓▒▒█▒▒███▓░▓

█████▒▒▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒▓███████████████▓▓▓▓▓▓▒▒█▒▒▒██▓░░

>t. literal retard

If you were actually secure, people knowing your IP wouldn't cause any problem. Please tell me you're trolling. Please.

>>916406

>If you were actually secure, people knowing your IP wouldn't cause any problem.

Yes anon I just love the idea of every random fucko being able to read everything I have downloaded on IPFS. I love that they can do an IP lookup and see what city i'm in. I just love that I cant setup private files on IPFS where only authorized users can download them. This is a great system real good job. Tor and IPFS support still in development years later. Total bullshit.

>>916406

IPFS developers are wasting all their time on crypto coin scam projects so they can make some shekels. This is why IPFS will never be done. Muh filecoin.

>>916407

>>916409

>the state of IPFShills rn

Based anons exposing this shit for what it is. Everyone who isn't fucking retarded abandoned this shit months ago. It's laughingstock nigger-tier, OPis a fag.

>>916407

>read everything I have downloaded on IPFS

Eggsblein bleez

>only authorized users can download them

IPFS is for file sharing. If you want to have only (((authorized users))) be able to download, I recommend you use SFTP with an appropriate password.

I do however agree with you that IPFS development is slow. The software is bloated and written in a shitty meme language (Go). However, I can't make anything better myself, so I won't complain too much.

>>916407

Well, considering the popularity of public BitTorrent trackers, I have this eerie suspicion that most people are okay with this.

>>915966 (OP)

>Is it a meme?

>You be the judge.

>It has implementations in Go (meant for desktop integration) and Javascript (meant for browser/server integration) in active development that are functional right now

I have judged.

>>916435

You are right they clearly don't know what they are doing. They should have just written it ONLY in javascript. Then they could just use NodeJS on the backend having halving their work.

>>916411

>look fellow anons, I'm trolling using cutting edge shitposting lingo

>rolf I btfo'd those guys so you should totally listen to me

You're the same fag that derailed the last thread. You obviously have an ulterior motive since nobody can naturally be this retarded. You're just slapping together a bunch of shitposting vocabulary without even knowing how to apply it correctly in context.

Who hired you to try to shut IPFS threads down?

>>916413

IPFS can have private swarms.

>>916453

>IPFS can have private swarms.

Bullshit no it cant. Their solution to "private swarms" is configure your node bootstrap list to only have nodes you like and then hope no one else connects. There is no authentication or particular support for it or anything. And if you do do that you cant also use IPFS for public files.

>>916453

>You obviously have an ulterior motive

No way anyone could actually be against a shitty meme software written in javascript and go (((Google))) that has the development pace of a snail still missing basic features because the devs are money grubbing on other projects.

>>916473

I agree. Point me to a better solution which contains the same functionality, then. Until that happens I will continue to use IPFS.

>>916476

It is harmful software that solves nothing. It has not a single use case. You don't use it. You don't store files on it. No one does.

>>916536

>Point out the specific deficiencies of IPFS.

Did that already.

>Point out a better alternative as well.

Torrents

░░░░░░▒▒▓▒▒▓▒▓▒▒▒█████████████████████████████████████████████▓▓▓▓ >>916472

░░░░░░▒▓▒▓█▓▒▒▓▒▓██████████████████████████████████████████████▓▓▓ >muh private file sharing

░░░░░▒▓▒██▓▒▒██▓▓██████████████▓▓▓▓▓▒▒▒▒▒▒▒▒▓▓▓▓████████████████▓▓ IPFS is for content distribution. If you don't want it

░░░░▒▒▓███▒▓█████████████████████████████████▓▓▓▒▒▒▒▒▒▒▓▓▓██████▓▓ to be public, don't bother using IPFS. Use SFTP or

░░░▒▓██▓█▓▓█████████████▒██████████████████████████████▓▓▒▒▒▒▒▒▓▓▓ something like that.

░░▒███▒████████████████▒▒▓████████████████████████████████████▓▓▓▓ >>916473

░▒▒█▓▒███████████████▓▒░░▒████████████▓▓████████████████████████▓▓ >shitty meme software written in javascript and go (((Google)))

▒▒██▒▒███████████▒██▒░░░░▓████████████▒▓█████▒██████████████████▓▓ IPFS is a protocol. As long as the protocol is implemented

▒▒█▒▒▓██████████░██░░░░░░░░██████████▒▒░████░▒▓██████████████████▓ correctly, I don't give a shit about what language it is

▒█▓▒▒██████████░▒▒░░░░░░░░░▓████████▒▒░░▓██▒░▒▓██████████████████▓ written in unless I have to write code which interfaces with

▒█▒▒▒█████████░░░░████████░░███████▒▒░░░░██░░░░▓█████████████████▓ it. Google has nothing to do with IPFS.

▒█▒▒▓████████▓░░░░██████░░░░▒█████▓░░▒▒███████▓▒█████████████████▓ >>916539

▒█▒▒█████████▒░░░░█████▓░░░░░▓████░░░░░██████░░░▓███████████████▓▓ Torrents don't have content-addressability. Webpages can't be

▒▓▒▒█████████░░░░░█████▒░░░░░░███░░░░░▒██████░░░▓███████████████▓▓ conveniently hosted on a torrent. Torrents reveal your IP

▓▓▒▒▓████████░░░░░▓██▓█░░░░░░░░▓▒░░░░░▒█████▓░░░████████████████▓▓ to everyone who is uploading or downloading from you.

▒▒▒▒▒████████░░░░░░░░░░░░░░░░░░░░░░░░░▒█████░░░░████████▒░██████▓▓

▒▓▒▒▒▓███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████▓█▒░██████▓▓

▒▓▒▒▒▒▓██████▒░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▒██████▒█▓▒██████▓▓

▒▓▒▒▒▒▒▓█▓████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓██████▒░████████▓▓

▒▒▒▒▒▒▒▒█▒█████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓█████▒▒█████████▓▓

▒▒▒▒▒▒▒▒▒▓▒██████▓░░░░░░░░░░░░░░░░░░░░░░░░░░░░░▓████████████████▓▓

▒▒▒▒▒▒▒▒▒▒▓███████▓░░░░░░░░░░░▒▒▒▒░░░░░░░░░░░░░█████████████████▓▓

▒▒▒▒▒▒▒▒▒▒▓██████████▓░░░░░░░░░░░░░░░░░░░░░░▓███████████████████▓▓

▒▒▒▒▒▒▒▒▒▒▓████████████████▓▒░░░░░░░░░▒▓████████████████████████▓▓

▒▒▒▒▒▒▒▒▒▒▓██████████████████▓▒▒▒▒▒▒▒███████████████████████████▓▓

▒▒▒▒▒▒▒▒▒▒▓█████████████████▒▒▓▓▒▓▒▒▒▒▓████████████████████████▓▓▓

▒▒▒▒▒▒▒▒▒▒▓██████████████▓▓█▓▓███▓▓▒▒▓▓▓▓██████████████████████▓▓▓

▒▒▒▒▒▒▒▒▒▒▒████████▓▒▓▓▓▓▓▓█████████████▓▓▓▓███████████████████▓▓▓

▒▒▒▒▒▒▒▒▒▒▒█████▒▒▒▒▒▓▓▓▓▓▓▓▓▓████████▓█▓█▒▒▒▒█████████████████▓▓▓

>>916550

>Webpages

<What is webtorrent

>>916563

>Magnet links contain the hash of the torrent file itself, not the actual contents of the torrent.

Why the fuck do you think that is a response. You said:

>>> Torrents don't have content-addressability.

That is exactly what a magnet is.

>Actually you can use IPFS to share private information. Just be sure to encrypt it first.

Or they could implement basic access controls. Having to encrypt everything would be a massive PITA making the whole easy filesystem part worthless.

>You can use IPFS through tor, I believe.

Its been "in development" for years.

>Use the Go implementation then.

>GoogleLang

>But webtorrent is javascript and according to you javascript is bad.

Yep, thats why I don't use it or any of the web shit.

>>916566

>exactly what a magnet is

The magnet allows you to address the .torrent file. It does not allow you to address the content (i.e. what you want to download), whereas IPFS does.

>implement basic access controls

And how are they going to do that? How are they going to identify who has access to the file besides by using a password? If they did that you might as well use encryption from outside IPFS, which is simpler and more secure.

>"in development" for years

You can use IPFS gateway through a hidden service. However, yes it is true that the IPFS daemon itself can't be run through Tor without deanonymization.

This is still better than torrents, which can't be run through Tor at all.

>GoogleLang

The language being invented by Google isn't a reason not to use all software programmed in it. It's a compiled language so it's all the same in the end.

>thats why I don't use it

You can't use webtorrent, but you can use IPFS (through the gateway) without using any javascript.

>>916571

>It does not allow you to address the content

Again this is just wrong.

>How are they going to identify who has access to the file besides by using a password?

Via the node key they already generate for authentication purposes. Or a password.

>You can use IPFS gateway through a hidden service

<You can run an http server that you proxy all your requests through on tor

durr

>It's a compiled language so it's all the same in the end.

<All compiled languages are the same

<Trust google

>through the gateway

Anon torrents have "gateways" too.

>>916575

You've fucking run out of arguments now.

>via the node key

so they would need to send me a long ass public key, instead of telling them a simple password? nice "solution" you got there.

>blah blah cant use through tor

You can download stuff from IPFS through tor. This is a fact and you can't ignore that. That is already better than torrents.

>all compiled languages are the same

>trust google

Now you've run out of things to say. The Go compiler is free software. Many people from non-google backgrounds work on it. The fact that the best implementation of IPFS is written in Go doesn't affect you in any way whatsoever, unless you are working on the code, because it's a compiled language which generates static executables.

