by joe
on 25. May, 2010
in Misc
with 5,000 views
A few people have asked us how TrustFabric compares to Diaspora. Here are some thoughts.. mostly Joe’s thoughts.
The short attention span the answer is:
It sounds like they are planning to build what we built 3 months ago.
“judge results not intentions” — A little Tom Peters gem which has stuck with me over the years.
The longer version:
There have been a few reality check posts about Diaspora.
I’ll start with the positive stuff:
- Diaspora has the right idea. Distributed and Open makes sense. I think the way social networks are “owned” and live in a centrally hosted environment has bugged most geeks for some time now. We should own and control our information.
- They have the right model. Seeds. We call them Agents. Many people have a distributed social network in the browser idea. Bad idea, I have more than one browser and I don’t want to worry about a very complex thing (my browser) being exploited. I don’t want to keep all my personal information in it.
- They are doing it the right way. Open Source. AGPL.
The list goes on. TrustFabric and Diaspora are pretty similar here. I’m ignoring the little Python vs Ruby kind of details.
Next, some challenges:
- Expectations. I think there is a HUGE disconnect between what people expect from Diaspora and what they are planning to build. People think, new Facebook (pretty things). Diaspora seems to be saying.. open distributed Framework (geeky things). Geeks like to build API’s and frameworks, they don’t like to build the things which users expect a new Facebook to do. Photo tagging and pretty interfaces is not the same as moving GPG messages around with REST.
- Is it sustainable? If people give you money they usually think they are getting something for their money. You generally get one shot. Most projects (and businesses) take about 3 years to gather enough momentum to be self sustaining. I wonder about the longer term plan. Maybe some form of business model would be useful.
- Adoption. How do you get people to us it? Diaspora (when the code exists) is far from the only open distributed social framework out there. Have a look at foaf+ssl for example. Are people using it? Are people using something as simple as OpenID? Do they have a reason to?
- By the Geeks for the Geeks. Is building an open (source) social network the same problem as writing open source software? I don’t think so. You can hack up your own uber fast new web server, just for fun, just because it’s useful to you on day 1. You can share the code and see if others find it useful. It’s a scratch my own itch model. Social networks don’t work like this.. it’s about the network effect, their value is in the number of users. It’s more of a popularity contest model.
I’d like to think TrustFabric is a bit more mature on the points above. We’ve been busy designing this thing since early 2009 and a lot of the thinking comes from even earlier. We have running code and users. We have a practical application and a fairly long list of interested customers. Also, we’ve had a number of discussions with people who want to see a business model and a solution for the adoption problem before they invest money. We’ve had some practise at answering the tough questions.
Make no mistake there is a wave out there. The Open and Distributed wave. We don’t have all the answers, but we have some ideas on how to ride this wave.
We’re giving a (technical) CLUG talk on 25 May (today). If you can code or if you have a brilliant use for TrustFabric, please attend and get involved.
by joe
on 25. May, 2010
in Misc
with 2,318 views
I attended the ISOC INET conference yesterday. A few people asked me: What is TrustFabric?
This is my current 2min summary:
TrustFabric is two things:
1) A Personal Information Management system (PIM)
One place to keep your personal information. An easy way to selectively share your personal information. Think: better way to fill in forms.
2) A Vendor Relationship Management system (VRM)
The other half of CRM. A way for you to keep track of the relationships you have with vendors and service providers you buy from. A way to keep track of the information you shared.
Obviously it’s a classic case of the whole is bigger than the sum of the parts. To do the two things above we have to build on a few other concepts: Identity. Security. Authenticity. Web of Trust. Open Source. Open Standards. Extensibility. Distributed network design. The list goes on, but you get the idea.
What TrustFabric is NOT:
The most frequent wrong assumption is that we are yet another web service which collects personal information, retains personal information and then sells this information to 3rd parties. No, no, no and no.
We’re not a centralised web service.. the idea is that you can run your own instance of TrustFabric (an Agent), in a bunker, under the sea. It’s a distributed system. We never collect your personal information.. it’s your information, it’s inside your Agent. We don’t retain your information. We can’t sell your information because we don’t collect it.
