Comment by sanity
2 years ago
I didn't post this but I'm the guy in the video dressed like Tom Scott. This talk was 8 months ago but everything is still relevant. We're very close to an initial release which I'll (of course) announce on HN. In the meantime I'm happy to answer any questions. You can learn more about Freenet at https://freenet.org/.
Could you change the name? Since there hasn't been a release yet, it would break nothing, and there's already a project (that you also started, but that is used by more people, since it is released software) that was named Freenet, that would probably like its name back.
I like this new project though. It seems cool.
Seconding this.
Reusing the name of a longstanding software project for a similar but distinct project with wildly different security guarantees is hazardous; it means that existing documentation which directs people to use "Freenet" in a specific way may expose them to unexpected risks. Please reconsider.
The name belongs to the non-profit and not a specific codebase, the previous codebase had itself a number of fundamental rewrites (eg. with the 0.7 release) and retained the name Freenet through them - this is no different.
The name change decision was made over a year ago after a long debate and very careful consideration. There is risk but risk is inevitable if you want to make progress.
4 replies →
tribbling this.
'freenet' is an isp in my country, and if this does not hint at it enough, too generic of a term
Thanks re: liking the project.
I'm not going to change the name again. I carefully weighed up the pros and cons over the course of a year - debating the issue with those that disagreed. Eventually I made a call as the architect of Freenet. It's not without risk, but risks are sometimes necessary.
People are entitled to disagree but I'm not going to relitigate it at this point.
Very nice to see progress on this "modern" version!
I have one question for you: how would you compare Freenet to Veilid (https://veilid.com/)?
It's always tricky to do head-to-head comparisons as there is so much detail with these projects.
Freenet is designed to be a complete drop-in replacement for the world wide web, handling the discovery, distribution, and execution of an ecosystem of decentralized software. It's a platform. Similar to how installing a browser gives you access to the entire web, installing Freenet gives you access to all decentralized services built on it.
Freenet's most unique architectural feature is that it's a global key-value store where keys are cryptographic contracts that control what values are permissible under the key and how these values can change. This is the key (pun intended) idea that makes it so general purpose.
My impression with Veilid is that it's more a set of tools and libraries that can be incorporated into software - but it isn't a platform in its own right that can allow software to be discovered and distributed.
Think of it like the difference between buying a car (freenet) or a crank shaft that must be integrated with other components to be useful.
Using hash-of-validation-WASM as a state key is a really cool idea and I'm excited to see how it goes operationally.
Using (blinded) proof of donation as an anti-abuse signal is a clever band-aid but seems fraught in the long term. Is there a plan to switch off of it to something less centralized, or is it staying this way until it causes problems?
> Using hash-of-validation-WASM as a state key is a really cool idea and I'm excited to see how it goes operationally.
Thank you :)
> Using (blinded) proof of donation as an anti-abuse signal is a clever band-aid but seems fraught in the long term.
Agreed.
> Is there a plan to switch off of it to something less centralized, or is it staying this way until it causes problems?
These blind trust tokens are a proof-of-concept because while it is anonymous and has the benefit of funding the project, it is centralized which is far from ideal.
Proof-of-work could be used as a decentralized mechanism for mitigating abuse but I don't like deliberately burning energy. I quite like a concept called "proof-of-trust" that I describe here [1], but the design needs to be fleshed out.
[1] https://freenet.org/blog/799/proof-of-trust-a-wealth-unbiase...
The primary non-technical challenge to all these distributed storage system is the storage of illegal data. How does this protocol deal with this issue?
Freenet side-steps this problem because it is more of a communication medium than a storage medium. It also isn't optimized for the distribution of large files that are the most likely to pose copyright issues. It's not a substitute for BitTorrent.
I read the info on the main page and don't see how it's different from web3 and some other crypto related technologies.
Crypto is mostly about building global decentralized ledgers, think of Freenet as a much more general global decentralized computer. I suggest watching the video if you can.