>>916580

>You've fucking run out of arguments now.

Nice try anime poster. He cries out as he strikes you.

>instead of telling them a simple password?

Nice way to cut off that quote

<Via the node key they already generate for authentication purposes. Or a password.

<Or a password.

LOL

>You can download stuff from IPFS through tor.

Yes you can go to a website and download a file through tor. Great feature! Come back when I can run IPFS on tor.

>The Go compiler is free software

Yes goy if its free software it must be good.

> unless you are working on the code

Ah yes what things are written in has no effect on me what so ever.

>>916586

>GPG symmetric encryption is very easy to use and if you can't figure it out you should read the fucking man pages.

The FS part of IPFS would be useless then. I'm supposed to just be able to DL a file and have it work. Not to have to decrypt and copy it somewhere else. Total waste.

>Point out a better alternative.

Torrents.

>I will trust it until someone has evidence to the contrary.

Well thats retarded.

>so it doesn't matter to you.

I suppose if I buy a car and its made of inferior steel that does not matter to me either. I don't build cars after all.

>>916589

>Are you too lazy to run gpg --decrypt <file> and type the password?

With IPFS i'm supposed to open a file on my file system and its supposed to fetch from the network and open automatically. So yes.

>whereas you can't do that with torrents.

No anon I can connect to a server through tor and download my torrents. Same thing as IPFS. Also torrents work through I2P a network superior to tor.

>are all engaged in secret activities

Anon its a bad language. Its not backdoor-ed (I think).

>you get a headache when trying to read and write the code.

Yes of course nothing could possibly happen when someone fucks up a program. It could not delete all my files, allow remote code execution, leak my key, etc.

>caused explicitly by its choice of programming language.

In literally any language retards are always like "Muh bad programmer muh error muh blah blah". C fags do this all the time.

>>916591

>With IPFS i'm supposed to open a file on my file system and its supposed to fetch from the network and open automatically. So yes.

You still have to enter a password with your shitty solution of including encryption/decryption code into the IPFS daemon.

>connect to a server through tor and download my torrents

I don't believe this is the case, unless you operate some server somewhere which you torrent from and then transfer the files to your torified computer. Actually torrenting through the tor network, while possible, deanonymizes you in the same way as using IPFS over tor does.

>security bugs

>muh bad programmer

Find a security bug. Go ahead, find a security bug which was caused by it being written in Go. I believe Go is more "memory safe" than C, and you're using (either directly or indirectly) many C programs right now (and trusting them with private data) without screeching autistically about them.

>>916597

>You still have to enter a password

Yes I agree passwords are a bad idea. I don't know why you thought that was good. Better to just use the public

> unless you operate some server somewhere

Yes thats what IPFS does.

>Find a security bug

Anon i'm not going to go audit a project for you.

> I believe Go is more "memory safe" than C

There are few things worse for memory safety than C.

>>916599

>public

*Public key system

>>916597

>You still have to enter a password

I can enter that in the daemon in one command and then get access to the whole repository that is being hosted. Much different than a manual PGP layer.

>>916608

>If you're talking about the autogenerated IPFS key then it would be necessary to acquire the public keys of everyone who you want to share files with.

Yes anon this is a basic property of access control.

>This is more inconvenient than typing in a password, ONCE, when you download the file, and then storing it normally on your own hard disk.

It totally breaks the convenience of the whole system.

>Nice argument

Its not an argument. I am not your personal auditor.

>Not if you program properly

You can say that about literally anything. This logic has done untold amount of damage.

>as I said, you aren't complaining about the multitude of C programs you

Not in this thread.

>No you can't you imbecile

Anon typing a password into a CLI is fine.

>And if you don't there is no difference to using gpg.

With GPG I have to go manually decrypt everything and then copy it to a different directory. Compared to IPFS now where I just open it and it downloads and thats that.

>>916608

I hope the (((IPFS))) devs, are paying you a bunch of their scam coin. The money you make from this shilling may counterbalance the utter lack of features their system has.

>>916633

>n if you aren't going to be contributing to the code

Which is obviously wrong for the reasons pointed out.

>No. Encryption is built on top of a general web protocol

Anon encryption happens BELOW HTTP FTP and all that other shit. It should happen below the IPFS layer all teh same.

>>916641

Anon its easy. Its called THE EXISTING KEYS NODES HAVE, and passwords. Done.

>>916641

>but the (((Certificate Authorities)))

Woah anon i'm not adding a human readable name system here. Totally different problem. I don't need to certify a name I just go get they password from them.

>>916646

Yes, private trackers. You can have an authentication server (that works ever tor / I2P) and then use the torrent part normally.

>>916690

I agree torrents are annoying for this. SyncThing and BitTorrentSync (proprietary dont use) use the torrent protocol and do this very nicely though.

>>916735

Keep shilling for your meme software that literally nobody on this board but you thinks is anything worth shits.

>>916735

>>916741

anon thats not me

>>916741

The people with actual brains actually had arguments against your shitty meme protocol, unlike your "NOOOOOOOOOOOO STOP HATING ON IT NOOOOOOOOOOO WAAAAAAAAAA :'("

>>916743

What kind of retarded /g/ happening brought you faggots here?

>>916784

0.000000000002 Filecoin has been placed into your account

>>916967

>I wonder why they're derailing the threads now of all times

Yeah, I wonder why people are JUST NOW catching on to the (((IPFS))) scam. Controlled opposition at its finest.

>>916982

We all know you're a FUD troll but why put IPFS in echos? What's the scam? If you're salty about the $252 million they raised for Filecoin why bitch about it now? Why didn't you whine about their ICO when it happened last year? Only (((accredited investors))) could buy in but that has no impact on the open source protocol or implementations they're coding. They outright say everything is still in heavy development and is not production ready. They're in the process of having the code audited.

>>917048

Nigger if you're going to be a shill at least tell me what FUD you're slinging. You keep rambling about bitcoin so link me to something to prove your point.

So I recently claimed the /ipfs/ board. There probably isn't enough of a community or interest on here at the moment to warrant a board, but hopefully it can be used as a place for anons to post ipfs related content or share whatever they're working on.

>>>/ipfs/

Ultimately, I think an ipfschan would be pretty cool, so I'm gonna do some research on that and post progress on the board. If anyone has any ideas or suggestions, I'm all ears.

>>917073

>ipfschan

Personally I don't think IPFS is good as a framework for an imageboard, but besides that there is already an IPFS-based imageboard called smugchan.

>>917073

Don't worry, anon. Once normies ipfs is a streamlined version of BitTorrent, people will come in flocks to share their files. I'll post on your board.

>>917073

Please actively moderate the board so it doesn't turn into /tech/. Years ago we had great discussions here but now that /g/'s here to stay I'm running out of retard-free places to go.

>>917084

There are retard-free places to go. Of course I won't mention them on this cancerous board, but they do exist if you look for them.

>>917080

>there is already an IPFS-based imageboard called smugchan.

Huh, I'll have to check it out then.

>>917084

>>917085

>Please actively moderate the board so it doesn't turn into /tech/.

Easier said than done, but I'll do my best. Personally I think boards can just about self-moderate themselves if the communities occupied and there are things to do that aren't shitposting and template threads. The Libbie threads were a good example for that type of thing, /tech/ was really great those few weeks because people could put their energy into something.

>>917087

True. I also believe that people shouldn't depend on moderation but rather filter anyone who's an obvious troll, instead of replying and derailing the thread.

>>917091

>What is a sybil attack

LOL

>>917243

That functionality already exists in 8chan. It's called "hide post", or "filter by ID".

>>917249

No anon. Thats not how it works.

>I doubt they're in financial need with the support they have.

Yes anon thats whats so bad about it. They have all they need and they are money grubbing for more.

>>917253

That doesn't make any sense. You act as if something that isn't even released yet, is inherently valuable. Knowing how cheap Amazon instances are, and the purpose of Filecoin, I can only imagine the monetary value of these things will be low. I can't see how this would be a detriment to anyone other than big hosters. People like me who leave their ocmputer on anyway and pay for fast service anyway at least stand to make a few bucks while sleeping.

I'm real curious as to how you see this as a negative thing or how the team can even exploit it for profit. Even beyond that, if they do profit, I'm still curious as to why this is bad too. I would rather open source projects get funding than Amazon, Google, et al.

Maybe it's there but I can't see any negative to this for individuals, only for corporations.

>>917258

The negative is that they are wasting all their time on it instead of implementing basic IPFS features.

> is somehow vulnerable to a sybil attack

They are.

>>917256

>Filter by ID

>ID is free

LOL

>>917266

>ID is free

What the fuck are you talking about you utter moron. Filter by ID is to hide all posts made by the same user, if he's being insufferable and hasn't bothered to switch IP.

>>917265

I think that's a bit unreasonable. The progress of go-ipfs alone has been great, for an alpha release they're doing all the experiments they can and improving things all the time. The js-ipfs people have done what's necessary to get browser vendors to adopt at least preliminary support, which is necessary to be a viable alternative to HTTP at all. The Filecoin project handles the incentive for hosting and distributing content, much like CDNs today, a necessary component for people who would otherwise not be able to afford to host something all over the world, and 24/7 with their own hardware. For people who are comfortable with that, go-ipfs exists already. I see nothing wrong with creating options for people, especially individuals.

Until electricity is free and networks are reliable, an incentive to host content feels like a necessity. For those with disposable resources (a good amount of people), we're already ahead of the curve.

To put it another way, in the same way HTTP is worthless if the server is down, IPFS is worthless if the content doesn't resolve.

With HTTP your only options are CDNs, with IPFS you have a CDN-like option or a self distributed option, over whatever networks you want to boot.

