GhostKat is Immortal, Impossible to censor, and globally available!

Looking for collaborators, web developers, release groups, seeders, and to be ripped apart.
Our explanation and walk through is here
github.com/GhostKatIII/GhostKat/blob/master/Contract_ABI
Our current interface is being served here:
ghostkat.org/

Other urls found in this thread:

github.com/webtorrent/webtorrent
github.com/GhostKatIII/GhostKat
securityweekly.com/2017/06/07/exploiting-client-side-node-js-with-moses-hernandez-pauls-security-weekly-516/
butterproject.org
twitter.com/SFWRedditGifs

bump

in browser stuff is so neat, but can't they shut down the main server with the file list?

I'll take a look tomorrow, am a webdev myself.

Have a bump sounds cool but outside of my scope.

>can't be fucked to make a free SSL certificate

Come on lads.

Also bump for a good idea.

OP Here.
Our magnet links are stored on the blockchain, meaning our database is replicated across 20,000+ nodes across the world. The user queries the blockchain client side using infura.

I feel like an idiot for saying this, but I have no money in the bank. If I had $9 to buy a certificate, I would.

Bump.

Dude. Look into let's encrypt. It's free. Jfc. This project really is in need of help.

Fair enough. The main reason it is in the back of my mind is that our webserver does nothing. There is no two way communication between the user and server besides our serving of a static html page.

looks cool, have a bump

Wait, how do you even dev in Solidity if you don't know how to acquire a free SSL cert?

maybe im missing something but isnt this basically just streaming torrents on a browser without a vpn? one of the reasons people flock to streaming services that are hosted by a single entity instead of p2p is because the punishment/consequences will rarely fall on the end user. with p2p as long you are a host and your ip is visible to the swarm. i guess im struggling to see the point of this project

I'll take a look tomorrow, am a consumer thread poster myself.

I cant figure out how to get a cert or how to make this work. Otherwise I'd dump so much support into this

>tfw too brainlet to support good stuff

i was talking about the basic page mentioned here . can an end user just download a static html page and load it from disk using ctrl+o instead of visiting your server?

also i am concerned about your blockchain dev quality. could a malicious torrent get access to the end user's browser and take control? how do you add magnet links to the blockchain?

I'm interested in helping out. Where are you struggling the most?

Also, you probably want a service worker for this to function as you intend

Yes, you can just save our source html as an .html file, load it locally to connect to the swarm through your internet connection. The system is not reliant on our server at all.

To your second point.
The user can post only a string through the system. That string is handled in Javascript as a string variable being passed to webtorrent. We parse the torrent name from the magnet link, and generate text nodes we use in the html. I have attempted to code inject the site by posting strings with escapes and it doesn't escape. It just returns an error like it should. That is because we never quote the variable, we only ever let it fall within an ascii string scope. Feel free to test out the interface, and our blockchain code using our free to post function, "ridin". It allows for anyone with .30 cents usd in ether to post whatever string they want. It will likely never ever be deleted. Post whatever horrendous shit. My point is it will work, and it won't break due to anyone using it to its fullest extent.

This is the core functionality of the system.
It is very simple.
But it also works very well.

I want our interface to be way better, but we need to have people seeding content before we start expanding that.

Our github has a full walk through on how to post Mr.Bones, and the rest of our shit.

I am not really struggling except my interface crashes on mobile and non chrome browsers. I hate that shit. But it is a current limitation of the libraries I'm using. Within the next few months, and a little modifications to the library myself, I think we can overcome it.

The other thing is that I can't upload media myself, because I want the .org to be legally free from liability. I hosted it on DMCA ignored hosting so it can't be taken down, even if people post popular shit. And it won't slow down too much, because once people cache the interface, they are just talking to their peers. The more people watching ghostkat, the faster it actually gets.

Our Github readme walks through all of this.

Why does no one read the readme?

parsing multimedia from a torrent sounds like a lot of complicated code handling in a variety of formats and that usually leads to bugs. how well tested are your libraries, meaning how old are they and what's the version number? which ones do you rely on? what do you use for the multimedia inside the browser?

I don't think people understand that our webserver doesn't do anything. Yes it is the exact same thing as running a torrent client on the desktop, but the database is decentralized and impossible to take down. Getting the html client to work in a authoritarian world is the same as using a vpn now. The point is its way more user friendly, and the database is immortal.

WebTorrent actually does all of the media work. It is an in browser torrent client and can render almost any kind of media directly into the DOM.
Read their ReadMe,
github.com/webtorrent/webtorrent
Read my Readme
github.com/GhostKatIII/GhostKat

so it uses mediasource or the browser tag. no wonder javascript is such a popular target now.

securityweekly.com/2017/06/07/exploiting-client-side-node-js-with-moses-hernandez-pauls-security-weekly-516/

So has anyone posted anything yet?

doesn't look like it

might want to post something public domain as a proof of concept because right now there's nothing to even demonstrate that it's working

so this is basically butter with magnets?

wat is buttare

butterproject.org

>webtorrent
it has issues getting through NAT to establish a p2p connection, and it relies on servers to relay data (so it's not completely p2p). also, it can't connect to normal BT swarms (unless someone finishes the libtorrent integration).