You don't read code, you read comments, and they tell you how fucked up the code below is going to be.

True coders don't bother reading code anyway, because they read the first line, decide that it's total shit, and rewrite the entire program.

ok ok ok, where's the video???

You mean something like creating our own hardlines? That would be great, but feasibility would prove near science fiction. It may come to steganography over tcp/ip, disguising packets as approved data... Or in the future possibly quantum entanglement, but I'd guess quantum tech for common people is at least 20 years out.

I agree that the contributors today can already be anonymous, but I'm curious about protecting the host and the repo server. Usually today you can just google the repo location, anonymously commit or clone, and you're done; however the repo server is "known" and easily attacked/raided if big brother doesn't approve, and the server owners can be located if they're not experienced in IT ghostery.

Yeah, I would love a decentralized everything.

Yes, I get this concept, and it definitely works for smaller teams as long as the team is coordinated. But yeah, larger projects where you would like to have anonymous submissions would be challenging. You can definitely have groups of friends all exchange their repo locations, give them ssh access, etc.. But let's say you want anonymous contributions for a large project, yet prevent people from knowing where your server is located... For this case, I'm wondering how to manage that situation, and if Git is capeable of working in that environment.

My concern here is that the "blessing" of one repo may need to be changed quite frequently if the server is constantly raided or censored. Is there a simple way to transparently redirect people to the latest repo? Would the developers easily manage this without confusing contributors? This process would hopefully be simple enough as to not discourage contributors And could this repo exist centrally, or would it need to be decentralized? I'm not fearmongering, I'm legitimately curious for my own sake; I'm interested if there are simple methods to make coding with Git resilient, anonymous, and secure without imposing difficulties on contributors. Although it would be cool, I would be pissed if we had to socially code via carrier pigeon.

No, I'm speculating, but I imagine one day we may need unstoppable social coding. For example, with everyone moving to Linux recently, what if someone decided to build an open source excel clone (operationally, not code wise)..I mean a perfect clone, because I know tons of people who build their lives upon Excel and would never try switching unless it was perfect. Then let's say Microsoft sues the pants off the developers etc for copyright/patent infringement...they would attack anyone and everyone associated with it, and our government would help them of course, because they have a legal case. But if the code was decentralized and anonymous, then they wouldn't have anyone to attack. Their efforts to stop the software developers would be just about as successful as the MPAA stopping movie torrents..... It's just an example, but I'm really wondering what is possible with Git. It seems to me that the failure point might be that the "authoritative" repo in Git must remain known, trusted, and contain the latest changes. Today, it's easy to know where the main repo lives, where to send your updates, etc, and the main developers easily publish that location. But what if the code could not live in one place, because the Feds would raid every server you set up as the "trusted" synchronization point? What if the Feds aggressively attack any server hosting a repo that contains a new unbreakable encryption library. Or like what happened in China, they force someone to shut down their Github repo because they're publishing code to subvert their "Great Firewall". I'm clearly speculating, but I'm wondering if Git is flexible enough to overcome such challenges.