>>917314

>I-IT'S MAKING PROGRESS I SWEAR

Kys

>>917269

>filter by ID is to hide all posts made by the same user,

<ID is still free

LOL

>>917314

> an incentive to host content feels like a necessit

I agree, now let them make a platform thats not shit to build it on before they try to get a bunch of shekels out of it.

>>917568

They fix bugs, conduct experiments, add features and reach out to vendors. I'm not sure what you're expecting. What area are they neglecting for a distributed filesystem? From my perspective this is the most active project for distributed filesystems. All I ever hear about ZeroNet are the limitations and security concerns, bittorrent doesn't have the flexibility to compete here, Upspin may as well only exist on Google's campus, and anything else is just a whitepaper with no working implementation.

IPFS is usable right now, in multiple languages, and has preliminary browser support. When is the last time you remember browsers adding protocol support like this? That sounds like vetted progress. The libp2p project (what IPFS relies on) were posting about the possibility of writing a kernel module for Linux, that would mean inserting their networking at the OS level.

You're going to have to point out the slack if you want to say they're not making progress. As someone who really wants something like this and has investigated a lot of the alternatives that existed before IPFS, this one is the only one I've seen that has reasonable progress, in a field FULL of hypothetical competition (the majority of projects like this never make it past concept let alone YEARS of regular commits).

>>917622

>now let them make a platform thats not shit to build it on before they try to get a bunch of shekels out of it.

If you're saying that Filecoin should be stable before they release it, that's what they appear to be doing. I don't know what you think is happening.

Maybe you're conflating Filecoin and IPFS? They're separate projects with separate teams. The only thing in common is some of the developers overlap and the parent company is Protocol Labs for both.

These things should be done in parallel, otherwise you end up in situations where IPFS is released and people don't adopt it because there's no CDN-like equivalent. For a p2p network, adoption/popularity is obviously important. They're most likely trying to finish everything at the same time.

IPFS on the desktop, server, browser in combination with Filecoin for non-tech businesses who want to host data worldwide.

For reference, Filecoin is a blockchain, the only thing it likely has in common with IPFS is libp2p, having 2 applications of libp2p during development is valuable in itself since it will likely have different needs and encounter different problems to fix before release.

IPFS is just a decentralized datastore with content addressing. You could implement a blockchain on top of it but I'm willing to bet Filecoin is Ethereum based or very similar.

>>917758

>If you're saying that Filecoin should be stable before they release it

No. IPFS should have more basic shit working before they build filecoin.

>>917758

> The only thing in common is some of the developers overlap and the parent company is Protocol Labs for both.

Totally wrong

> Filecoin works as an incentive

layer on top of IPFS

https://filecoin.io/filecoin.pdf

>>917767

fucking newline

>>917758

>write a tl;dr denialfagging post

>everyone ITT roasts the fuck out of you

oh no no no no

>>917783

>two anons give brief responses

>this is somehow roasting the fuck out of him

>actually using oh no no no no

>>>/cuckchan/

>>917765

>No. IPFS should have more basic shit working before they build filecoin.

Why? I can see the benefit of developing them in parallel, I can't see a reason not to do it.

What I'm asking is, how is Filecoins existence detrimental to IPFS, especially when we both agree it's a necessary component to have early on for adoption.

Implementing something on top of IPFS is a perfect way to actually test the stability and limitations of the base. How else would you do it?

>>917767

>The Filecoin protocol is a Decentralized Storage Network construction built on a blockchain and with a native token. Clients spend tokens for storing and retrieving data and miners earn tokens by storing and serving data.

Being built on top of IPFS does not mean it is similar to IPFS, it means it is an extension of it. They're implementing things on top of IPFS that IPFS doesn't have itself. IPFS obviously isn't a blockchain and doesn't have a token of its own, but you could easily implement them with merkledags and libp2p. That's what they're doing.

>>917783

I think we're just having a discussion, Anon. I'm not interpreting the replies as "roasts", they seem legitimate. As someone who is interested in seeing decentralized networks take off, I'm very interested to hear the negatives of IPFS and what alternatives we have for it, but from everything I've seen and heard I'm banking on IPFS for now.

People are saying something is wrong but not explaining it well, I'm trying to discuss it with them to figure things out for me and anyone reading. From my current perspective a lot of the flak seems unfounded but I'm giving people the benefit of the doubt.

>>917837

>Implementing something on top of IPFS is a perfect way to actually test the stability and limitations of the base. How else would you do it?

More importantly, this should be done before they freeze all the APIs and decide on defaults. Now is the only time to do that. Otherwise you have an implementation meant to build on top of without that having actually been tested first. I don't just mean technical testing either but conceptually too.

>why can't it do x

>why does y happen when I try to implement z

etc.

If they dogfood their own product now, it should benefit everyone currently and later.

>>917832

>assuming the gender of /tech/ is male

lel

>>917837

>Why? I can see the benefit of developing them in parallel, I can't see a reason not to do it.

Because they are working on too much shit and IPFS needs more work.

>What I'm asking is, how is Filecoins existence detrimental to IPFS,

Because its the same developers and company.

>Being built on top of IPFS does not mean it is similar to IPFS

Exactly. Which is why they need IPFS to be good before they build more shit on top of it.

Ok, I like ipfs, I like where it is going, it might be shit currently but luckily that's only because of its slow development, it's not bad by design, however, I REALLY do think that Tor support is a priority, or any sort of anonymity mechanism, the point of IPFS is to have a permanent uncensorable network, but it being uncensorable is pointless when you can't share files anonymously, in other words, what's the point of it being uncensorable if you can't share files that someone would want to censor without getting in trouble.

Ok, someone blogposts about something and regrets it 10 minutes later, 9:30 minutes too late, everyone is already hosting the file on IPFS, "haha, lel, sorry friend but you can't delete things on the internet :)", the blogposter says "ok" and sues everyone doing it.

I know you can't just sue an IP (well it depends on your ISP/Country), but you better bet this will be a thing if IPFS becomes mainstrean.

>>917845

Oh, and additionally, if they do implement Tor support, there should be a way prevent folks from seeding "risky files" (stuff that might get you in trouble) on accident, say, you share a risky file on the network and mark it as a risky file, now everyone who tries to get that file from you gets a warning saying that they should use Tor to get the file or just get it with Tor automatically, once they get the file and save it to the to the datastore it's automatically marked a risky file, of course there will always be people who are gonna shit it up so you could just make the warning show what percentage of the swarm marked it as a risky file, so if for example 30% of the swarm marked it as a risky file you get the warning.

>>917841

>Because they are working on too much shit

I think we just disagree on this, it's hundreds of people spread across like 4 or 5 projects with what I would call steady progress. They've only gotten faster with more people.

>IPFS needs more work

This is what I'm curious about, can you be more specific? What in particular are you missing right now.

>Because its the same developers and company.

I don't see how that's a problem, or even how it's anything other than standard. Even in my personal life I work on multiple projects. If you don't do this, you go insane or burn out. My current employer is the same way, we have overlap between teams but it's not 1:1.

>Which is why they need IPFS to be good before they build more shit on top of it.

Strongly disagree, I think building on top of the foundation and testing it is what strengthens said foundation.

I guess we have to agree to disagree. Even if I did agree though, I don't know how you solve the problem. Like you can't just throw more devs at it and hope for the best. Other people with experience are probably already being paid to work on simillar proprietary tech at other companies. I don't even have to mention why you don't want inexperienced devs working on it.

It's a tough problem for any project but I always preffer a slower and steady approach to a fast and loose. Being wild about it isn't going to get you into any web standards meetings (maybe Mozillas because they just don't care anymore).

It seems like the major complaint is just that people wish it was already done, which sounds better to me than other types of complaints.

>>917845

They've started treating libp2p like its own project recently. Bolstering efforts there may contribute to interoperability with things like TOR, I2P, etc

The only thing missing from the tor transport is an audit, but it's functional otherwise and has existed since 2016 without anyone making complaints.

Obviously a security audit isn't going to happen until right before release.

>>917860

The OpenBazzar people have been using the IPFS over tor for a while now. And they openly sell drugs and shit on their platform.

https://github.com/OpenBazaar/go-onion-transport

https://github.com/ipfs/notes/issues/37#issuecomment-247717781

That doesn't mean it's secure but also nobody has been busted yet.

>>917091

As you can see, self moderation only impacts those who were never going to reply in the first place. That makes it essentially useless in preventing replies to trolls. There needs to be an active force against serious shitposters or it'll keep happening.

>>917845

You beat me to the punch. The tor transport isn't officially recognized because it hasn't been through an audit yet.

>>917870

>There needs to be an active force against serious shitposters or it'll keep happening.

We should go back to old etiquette and tell those replying to trolls to lurk more. Trolls were never a problem, replying to them is. Humiliate people who reply to bad posts, not the people who make bad posts to begin with.

It will be worse short term, better long term.

Better moderation in the first place would help but in multiple decades it has been proven to be something we can't rely on, on any site, in any bbs-like service. The only thing you can influence is the community, not the moderators.

>>917866

>What in particular are you missing right now.

Tor, I2P, access controls. Just to name a few.

Keeping this thread alive

>>917874

>We should go back to old etiquette and tell those replying to trolls to lurk more

>The only thing you can influence is the community, not the moderators

I agree that shaping the community is the best line of defense against degrading post quality. That said, there's a tipping point where outsider's low quality content becomes normalized. On large imageboards we're only delaying the inevitable. It's beginning to happen here.

I'm just saying we need all the extra help from mods we can get (but from the looks of it they don't care which is a real shame. If only they enforced the questions/support and consumer advice stickies).

>>918690

Don't bump the thread until it's on page 14 or 15. Otherwise you're just killing it sooner.