I hope these three ideas make TrustFabric easier to explain to your friends.
Your information is everywhere: at your bank, your insurance company, your mobile phone operator, the stores you have loyalty cards with, all the websites you’ve ever had an account on and probably many many other places you probably wish it wasn’t.
There are of course privacy issues with this distribution of your data, but perhaps more annoying to the average user are the inefficiencies bred by this “system”.
Your mobile phone number changes and all of a sudden you’re having to wonder if the your child’s school will be able to get hold of you in an emergency. Your dentist moves offices and because you’ve always been a cash customer he is unable to let you know where his new practice is.
This inefficiency is being amplified daily as more and more users move more and more of their lives online. What was once just internet banking has now become a myriad of services ranging from personal accounting to online dating. We shop online, we work online, we study online, we store large parts of our lives online.
The world has come so far… there are astronauts tweeting from outer space… we now have relationships with hundreds of businesses and likely thousands of individuals online…. and yet all this information is stored in the the private databases of your service providers. Flickr, Last.fm, Facebook, Twitter, Google Buzz, Linked-in… all ships passing in the night. If you think about it, they are only keeping track of half the relationship. There are no tools for you to manage these relationships from your point of view.
These relationships also last longer than we ever expected… like the y2k bug, the programmers behind your internet banking system probably never thought you’d still be using their software 15 years later. These relationships don’t even have to be as formal as internet banking to be long lasting. The business you booked that great vacation with 10 years ago is still in business but you can’t remember their name and you need a holiday.
As seen recently with the recall of millions of Toyotas, some relationships are more important than others and yet it is often these important relationships that are the most mismanaged. Does the company you bought your current car from still know how to get hold of you?
There is however a revolution happening. Users are starting to make choices based on the efficiency of their interactions and not solely on the financial implications. People want more time with their families and less time being frustrated by inefficient systems.
Unfortunately there aren’t always options and even when there are options the best you can hope for is a decent website or friendly call centre agent.
The world is wrong… and it’s ready for a revolution.
by joe
on 20. May, 2010
in Misc
with 748 views
In Rich Mulholland‘s NetProphet presentation last week he suggested we should not just label all social services as “social networks”.
What’s the niche? What is the core value in each?
Twitter is an information sharing network. You post a link, people RT. That’s the value. Simple. It’s not a conversation tool because there is no room for context information and personally I think DM should never have been a Twitter feature.
What is LinkedIn? It’s your online CV. A business reputation network.
What is Facebook? It’s where you build Me, Inc. It’s where you build the brand you want to be. Rich thinks it’s a Social Operating System.
I agree.
I see Facebook as a monolithic Windows style OS. It has networking, a window manager, apps etc.
Maybe the challenge for Open and Distributed social networks is to be a Social OS built the Unix way. Small is beautiful. A social file system (ACLs etc), networking (name resolution, addressing, protocols), a few window managers to choose from, apps, some form of package management.. all the things we like about Debian/Ubuntu.
We released the second Alpha version of the TrustFabric hosted service today.
Firstly, thank you for the testing, feedback and support so far and keep the suggestions rolling in.
New features in the Alpha2 release:
- Document definitions and validation – Our Document Definition Markup Language (DDML) makes it easy to define XML documents with Forms and Validation.
- Multiple documents – Users will notice a new User Profile document and we’re busy developing some other documents aimed at the first practical TrustFabric applications.
- Binary documents – So you can store photos and PDFs etc.
- Permissions – You can now control permissions at a per group level.
- User profile document – This is the first “real” document. You can now upload a profile photo.
- Email notifications – Your agent will now email you when you receive a relationship request. (We use the email address in your “User Profile Document”).
- Relationship request improvements – When sending a relationship request, we’ll send along your “common name” and an “introduction” so that the receiving party knows who you are.
- UI improvements – Many layout and CSS improvements. We merged the webui and www sites and used the webui look on the new landing page.
- Relationships page improvements – Grouped relationships and actions.
Alpha1 users, please note:
The new document permissions logic is set to not share any of your information by default. In Alpha1 we shared your business card with all your relationships, however you’ll now need to explicitly set which documents are shared with your groups.
