Following up Alex's possibly naive handwaving. Read at your leisure. On Friday 14 Jan 2011 13:19:55 Michiel de Jong wrote: > Hi! > > My name is Michiel, I originally founded Unhosted, about three months ago. > One important thing for you to understand about Unhosted is that there is > hardly any code (yet), and although there are a few proof-of-concept > applications that use unhosted technology, these are not real-world / live / > production websites. We hope to create and convert websites to unhosted > architectures during the course of 2011, but as yet this project has only > about 2 man-months of work done on it. So very early days. > > Also, I must warn that despite what many people seem to associate it with, > unhosted is a project that aims to break web2.0's monopolies by giving > open-source websites certain scalability techniques (so that they only have > to publish their source code and don't need to supply so much server power). > it is not at all about providing anonymity (nor to the publisher, nor to the > viewer)
Right. Until and unless Freenet has overwhelming advantages in e.g. performance (which I once believed was possible, I dunno now, we'll see), most unrelated projects which don't have closely related goals will not want to associate with it. For much the same reason that the Wikimedia foundation is distancing itself from Wikileaks. Having said that some of the technologies may be common. HOWEVER, what Unhosted is mainly about according to Alex is applications. Freenet isn't interested in distributing applications. That is, our anonymity/security/privacy goals make distributing applications extremely difficult. Freenet is about data; apps are layered on top of the network, which provides get and put operations (and not much else). > > On Fri, Jan 14, 2011 at 12:43 PM, Alex Rollin <alex.rollin at gmail.com> > wrote: > > > Hello, > > > > I have been looking through the site and wiki for Unhosted: > > http://www.unhosted.org/ > > https://github.com/michiel-unhosted/unhosted > > > > and comparing the desired decentralization goals with that of Freenet: > > http://freenetproject.org/ > > > > I wonder if these two projects are not links in the same chain for > > higher availability complex behavior in anonymous browsing and > > small-world networks? > > I'm sure there are things we can complement! Please explore these, and I'll > be happy to answer any questions you may have about what unhosted is > planning to do, and also to change course on things if that makes more sense > given what you are doing. You are the existing project, and we are the new > guys, so we are the ones who should refrain from reinventing the wheel if we > are investigating similar things. Cool, good to have an expression of mutual respect. > > As for the details of the rest of your post, I must admit I don't understand > it very well. I thought freenode was a way to make sure "no one knows you're > writing" instead of as you say "no one knows you're reading". or maybe it > automatically is both? We aim to provide anonymity for both. However in practice we provide only limited protection against a determined attacker, unless people use darknet (friend to friend connections). > > Can you elaborate a bit on the change that is happening in freenet towards > 'formatting the data'? Chat systems are inserting XML for various reasons. There are also distributed search mechanisms (fetching by keyword isn't always ideal for e.g. space reasons so we've implemented a yaml-based btree). > > What unhosted does is propose a list of per-user resource protocols, and a > list of discovery mechanisms for them. for now, we propose a simplistic > 'SET/GET' KeyValue protocol, and a simplistic 'SEND/RECEIVE' MessageQueues > protocol (so two modules so far). Discovery mechanisms so far we have only > one, which is WebFinger-based. Set/get key:value from the user, stored in a distributed fashion? And send/receive to a user from others? Freenet has somewhat similar things in some of the higher level apps, however it has to deal with anonymous spam, so the chat clients use a Web of Trust. > > Since unhosted is a focus on per-user resources, and freenet is a focus on > anonymized resources, maybe the two are even opposite in that sense if you > know what i mean. But they can coincide if we talk about 'per-nick > resources' instead of per-user, where the link between nick and physical > person stays secret. This immediately breaks down however if you would have > to identify yourself on a traditional website using your supposedly > anonymous nick. IMHO there is almost always an identity of some kind linked to resources, even if it is only a public key that was never used for anything else. For instance, in future Freenet will announce uploads via its chat system. Truly anonymous announcements are extremely difficult, given spam: It is far more efficient to create an identity and reuse it, gain trust for it etc. > > Unhosted is essentially in-browser. I heard of an in-browser anonymity > network called Veiled. Maybe they are more interesting to you than we are. It is in-browser yet it communicates between different nodes? I have heard of such things... Generally Freenet needs good uptime so in-browser doesn't work too well. Except on opennet, but even there low uptime means massive storage redundancy, which we don't really provide, and opennet is so easy to attack that I'm reluctant... Freenet in a java applet would be fun (how do you allow connections to other nodes??) and could get a huge audience, but it'd introduce a lot of new problems... > > Anyway, I'm available for whatever you need from me. > > Cheers! > Michiel > www.unhosted.org -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part. URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20110212/025974e8/attachment.pgp>