>>918739

bumping to kill thread

WE NEED MASS ADOPTION

>>919324

Then buy a drive and start hosting a shit ton of stuff on it. It would only take a few people doing this to make it practical.

>>919384

>have an extra drive laying around and ipfs running on a raspberry pi

>drive takes 900mA, while the usb ports can only support 200 tops

Guess I need to get a powered hub or some shit. Or maybe I can just use an old laptop that's lying around and skip the pi altogether. That or I need to bite the bullet and purchase an actual home server one of these days.

>>919490

Buy an old thinkpad with an i5 for $100 on ebay and leave it plugged in somewhere. Even has a built in battery backup.

Now I remember why I stopped browsing this fucking board months ago. The IPFS thread used to be a nice little oasis from /g/ but I guess no longer. Why is the OP numbered like some faggy general? There's been far more than 2 IPFS threads.

>>919490

If you had purchased a decent SBC instead, you wouldn't have this problem. I used to run IPFS on a Banana Pi Pro and external hard drive (with an AC adapter). It even has at least enough power to run a 2.5" SSD.

Threadly reminder that >>>/ipfs/ is a board. If you're working on an ipfs-related project or just want to post some cool hashes where people can find them, there's no better place.

>>919578

>It even has at least enough power to run a 2.5" SSD.

Nice, I might pick one of those up then. I like SBC's, but I guess rpi's turned out to be the shittiest of the bunch.

So why doesn't everyone use Zeronet?

>>919689

Fag.js, that's why

>>919384

>>919490

>>919578

Not just that, we need "trackers", "search engine" and "clubs"

like https://pantsu.cat/ or https://nyaa.si/ or https://www.tokyotosho.info/ or https://www.shanaproject.com/ or https://www.shanaproject.com/ or https://www.anirena.com/

but with IPFS and not Goddamn BitTorrent

>>919629

>>919784

Made a thread on >>>/ipfs/9 about IPFS + Bittorrent indexes. I think I've come up with a pretty good decentralized solution.

>>917621

>what is a dynamic IP

A lot of people can change their IPs and clear their cookies in less than a minute, which changes their IDs and bypasses any kind of block. Mobile carriers will give you a new IP every time you reconnect to the network (you can force this by turning airplane mode on and off), IDs aren't as good as you think.

>>917073

How would you use IPFS with a dynamic website like an imageboard? IPFS websites don't even have search engines afaik.

>>920869

>and clear their cookies in less than a minute,

Yes anon that was my exact point

>>920869

>How would you use IPFS with a dynamic website

Very easily, you have a database consisting of JSON files. I explain it in my post on >>>/ipfs/9 for use as a real-time decentralized database for a bittorrent/IPFS sharing site. That said, it might not be worth the trouble for using it in an IPFS imageboard since not much is lost if threads get nuked because we (hopefully) archive the important ones.

Decentralizing data and changing the authority model are two completely different things if that's what your aim of using IPFS is for.

>>920577

I could see it happening to bitchute as well

>>919324

We'll see it when it's actually out of alpha. Until then this is a gentoomen exclusive.

>>919689

Zeronet is just bittorrent, it's not as exciting because it doesn't really change or add anything, it's not nearly as versatile as IPFS. Not to mention it feels like an experimental project, not a serious attempt at replacing HTTP. I'm not saying it's bad, just that it doesn't seem like anything to be excited about, we've had bittorrent in the browser for a long time, even before zeronet.

https://en.wikipedia.org/wiki/BitTorrent_(software)#BitTorrent_DNA

>>920869

You should look at the smugchan source code if you want to see an example of that exact thing.

>IPFS doesn't even have search

There's a few search projects, some crawl sites and others scrape the dht to build an index.

http://ipfs-search.com/

>>920071

Cool stuff. I'm more interested in all the shady projects that use IPFS more than any of the legitimate ones. I'm waiting for someone to host some video game romsets that people can just mount.

Like this but for everything:

/ipfs/QmacAqRVhJX9eS7YJX1vY3ifFKF9CduDqPEgaCUSa4x5xb/

>>920577

JUST

>>921413

> I'm more interested in all the shady projects that use IPFS more than any of the legitimate ones.

id also like more sites like this, but i am afraid of creating one because it could damage IPFS' reputation

>>921107

>we all remember when SMTP got sued over spam

Except the public mail routers did get sued and then all shut down and now we have non public alternatives.

Juse use OneDrive, theres no point in this shits.

>>915966 (OP)

>>916608

>it would be necessary to acquire the public keys of everyone who you want to share files with.

Then use plain AES encrpytion with a password goy

>>916641

A couple of solutions for decentralized resolution already exist, mostly work by ownership of a private key. But i hear there is already work on something called IPNS (Inter Planetary Name Service).

Regarding the shills it is going to happen. Remember all the fakenews when Let's Encrypt was invented? oh-no mah scammer website can gets SSL. Education and facts are to be used against shills and fake news. In the case of Let's Encrypt it was LE doesn't authenticate the motives of websites, just the website itself. Laymen: Encryption does not mean people will not scam you, it just means nobody else can read the interaction between you and the scammer, not a third party. (unless you give up that information)

>>921914

More clearly, It doesn't mean your not going to get scammed, It means your communication channel with the scammer is secure.

We need Windows installer for IPFS+I2P for mass adoption

>>921904

>Then use plain AES encrpytion with a password goy

Then you have to share the password for every file with every person you want to share with, and then they have to manually do all the decryption fucking up the caching. OR we could go with public keys that already exist and have IPFS seamlessly decrypt.

>>921948

It won't.

>Whole of *booru can be downloaded with ease

>Every video and ebook can be mirrored

>The wet dream of >>>/ipfs/ when combined with I2P

https://github.com/richardschneider/net-ipfs-mount

https://github.com/marcin212/ipfs-gui

https://github.com/bogdoslav/ipfs-windows-installer

>>921953

I meant >>>/pdfs/

>>915966 (OP)

>http://glop.me/

It's ded.

>>922001

The silver lining is here

>>>/ipfs/9

/ipfs/QmY3eTAm48UDEdu8vPVhvvPZu9ra5SBKH4ZcLcYqaytRnN

Most recent /hgg2d/ translated games repo.

I wish they had a TODO page somewhere, I can't get excited for something when I don't even know what they are working towards.

>>922064

The teams have public live-streams where they say what they're working on, show demos of their projects, and answer questions

https://github.com/ipfs/pm/blob/master/README.md

Notes are taken during the session and published when they happen

https://github.com/ipfs/pm/blob/3a4bdabc7fb4423565e38e8f4868858f13877589/meeting-notes/2018/2018-05-07--all-hands.md

there's also this

https://waffle.io/ipfs/go-ipfs

https://waffle.io/ipfs/js-ipfs

etc.

>>921948

Don't friendly up good software or hardware too much. This is what happened to the Internet, now look at it. slopping along the floor half dead with FaceBook aids ,goobermint regulation, and spying. Or mobile computing. You want an emote bar right and no IO right?

Their Rabin Chunking is just bad, can't be used to save space if they want it to.

>>921957

THIS but more refined

Any ideas on these?

https://zeronet.io/ and https://blockstack.org/

Requesting tutorials for IPFS+I2P setup on Linux.

Do it for Windows as well if you want mass adoption.

>>923070

Zeronet is more for dynamic content. It also requires js, which is problematic for some people. Finally, it's supposed to be decentralized, but ZeroID ends up being necessary for a lot of websites, which is centralized. You also end up being pseudoanonymous on most sites because of it, but I think the identity only applies to that specific site. There's a screencap of a post claiming that you can track people across different zeronet websites, but I don't think that's true anymore.

You can route it through tor, but not i2p yet.

I've never heard of blockstack

https://github.com/ipfs/archives/issues/137

>>923456

See >>923584

>>923584

The problem I'm having, is figuring out where the problem is. That is to say, I can't tell if they have bad results due to a faulty implementation or if the method is inherently worse than fixed sized. Another variable is obviously the dataset.

Ive seen people talk about rabin fingerprinting outside of IPFS, as something that greatly improves efficiency in one way or another. I think mostly for getting deltas between binaries. That seems like it would fit well here and I'm curious as to why it doesn't.

I wonder what de-dupe methods other people are using, like ZFS, BTRFS, etc.

I know people seem to complain about memory usage for ZFS when de-dupe is enabled.

If you watch the Apple WWDC, then youll realize that this will never catch on, because Apple is ending the personal computer forever. /tech/ is starting to realise this and give up on fighting against the cloud. All will be centralized on the cloud. Conform or be vaporized.

>>925012

Lmao this

>INVALID

>PENILE

>FAGGOTRY

>SHED

>>925012

Fucking this.

People in this thread >>924688 are now talking about it. It's fucked forever.

>>925022

Watch some cuckfag call us CIA niggers just because we actually see shit realistically instead of living in the giggly happy fun time ideal future rose tinted soy-o-vision world that decentralcucks see.

dead thread

GEE I WONDER WHY

Shit thread, dying project. NEXT

Hopefully the WWDC shit recently shut this delusional thread up.

>>925088

didn't take long >>925031

>>925095

Read this >>925019 and this >>924970 conversation. No shilling here, just straight truth from people with a realistic view of the world.

>>925012

> what is porting software to another platform

i hope you know that js-ipfs is being developed, and that's one thing Apple and Microsoft cannot kill, the browser

Can someone please write up an instruction book on how to use IPFS with I2P?

Test input (Some latex 2hu doujin) (pin if you want)

/ipfs/Qmc9ritHTbf4nzNo4kE1ezuaY8fTR22q4G2hPLJB5Si3Q7

>>925227

Go ahead and test it yourself. Stop begging for content and start making some.

