Commander Zot wrote the following post 3 months ago:

In order to implement controllable privacy, one needs to start by having identity. This is an easy problem to solve in a centralised network. One service contains all the login information - and therefore knows who everybody is and can assign roles and access rights between them.

On the decentralised web (which is the way the web works incidentally), privacy is a lot harder because identity is a lot harder. Your machine might know who you are, but my machine doesn't. If I need to make a photo private so only you can see it, I need to know who you are.

How to solve this problem? Distributed authentication.If I need to know who you are, I'm going to ask your server to prove to me who you are. Then I'll show you the private photo.


Friendica does this well.

Red takes this to another level entirely. As you browse decentralised services across the internet, you will casually notice that you are logged into some of them and they greet you by name. You didn't have to do anything to make this happen. You're just surfing the web, and you can see stuff which is only visible to you - because the site you are visiting is able to verify who you are. But you needn't worry about your privacy because we can't track you across the internet. All we know is that you are here. We don't know where you go from there. There is no central website or data harvesting ability. It's just a bunch of independent computers with independent content. And you can move - you aren't authenticated to a single server or DNS address, you're authenticated as you. No matter where you are.

And anywhere you go within the Red universe, we can make content visible to your eyes only.

What kind of content are you likely to find on Red?

Anything. We're building an open source content management system on top of a decentralised communications network - with internet iscale access control. We don't have any concept of "friends". That's sooo 1995. We just publish stuff privately and create channels of things we wish to share, and share these with whom we wish, because we know how to prove that those viewing the things we share are who they say they are. This could completely change how you look at the internet. It is how the internet should work. Blogs, forums, small business websites, dating sites, wikis, customer relations sites, social networks, photo sharing sites, issue tracking sites and more. Just about anything you can do on any other website, we'll be able to provide in Red. Except we can also communicate with other sites and control who can see stuff. If you use a CMS today, you're limited to providing communications and permissions with those that register on the site. Red is a matrix of independent CMS installations that can use communications and permission controls to become a part of something much larger than any one website. You control the content. There are no user agreements or terms of service - unless you use somebody else's server. If you do this and violate their terms of service, you can take a thumbdrive containing the essence of your identity, and find another site. All your permissions, all the channels you communicate with - are still there.

When somebody asks me if I'm on a "social network" I can only laugh. That's like asking me if I have a telegraph.

• Create a field in the profile for email addresses. Multiple addresses can be specified with some UI to add more addresses, one per field.
  
• Make it possible to decide whether this email address should also be shown to profile viewers or only be used for searches. Or never show it at all. (If the user doesn't want to be found by email address, he simply doesn't fill in any.)
  
• Create a hash of each address and put it into the directory.
  
• Give the user the possibility to upload an address book.
  
• First step: Create hashes of all email addresses in the address book and look for them in the directory, find matches. Don't save anything.
  
• Maybe later (not sure if that desirable): Save the names from the address book and the corresponding email address hashes (but never the actual addresses!). Notify the user when someone from the address book joins Red.

• Searching the directory for possible matches from a large address book might be quite some task. I, e.g., have an address book with about 600 members, many of them with more than one email address. And some people have thousands of email addresses in their address books. I'm not sure if and how such a search could be done efficiently.
  
• Is it good to give users the possibility to upload such private data, easily abusable by evil Red server admins?
  
• At least if the data is saved, it gives admins more possibilities to create relationship graphs even for people their users are not (yet) connected to.

Commander Zot wrote the following post 4 months ago:

You're right - nobody is going to take down the Facebook in its own terrain. It's the alligator. If you're in the water, you're in the alligator's domain - and you're going to get eaten. So if you're building a "social network" - good luck and all that, but you're stuffed. Facebook doesn't have a bright future, but contrary to what some "visionaries" predicted - the future social web does not currently belong to decentralisation. At least not today. It will when the sleeping giants (the telecoms) wake up and realise that they're pulling all the strings and that they're the ones who really own the web. Not Google, not Facebook, not Amazon or Microsoft or Apple. This is also why Google is trying to learn how to be a telecom, because in that terrain, Google isn't predator - but prey.

But that web revolution is probably 5-10 years out.

Friendica is basically a social network, but with some "free and open" concepts built in. So it's in the water. A very small minority will find its freedom and privacy useful. Most have already moved on. Even an alligator might let a viper go as long as it isn't too hungry and the snake isn't seen as a threat.

I still don't think you've been listening because you're still thinking of Red as being in the water. I'm not playing the Facebook/SocNet game any more. I'm playing a different game - and it's not in the water. Facebook this, Facebook that - yada yada... I'm not building Facebook. I'm not building a Facebook competitor. Red has nothing whatsoever to do with Facebook (or G+, or Diaspora, yada, yada).

Red isn't a social network. It's website software which has the ability to matrix into something larger than itself. That's pretty much it. That isn't sexy to the masses, and it certainly isn't sexy to any VC's or even the Robert Scobles of the world. It's just a server appliance which we'd be particularly good at building after the Friendica experience, and it puts together some things we happen to do well into a combination which doesn't currently exist. It also starts over on things which we didn't do well and does them better - and provides a framework for others to build on. Some people might find it useful, especially those running niche online forums and blogs which are stuck in their centralised little islands. We've mentioned dating sites and small businesses. We can also cater a bit to that small minority of people who found freedom and privacy useful and found Friendica to be a safe refuge from Facebook - but there aren't many of these.

But this isn't Red's focus or raison de etre, it's just one thing that Red can do. For better or worse Facebook shaped the communications landscape and altered people's expectations of online communications and what features it should provide. Most blog and forum software are "pre-Facebook" and provide somewhat primitive communications. Many people demand more. We can do better than that, again based on the Friendica experience.

Decentralised access control is our terrain. Some/most would not see this as a market or audience worthy of interest.

But they'd be wrong.