JS-IPFS v0.29 is released

https://github.com/ipfs/js-ipfs/issues/1320

>Bitswap is 30% faster

>js-ipfs now works with uglify-es just fine

>Circuit Relay was merged in

https://github.com/ipfs/js-ipfs/tree/master/examples/circuit-relaying

>S3 backend for IPFS

https://github.com/ipfs/js-ipfs/tree/master/examples/custom-ipfs-repo

>interface-ipfs-core keeps getting more API calls documented and tested

>Bandwidth stats now available

>Fully Async PubSub API

>Ping API implemented

>Electron Main Process support with Electron 2.0

Circuit relay was one of the two features that js-ipfs was missing to get full connectivity. Alternate backend support is something I didn't even know existed. It's a pretty awesome feature if you combine it with Minio (self hosted S3) or Ceph (massively clustered S3). I wonder if it's possible to switch your go-IPFS backend to something similar.

>>925530

Yes. We are getting closer of having in-browser IPFS similar to Torrent Tornado. I like this.

>>925530

JS is getting good?

https://www.sitepoint.com/10-languages-compile-javascript/

https://codeburst.io/8-javascript-alternatives-for-web-developers-to-consider-22f8d38bdfa9

bamp

We. Need. More.

>>927157

>>926348

Stop. You're only spamming the thread.

There's no need to bump unless it's on the last page. Everyone already knows the thread exists.

Site: https://orion.siderus.io

Repo: https://github.com/Siderus/Orion

>>927558

thought this was that peersub orion chat project, really confused me for a while

>>927558

>using electron for a simple program

Do people forget webview exists? There's no reason he should be using electron.

>>927616

I think a lot of people use it just because it's the flavor of the month. The main issue with the application right now is the fact that (at least on Windows) it gives you no option to choose a custom directory during installation if you use the .exe installer. Also, it doesn't seem to be respecting where I told IPFS to put the datastore. Thus, my datastore is now on my SDD instead of the large HDD I pointed it to. I had this issue last time I used IPFS on Windows but managed to fix it but this client doesn't seem to respect the environment variable I've set.

>>927613

Well we should contact >>>/pdfs/ about this, such that more people use it.

>>927616

See >>927622 about "flavor of the month". I think it is more of a cross-platform thing and not a JS thing.

>>927622

Tell the dev and make an issue.

>>927631

I'll share it on /pdf/ sometime after I take care of this machine. Currently backing up everything on it to finally get rid of the Windows dual boot hell I've been suffering from. Had to wait until I upgraded the CPU for hardware passthru before I could get rid of Win7.

As for the datastore issue I think it was simply misconfigured. I'm going to do further debugging before I send off bug reports. I also think it may be the ipfs port for Windows and not this GUI. I'm going to see is a symlink solves the issue because ipfs doesn't seem to respect the environment variable for the datastore.

>I think it is more of a cross-platform thing and not a JS thing.

Exactly, I just think qt would be better than electron but this stuff is for normalfags so I'm sure they wouldn't care either way. Just prefer qt in these cases as it's more mature and better tested. This client is a good start though and I'd love to see it eventually implement search features. In time it could be turned into a full blown Napster/Kazaa/traditional P2P application clone with chat and text forums. It'd be nice to have something like that for normalfags so we'd have more people connected to the network and seeding files.

>>927634

I said >>>/pdfs/ not /pdf/... also http://inclibuql666c5c4.onion/ and https:// inclib.i2p/ could be a good candidate.

>>927635

* https://inclib.i2p/

about IPFS over Tor, hasn't OpenBazaar forked IPFS and made it work over Tor? https://github.com/OpenBazaar/go-ipfs

>>927650

Good, but no I2P.

>>927655

>This GUI for ipfs seems stable enough to shill to normalfags

Exactly. That is the goal.

Also do me a favor, if you like smut, >>>/rule34/38083

>>927659

> tfw finally have a place to dump my porn

>>927666

satanic porn trips, pump it to >>>/d/ and >>>/hentai/ as well.

>>927667

Fight at all fronts, >>>/pol/, >>>/pdfs/ and >>>/redstick/

>>927671

>>927678

The holy grail is here. I am glad that it is here. >>>/hentai/ is temporary, >>>/pdfs/ are forever.

>>927671

For IPFS itself you can set the environment variable IPFS_PATH

SET IPFS_PATH=X:\wherever

Maybe the software would respect it too.

>>927678

>Normalfags aren't going to put up with the frustration of setting all that up

Why not make a batch script?

>>927758

>nor does it support hosting files without duplicating them

That's still and experimental feature you have to manually enable. The gui should give users advanced options to tweak settings like that.

>>927922

>Why not make a batch script?

Batch script? Whats that anon? Is it a type of play?

>>927937

Are you trying to be funny?

>>927667

Why CC 2015 and not 2017/2018? Or are they the same thing?

>>927958

do you mean a batch file, .bat?

>>927758

Also for the monthly book pack, /ipfs/QmVs4izcbRuG9VSYuVFrRSjZeCk79W6VJVxRXbu3uM2p5D

>>927922

Well...

https://github.com/richardschneider/net-ipfs-mount

https://github.com/marcin212/ipfs-gui

https://github.com/bogdoslav/ipfs-windows-installer

>>928031

I2P has a non java implementation that works (I use it) just fine.

>>928039

Exactly, and we need it for a more VPN-like experience for IPFS (to handle some ISPs)

>>928136

>VPN-like experience

What's that?

>>928141

Some people wants their IP to be safe (OpSec), and some people can't really torrent through their ISPs so they need work-arounds.

Test Input (Some Railgun doujin) pin if you want

/ipfs/QmbCW35reZKktWUrg63hEALZC9WubysDyQZAVpnpwmmH92

Test Data Supreme (Evangelion Doujin) pin if you want

/ipfs/QmYofmy26TZs4j6Mso5wAzGLLe9i5f9qmMB2eFPrXspbRm

Is there something like a mirror of The Odin Project and the websites it links to?

>>928884

Be the man you wish to become

Can someone please give an update on DistributedMemetics?

>>916407

Well IPFS through TOR/I2P, damn it

>>933061

IPFS does not work over tor.

>>933136

But someone did it with I2P and did not post a guide.

Beatoraja, a rhythm game player, has an option to download/distribute songs using IPFS.

The game looks like this: https://youtu.be/63PM00d8Xm0?t=39m58s The song files store the key charts, all key sounds (each key can make a different sound) and the background video. Trying to find working download links for the thousands of songs out there is a pain in the ass, so IPFS is really helpful.

>>933158

evidence or gtfo

>>933459

>>933158

I highly doubt that would ever work considering the almost complete absence of bridges from I2P to normalweb.

It's simple, IPFS is Redditware. Designed with total disregard for privacy, security, anonymity.

But hey, they have a shithub, a cool name, website and they gave some talks, so it must be good right?

>>933598

>It's simple, IPFS is Redditware. Designed with total disregard for privacy, security, anonymity.

this

>>933136

>IPFS does not work over tor.

Why do people not read the thread? /tech/ is a slow board, you have days-weeks to read it.

bottom part of >>917866

>>933436

That's cool, I wonder if they use libraries from beets to do the syncing.

https://github.com/beetbox/beets

>>933598

>complete absence of bridges from I2P to normalweb

There's a gorillion tor and i2p tunnels, if someone has done it, it's obviously possible.

https://www.hiddenservice.net/

It's not a complicated process of just translating the requests or implementing some other hybrid method.

>Designed with total disregard for privacy, security, anonymity.

The baseline is the baseline, the high end is the high end, the benefit of IPFS is the ability to utilize the existing high ends.

See >>917245

>Something people bring up a lot is the fact IPFS isn't as secure as TOR or I2P which doesn't make sense either since that's not the goal, the goal is interop with them, if you like I2P's routing over IPFS's, you'll be able to just swap them out. It's not trying to compete, it's trying to be the gluecode between many existing technologies that are ALREADY proven and liked.

There's no sense in reinventing encryption or anonymous routing when solutions already exist. It's better to just use IPFS (as an interface) on top of it.

>But hey, they have a shithub, a cool name, website and they gave some talks, so it must be good right?

The best thing they have over everyone else is a working implementation that's practically useful. I'm sure this is what is securing them adoption. Nobody cares about promises, if you're older than 10 then you'll know that people present p2p networks all the time with either no implementation, or terrible ones. It's all whitepapers, concept documents, etc.

You can use IPFS right now, they have a clear roadmap, active developers, and publicly known funds. This isn't just going to disappear like everything else.

Any doubt I had in the project was pushed out when Mozilla added support for the ipfs:// and ipns:// protocol. How often does that happen?

>>933626

>Why do people not read the thread? /tech/ is a slow board, you have days-weeks to read it.

We have addressed multiple times at this point why your claim of tor support is bullshit

>>933638

Answer: That is not official, it is a fork that can raise a few eyebrows

>>933638

>If IPFS security and anonimity is supposedly bullshit, and the tor support is also supposedly bullshit, then how are these people getting away with it?

You live in a fantasy land where you think everyone has to use l33t haxor technology or the NSA / FBI van you at 3AM. In reality facebook groups and clearnet sites are what almost all (uncaught but nearly in the open) crime happens on.

>>933660

> there's no way something even more secure puts them at any risk either.

<You should not have trivial security features because people sell drugs on facebook

IPFS fags are all absolute retards jfc

>>933661

The issue with what you just said is that they don't have trivial security features, they have the standard with the option for you to use what you want. If you can just use what you trust, what's the problem?

What security feature is missing? Even by default you can have an F2F network with relay nodes. How's that different from the rest? And again, if you don't like it, you can just use tor underneath. I fail to see the gap that you do. I would understand lack of trust for new security, but you can use either a new or vetted system.

>>933662

>What security feature is missing?

Tor support, IPFS support, MAC, etc.

>>933662

> if you don't like it, you can just use tor underneath

Except you cant because it does not have tor support

>>933666

*I2P

>>933667

>>933666

Do you specifically mean by default? Because when you say it like that, you seem to imply generally, which is not the case.

Experimental forks exist now, and first party support is planned, so says the developers in the github threads.

When IPFS is released, if they have first party support for it. Would you be okay with it then?

The thing I'm trying to figure out, and I'm sure most people are. Is if this is technically better, so that we can prepare for it early.

I've seen all kinds of methods to solve the simple problem of "get data X". As a poorfag, I've been pirating content forever which means using the hotest meme clients and networks, while dodging DMCA without paying for a VPN.

Eventually you see something pop up that just incorporates all the benefits of A and B, or wraps them both in some way. Like IRC+DC clients, or eDonkey(MFTP)+Kademlia(DHT).

Given all the public knowledge, the statements of intent, the track record of fulfilling them, the support from other developers. This is looking like the next thing and if it is, I want to be familiar with it sooner rather than later, because I'm getting really sick of everything else being shutdown and censored.

If the Openbazzar can function, I think it's at least good enough for my needs. But it still looks like it will be secure enough for people who need it to, in the long term.

I'm not trying to bury it before it's released since this one shows actual potential to see adoption, which in turn would benefit the networks it plans to support, such as tor, i2p, etc. I think it could be a benefit for everyone.

More people using secure networks, more content on that network, no compromise in usability or migration costs.

So I'm coming from a frustrated perspective where I feel like this solves our censorship issue, which relies on security/anonymity. But the latter half is still in development.

I think it will resolve in time. But I'm confused as to whether or not you guys are implying the path they're taking isn't sufficient, or if you're just saying you wish it was in the default right now instead of later.

I don't want to recommend it if it's doomed, but they seem like they're on the right track. This tor experiment seems to be going well? I don't know what else can be done other than audit it yourself right now, but it does exist, and it does seem to work. If it's provably secure, I think that puts a lid on it and shows that it's a more convenient method of sharing data, anonymously.

ipfs daemon --init-profile=tor
ipfs get /the/data.xz

is convenient compared to say, setting up a proxy per application, per network, etc.

Look at Bittorrent clients like BiglyBT, they have a plugin that runs its own I2P node to fetch bittorrent content over it and you can select which network(s) you use for each transaction.

The foundation for that is in IPFS but it's not limited to just switching networks. When it comes to reach/availability, you'd be able to restrict it as narrow or wide as you want. If you only trust i2p you could just pull from it, if you're not concerned, you can pull from everyone on every network, but the paranoid nodes make no compromise since to them, since the traffic conforms to the standard protocol that they already understand and trust.

Openbazzars implementation should be as secure over the wire as the other implementations since they both have to conform to the same protocol/standard. Assuming they didn't stumble into a new flaw in the protocol themselves, which tor and i2p are relatively vetted.

>>933674

>Do you specifically mean by default

It does not have to have tor turned on by default no

>Experimental forks exist now

Right so its not a feature

>if they have first party support for it. Would you be okay with it then?

Assuming it supports the full feature set. That would mean being able to host files and not having go through a website.

>>933675

>Assuming it supports the full feature set. That would mean being able to host files and not having go through a website.

If I understand everything correctly, this seems to be the goal of the project. To standardize the interface across networks, without special cases.

If you want to use ipfs to host or retrieve data over tor only, to specific peers only, you should be able to do that and the tools will honor your restrictions without complex setup. But also, it can multiplex them if you don't care, meaning that as long as you can reach the data, you will reach it, also without complex setup.

`add` add's. `get` get's, etc.

The security and networking stuff is handled by the configuration of the daemon/libp2p. So even if the reference implementation doesn't handle it (unlikely since it's the reference), the capability is there.

To tie back into the previous example, it's like how BitTorrent(the client) doesn't do i2p native like biglybt does, but they're both using BitTorrent(the protocol) and a bridge interface between BT and i2p.

It's like 2 layers of APIs, the simple one for the interface, and the core one for configuring your node, which secures/restricts the api on top of it. It removes the chance for you to leak data easily for not knowing if something is going to respect your config or if you need to provide it with a special environment (socks proxy and shit). If you can trust the node, you can trust the tools using the node to not leak. *In theory

It seems like a solid concept and people are implementing it without obvious issues.

Obviously on anonymous imageboards, people are trying to figure out how to make anonymous services, and some people are making them with success using what's available now. I'm hopeful, but yes, it's not complete yet.

>>933660

>Those images

I've seen those images before, but riddle me this. If that is what they were going for, then how come the integration with tor/I2P is so troublesome?

>>933753

>based anon brings this up

>entire thread goes silent

>Redditware (((IPFS))) fucking exposed and BTFOed yet again

Fucking victory.

>>933674

>>933675

So the forks actually gets things done instead of the main project... also how do you IPFS+TOR/I2P in detail? Could you add that to Orion GUI client?

>>933924

>So the forks actually gets things done instead of the main project

This is typical of large open source projects. If the main project doesn't have the manpower to do what you want right now, nothing stops you from doing it yourself and trying to get it adopted into the main repo.

If they could just do everything themselves right now, there'd be no alpha phase. You can't just click heels and pray that it writes itself. At least with open source projects, anyone can contribute to the work as long as there are enough people to do patch reviews.

I doubt tor and i2p are a short term focus for them, they don't stand to gain anything but negative publicity from pirates, pedophiles, and perpetrators. Why choose that when you could implement literally any other feature? They've been improving the browser side of things recently which seems like a more obvious choice.

>IPFS works better in major browsers

vs

>IPFS now supports popular darknets officially

Tor, Freenet, etc are good projects who's adoption was stifled by this exact thing. Nobody wants to associate with it.

As long as you can do it now, and they plan to officially adopt it later, I don't see a problem for anyone.

>how do you IPFS+TOR/I2P in detail

There's nothing complicated about it, if the daemon supports it, you just listen on that transport. So if you want to listen on the onion network, in your config, just add the address you wish to listen on as a multiaddr and it will use it.

/ipv4/127.0.0.1/.../onion/...

use whatever addr, port, etc. you want.

https://github.com/ipfs/go-ipfs/blob/master/docs/config.md#addresses

if you want to listen on that exclusively, make sure to remove other listeners.

If you want to setup a F2F network you just do this

https://github.com/ipfs/go-ipfs/blob/master/docs/experimental-features.md#private-networks

Then you could have your own private network comprised of onion nodes, or any weird combination you want.

Combine it with

https://github.com/ipfs/go-ipfs/blob/master/docs/experimental-features.md#circuit-relay

and you can make exit nodes so non-onion nodes can still get content via relay if you want that, obviously if you don't, just don't set them up.

>Could you add that to Orion GUI client?

I don't even know which Orion you're referring to, but I'm not going to add it for you anyway. Ask the maintainer or look into it on your own. Again that's the point of distributed open development.

No offense, but I'm a little surprised to see that ignorance on /tech/ of all places. Are you not familiar with how Linux is developed? How most open projects are made? Even our own imageboard software was pretty collaborative at a time.

>>934052

>You can't just click heels and pray that it writes itself

No the reason its taking so long as the faggots are money grubbing with their scam coin instead of working on IPFS.

>>934063

That doesn't make any sense. They had an ICO for Filecoin that raised actual millions and they're likely using that to fund development. This was discussed earlier in the thread, you should read it.

Filecoin only stands to weaken Amazon and friends while giving individuals a way to make free money or get free seeding credit.

Please explain how it's a scam, especially when it's not even out yet. In concept it's literally just S3 tokens at competitive prices. I don't know how this is anything but a benefit, even if you still want to use AWS yourself the prices should stay competitive and your content should be more reliably hosted.

>>934078

It's more likely that Amazon introduces a pinning service if IPFS rises in popularity. Buy hosting time in fiat or crypto.

>>934078

>d they're likely using that to fund development.

Except they are not.

>Filecoin only stands to weaken Amazon

Only in the same way microsoft weakens amazon

>especially when it's not even out yet

Oh shit no way it can be a scam if it does not even exist!

> In concept it's literally just S3 tokens at competitive prices.

Its a centralized POS with a meme token on top. This is why the SEC yelled at them.

>>934082

>Except they are not.

My fault, I didn't realized you worked in their financial department. What are they spending the money on, sports cars? How nice is your office?

The public activity on Github, from new contributors with new big projects, begs to differ. I guess these people are just making contributions daily out of the kindness of their hearts.

>Only in the same way microsoft weakens amazon

Exactly. If you're trying to implying MS's Azure didn't steal a portion of the market, I don't know what to tell you. In any case, I fail to see how more options and competition is bad for consumers.

>Oh shit no way it can be a scam if it does not even exist!

I said it's not out, as in not released, not that it doesn't exist. You've presented no evidence that supports it's a scam, most likely because you know nothing about it. Is there some inherent flaw in their spec? Is someone being cheated out of their money right now?

What makes you assume it's a scam when you can't even review it yet? There's no basis as far as I can see unless you're witholding some insider info. If it's a scam, why not expose it?

>Its a centralized POS with a meme token on top.

I don't see how that's any different than what I said. At the end of the day it's a token that you can use to barter hosting time or try to exchange for something else. I assume most of /tech/ understands the basics of this by now.

I doubt your sincerity.

>>934087

I'm not excited about yet another AWS, GCE, Azure competitor and your shilling is no different than if Google reps came in here pushing their new shitty email platform. The ONE cool thing they had was this nice distributed file system called IPFS, and its what they are letting die on the vine with years of slow development while they work on other shit.

>>934095

You're the one that brought it up, dude. Also your perspective on this is warped. I fail to see how them gaining more funds hurts IPFS, you've given no reason whatsoever.

Please explain to me how IPFS is dying.

You're clueless but you act like you know details as private as their spending habits.

Talk about shills, I wonder what company you work for that benefits from spreading FUD like this. Either that or you're some impatient Windows user from /v/.

>but I want official support RIGHT NOW

>I don't know how to edit config files

>I don't know how to build it myself

>can you implement this in a GUI for me?

most jej

>The ONE cool thing they had was this nice distributed file system called IPFS, and its what they are letting die on the vine

I can't believe you're being this dramatic over nothing. You know nothing and cast assertions all over the place, what a jackass.

>>934099

>>934095

Let me put this in a more civil way, like I did before.

If you're interested in the project, and concerned with how they're managing it. I suggest you contact the developers, instead of trying to argue with me. Even if you managed to convince me, I have no power over them. Your job would be to convince them, not me.

Maybe they'll share their scambux with you and everyone will come out with something positive.

>>934099

>>but I want official support RIGHT NOW

I actually want official support for a bunch of shit years ago at this point like it was promised. Right now would be incredibly late.

>I can't believe you're being this dramatic over nothing.

Yes anon, over nothing. Because thats what IPFS is. Years of promised features without anything to show for it.

>>934105

> I suggest you contact the developers

I'm just going to try and warn people to stay away from this scam until they get their shit together or it implodes totally.

> tfw the only way this thread gets bumped is from arguments

wew

>>934107

>I'm just going to try and warn people to stay away from this scam until they get their shit together or it implodes totally.

You realize the cyclical nature of that right? You're problem is that they don't have enough eyes focused on the issues you care about, so you're telling people not to look at it.

You're doing everything ass backwards. Saying you have a vague problem with the project, dancing around explaining it for hours, then finally admitting that your intent isn't actually anything other than to drive people away from the project. None of that benefits anyone, not even yourself if you actually want to see this come to fruition.

You disregard the growth and focus purely on what's lacking. The problem seems to be your impatience more than anything else. Just because your specific feature set isn't implimented, when you admit to not reaching out about it. What a surprise that nobody is focusing on the features people are silent about.

>late

You're being completely unreasonable in your expectations coupled with your own actions (wasting time and driving people away) and lack of action (not speaking out).

I'll take late over never myself too. I'd say you have some semblance of a point if you presented alternatives, but there aint any. Everyone else making these promises has even less support. IPFS remains the most likely contender. And while they're not kissing your ass personally, they're for sure implementing features people want, at a steady pace.

Things people used to post about like no-copy adding, faster hashing, more efficient datastores, faster dynamic data, libp2p sockets, ipfs clusters, etc. were promised and then implemented.

But since you can't trade filecoin for pedo images over tor officially yet, I guess they've made no progress.

>>934117

Isn't the whole point of this place to discuss?

What else would you post? How else would you get a better understanding of the topic?

>>934121

>I'd say you have some semblance of a point if you presented alternatives

Lets see, torrents. That was easy!

>inb4 torrents are not an identical copy of the system

>None of that benefits anyone

You know what benefits anons? Warning them about a companies scam project so that they don't invest their time into using it.

>>934142

We have before us something that could be a better platform for anonymous, censorship resistant discussion, and the place that should be most excited about this is doing nothing but saying

>what if we could just stick with our current broken platforms?

That isn't a warning at all, it's a disappointment.

People here should know the flaws of the existing systems, seek out better alternatives, and hope to improve them.

Instead what do we have, people saying it's not worth even looking at because it's not 100% complete. On /tech/, an anonymous technology focused imageboard.

>>934141

The thing to compare would be bitswap to bittorrent, but those are just transports. It's obvious that IPFS is much larger in scope.

Projects like DNA were proprietary and failed miserably, projects like Zeronet stumbled with arbitrary limitations and weird ZeronetID bullshit.

Things like Bittorrent sync are interesting, but again proprietary and not very extensible or interoperable as a result.

IPFS is at worst on par with multiple protocols, like HTTP, as well as various P2P protocols like bittorrent, DC, etc.

But manages to wrangle all the advantages together under an open standard and open implementations.

The best feature of IPFS is the flexibility, if you don't like bitswap, use bittorrent under IPFS.

If you use BTSync, what options do you have?

If you use Zeronet, you're still limited to the limitations of Bittorrent itself unless you come up with a migration plan and then migrate all the nodes.

Look at all these bt trackers hosted on HTTP that are being shut down. If you want to use trackers on i2p you have to install more software, and make them collaborate. None of that should be necessary, and we shouldn't be locked into a single transport standard either. Look at how many networks came and went before BT. BT itself isn't the greatest transport of all time.

When the time comes to migrate from it, wouldn't it be nicer if we didn't actually have to account for it? No need to rehash anything, no new client, all the old tools still work. Wouldn't that be nice?

>>934156

Actually, a more concise way to put this.

With Bittorrent, you can't even share identical blocks across different swams reliably. With IPFS you can share whatever, across whatever, even spanning different networks and relay nodes.

The evolution in my mind is

old bullshit -> torrent files -> magnets -> multihash

It's not rare to see torrents with partial seeders, where you're forced to cobble together the full file from multiple torrent swarms manually. There's no reason for that problem to exist anymore, and we're talking in the same protocol, same network, same client. Trash.

>>934160

>>934156

FFS look at this shill, writing literal blog posts in response. We cant be expected to respond to the metric shit load of bad arguments one after another. Split it up atleast.

>>934168

I would recommend torrents over i2p.

>>934173

While it's not terrible, what I'm saying is, I think there's room for improvement and it looks like people are actually working on that. From both a technical and usability standpoint.

We can have an improved protocol and user experience (for both developers and end users) without compromise.

>>934184

We have a bunch of features that exist RIGHT NOW for torrents over I2P. And we have a bunch of features that barely work, or are promised for IPFS. IPFS devs have a long history of making progress sound like its just around the corner. You should NOT trust their expectations, and instead use what exists already for things you need to rely on. If you want to contribute to experimental future shit thats great and all please do, but we have needs now.

>>934187

I'm not sure who "we" refers to.

If you're not interested in experimental projects, you might be in the wrong thread.

Everything I just mentioned, exists right now as well. You can fetch blocks from peers across multiple networks and transports, publicly and privately. The only concern people seem to have is if it's "official" or not, which makes little sense to me when you look at things like third party torrent clients implementing these non-standard feature, however they chose.

This seems like an odd thing to focus on, because you're saying you need things "right now", while they're available, "right now".

Are you concerned about the features or the authors of the features? Because on the other side of things, they seem to be doing things the same way.

Question: could someone explain the relationship between libp2p and ipfs?

Is libp2p where the anonymous networks like tor and i2p will be added, or ipfs itself.

>>934187

>IPFS devs have a long history of making progress sound like its just around the corner.

Progress towards what? If it's specific features that they promised would happen soon, like missed deadlines and all, I can understand. But if it's just you expecting a certain feature, like i2p or tor integration and/or the DHT for ipfs-js, then that's not a fault on their part. They would just not be implementing the features you want. There's also seems to be a lot of other shit they need to do based of the size of the project. Could be wrong, so an example wouldn't hurt.

I would also suggest using torrents over i2p if you want to fileshare anonymously (or retroshare maybe), but I don't see the point in shitting on ipfs if others are fine not using those features for now.

>>934196

>but I don't see the point in shitting on ipfs if others are fine not using those features for now.

There is no problem with IPFS if you use the minimal subset of things it supports and you dont have expectations of other things things in the near future.

>>934193

I don't understand how you conduct experiments without use.

While everything may not be tip-top right now, it's in a usable state, enough so that we can build on top of it and experiment with it, today. I don't think anybody is in the dark that this is a pre-release alpha project, nobody is trying to say otherwise.

I figured this was implied given that this is /tech/, a place where people conduct experiments like this as a hobby and have done so historically in these threads.

Most recently we have people talking about building hash indexes, imageboards, and weird video streaming services.

People interested in this usually like to start building now, so that when release comes, you're already finished.

I don't think it's practical to assume a hundred million dollar company, with backing from major browser vendors, and backing from big web services, is going to suddenly disappear, and all the effort wasted.

Am I crazy for believing that the people responsible for building the tools of the internet are going to suddenly allow that? On an open source project that at worst would see a fork?

Like come on man, I'm not exactly trusting but this is beyond the line. We're talking about a project that has only seen an increase in developers and patches.

Everything is public and open

https://githab.com/ipfs/go-ipfs/pulse

All the funding, the promises, their meetings, their source code.

Where's the deception? Where's the doubt? I just don't see it.

>>934199

If you have even the slightest amount of comprehension of software development, you can use experimental features too though. And the people that want those features, probably should have that competency. I think that's where the confusion is.

>>934203

The excitement comes from the fact that, in its experimental phase, it's competitive with established projects. It's not polished but it is actually viable, right now, and has only improved over time.

This is demonstrated by people like Neocities, OpenBazaar, various pinservices, etc. running IPFS in production, with real users, real money, real instances.

While the project itself is still experimental, people are using it right now on real products. And the exciting thing is that they're doing so without problems.

Me and you can exchange data over a private f2f network, using IPFS, today. It may require some setup but the promise is that it won't always be this way. You have to employ the benefit of the doubt at some point.

>>934209

>over a private f2f network

Thats just it even trivial shit like that, DOES NOT ACTUALLY WORK. There are no "private f2f networks". What you can do is bootstrap off your own node that you keep secret and then HOPE no one else ever connects to you. There is no mechanism for private networks. You just hope a normal network stays netsplit.

>>934216

And if you want to take that hack netsplit system and actually download a file on the real network? Guess what, ur fucked!"

>>934209

>It may require some setup but the promise is that it won't always be this way. You have to employ the benefit of the doubt at some point.

Or at least don't be outright dismissive and say a project is useless before it's even finished.

If an analog to torrents over i2p is what you want, you can implement it right now and it would present a better experience. The only way to get there is through experimenting with experimental features, and I think those experiments should be encouraged, not dismissed.

>yes, you can pull the patch from this fork, follow these instructions, and if it works, try to make it more streamlined or get your patches into mainline

vs

>lmao no i2p support, ded meme

>>934221

>>yes, you can pull the patch from this fork, follow these instructions, and if it works, try to make it more streamlined or get your patches into mainline

No you cant. There is no full tor or i2p support.

>>934223

Nobody is saying there is, nobody is trying to pretend that experimental support is "first party anon approved support" whatever that constitutes.

But there are people saying that it's not supported, which again is not the truth. If you want to move data with IPFS over tor and i2p, you can do that right now.

You'll have to be more specific on what you mean by "full" support. What feature is lacking, and have you told the developers about it? If so what was their response? Do you trust that response? Why or why not? Is there anything you can do to change it yourself?

When you remain vague, I have no idea what you mean, where you're coming from, or what you expect.

The argument that there's no support, is flatout wrong and has been wrong publicly since 2016.

Now that people are referencing it, the argument has changed to

>no true fork would put sugar on their patches

so then the developers say

>we'll support this eventually

and people are going

>IT'S STILL NOT GOING TO BE REAL SUPPORT

This is absurd.

>>934234

>The argument that there's no support, is flatout wrong and has been wrong publicly since 2016.

Yes you could download from the HTTP website over tor that works fine.

>>934234

You know what does work? You know what many people have looked at? You know what has worked for years??? Torrents over I2P.

>>934238

That's not a good case to make. Why did you migrate to torrents over i2p in the first place when things before it worked?

I would assume there was some advantage.

People want to migrate to IPFS for the same reason, it has advantages.

Just because it isn't complete yet doesn't mean it's not viable, a superset can be on par while still being incomplete itself, especially when considering users on a case to case basis.

For me, several advantages already exist over bittorrent for sharing data, so obviously, I'm going to use it now.

Do you seriously have trouble comprehending why someone would use something ahead of release?

You think nobody used torrents or i2p in beta?

Even forgoing use, to act like we shouldn't be discussing it is even more absurd. As if people can't talk about something until it caters to you specifically.

Maybe it's not viable for you yet, but it is viable for some and these people should be allowed to discuss it. Eventually it should be viable for most, but there's no reason to hush discussion before then.

>>934242

>Why did you migrate to torrents over i2p in the first place when things before it worked

They actually did mostly just work. By the time that shit got into the general public it just werked unlike IPFS.

>>934244

All I'm saying is, if Openbazaar's users can take the place of the silk road, openly, for almost 3 years now, I doubt you're going to run into any issues with their implementation. I think it's unfair to say it's "not supported" at all.

Can I ask what your plans really are? I won't judge, I just want to know why the concern is there. What are you afraid of getting caught with that requires this much anonymity?

>>934248

>You don't actually need secure software look at these guys selling drugs on facebook

>>934249

I'm not saying you don't need it, I'm saying they offer it and it seems to work. I'm not seeing cases where the security is faulty because nobody is presenting them. The arguments so far are

>it doesn't exist

which isn't true

>it's not official

which I don't see the relevance since you shouldn't trust the ipfs team inherently anyway, so both would require a personal audit.

and now

>it's not full support

which doesn't make sense to me since it seems to work, and it seems secure.

I'm not saying it's bulletproof, but at the same time, nobody is presenting holes in it. I would understand the concern if someone posted some news article where someone got busted, or a link to the source code where a vulnerability is.

At the moment the whole thing boils down to people invalidating somethings existence because they don't understand it.

>>934252

> I'm saying they offer it and it seems to work

Again no they don't.

>>934255

https://github.com/OpenBazaar/go-onion-transport

https://github.com/OpenBazaar/go-ipfs

https://github.com/OpenBazaar/go-libp2p-kad-dht

>>934258

>not official

disregarded

>>934258

This does not support direct tor 2 tor communication

>>934262

If 2 onion nodes can communicate without conflict, how doesn't it support that? It should be 1 implementation talking to another.

Is this being discussed somewhere?

>>934260

If you're worried about security, I hope you know none of IPFS has been officially audited yet so there's no reason why not to use it. It's all still in alpha with no security guarantees.

>>934262

>>934264

If I'm reading the code comments correctly it does support onion-to-onion address connections.

>>934286

You don't want to use the unofficial tor transport. Why is that? As I said, if your reason is concern about its security, you don't need to worry since the whole thing could be insecure right now. People are using it right now for fun and testing.

>>934287

I'm guessing some time next year the go version and its libraries will be finished audited. Hopefully at that point the reliability and efficiency problems will be fixed too.

>>934290

I found it hilarious that you said

>It's all still in alpha with no security guarantees.

...

>so there's no reason why not to use it.

>>934293

I think they're mocking my earlier posts.

>>934278

Same as some BitTorrent clients and Retroshare right?

>>934258

Until someone make it official, this is just an experiment. (No matteer how I want it to be perfect)

>>934402

A previous version of Retroshare was audited so the core should be pretty solid.

>Until someone make it official, this is just an experiment

As I've explained numerous times, they're waiting for an audit.

>We want to get all of IPFS and libp2p audited before we start encouraging people to use it with TOR. Unfortunately, this will take a while.

https://github.com/ipfs/notes/issues/37#issuecomment-353223480

That said, they should really get off their asses and start spending some of that $260 million they raised on more programmers. This is taking way longer then it should.

>>934425

>That said, they should really get off their asses and start spending some of that $260 million they raised on more programmers. This is taking way longer then it should.

The IPFS shill will defend this. They are probably giving him a fraction of it just to post all day.

>>934514

It really isn't ready for use yet in its current state. The main issue is the fact that it duplicates all the data you share on the network. I have TBs of data I'd love to put on the network but I'm not devoting double the space just to share it with the few autists using IPFS at the moment. The shills will say it's a solved issue and point to a half implemented feature on github but it doesn't work like it's supposed to. If you attempt to use IPFS in a way that isn't the default configuration that "feature" shits itself and returns unrelated error codes. It wants to place the datastore on the same HDD/SDD as the OS in the default configuration which is retarded since most of the time the OS isn't going to be on the drive with the data you want to host.

These are the same problems I've had with IPFS every time I've attempted to use it and share things with people. It has been almost four years now and they've made almost no progress on fixing these issues. It really sucks because IPFS would make for a fine P2P network but until such things are fixed no one is going to take the time to put a large amount of data on the network. Eventually IPFS will need a idiot proof GUI to see widespread adoption among normalfags which yes we do need simply because it makes the network faster/better due to having more peers. If I'm having constant problems using the CLI for it how do you think someone that doesn't live, breath, eat, and shit /tech/ is going to fare using it?

They need to get off their asses and fix these fundamental problems if they ever hope to see it go anywhere beyond a toy for a few autists. Maybe someone that actually wants to do the work will fork it and turn it into something decent.

>>934514

The fact that the shill isn't even responding anymore just gives away that he got fucking exposed lol

>>934519

Doesn't matter what they do, SV will find that they're not kosher and centralized enough and destroy it effortlessly, because they have the fucking money. Press S to spit on grave.

time to make a new thread :^)

>>934514

>>934527

I'm pretty sure the only shill here is the obvious anti-IPFS one. I just criticized the slow progress but was called a shill for defending them on different points earlier in the thread.

>>934519

>It wants to place the datastore on the same HDD/SDD as the OS in the default configuration which is retarded since most of the time the OS isn't going to be on the drive with the data you want to host

I posted this before but for JS-IPFS you're able to store your datastore in databases. I'm not sure when the same functionality will be brought to Go-IPFS. Hopefully soon you'll be able to store your files once in your filesystem and just have the IPFS metadata in a SQLite or Postgres DB.

https://github.com/ipfs/js-ipfs/tree/master/examples/custom-ipfs-repo

>Eventually IPFS will need a idiot proof GUI to see widespread adoption among normalfags which yes we do need simply because it makes the network faster/better due to having more peers

I think this will come after some of the big problems you mentioned gets fixed. If you're concerned about it right now, it isn't hard to make a simple python Qt GUI wrapper.

>>916536

Freenet

>>934674

Freenet is a slow, resource-hungry piece of shit.

>>934593

Can public users see the bumplimit for a board? What is /tech/'s?

>>934700

bump limit/auto-sage = anchor icon

I think /tech/'s is 400.

>>934704

Ahh, I though the anchor was only for explicit bump locked thread (flagged by moderators).

>>934727

You know that they will just make another thread right? This is thread #2 after all.

>>934734

If someone came on here and made sure there was always an active thread on gmail features would you think thats spam?

>>934740

>alternatives to it for security and privacy. Such as IPFS.

IPFS offers neither security, nor privacy. Gmail is actually more private because not every person in the world can read your inbox.

>>934745

>encrypt email or textfile using pgp or openssl

>upload email to IPFS

>retrieve email at your leisure by keeping hashfile

>share password IRL to send messages and upload hash to public place to recieve messages

Already more secure then gmail.

>they think as soon as a thread hits the bump limit it should be abandoned

Shit like that is why you stay at /g/. We've had IPFS threads survive weeks after the bump limit.

Also OP is a huge newfag for saying this is thread #2. We've always had active IPFS threads since it was first released in 2014. Don't bring your pointless mannerisms from cuckchan over here. For 4 years everything was fine until the this and the previous thread which got spammed with shills and newfags.