Bag of Spoons
Just off the A1(M)

Thu, 10 Sep 2009

Still waiting for an open semantic social site

Long-term readers (anyone?) would know what I've been interested in the possibilities of the Semantic Web as a way to make some of the data on the web more useful by giving it some context. I've written before about possible semantic social systems utilising the FOAF data model. This is a file format that can store details of a person and their relations to other people. It can link to FOAF files belonging to those other people to allow their net of connections to be explored. One advantage of this over sites like Facebook is that it allows each person to control their own set of data without relying on a third party and another is that the data can be processed by various software or web sites.

What I would like to see is a piece of software, either running on a PC (or other device) or web site that reads my FOAF file and then allows me to do things like viewing the latest updates that my friends have made to whatever sites they use, based on data in their FOAF files. This could be blog posts, music they have listened to, photos they have uploaded or anything else that can have an RSS/ATOM feed. I can do something like that with friendfeed, but that relies on people joining that particular site or me creating profiles for my friends.

FOAF has been around for a while now, but has not really become mainstream. I know that a number of sites can generate FOAF files based on the data they hold, e.g. the identi.ca microblogging service (my data).

I think one reason for the lack of adoption is that it requires a bit of technical experience to get started. There are various sites that can generate a file based on data entered into web forms, but then you have to upload the file to a web server you have some control over. My own file started with one generated by FOAF-a-matic and has since grown as I learnt about more possibilities. I tend to just use a text editor, but the RDF format is very strict about the data structure and mistakes make the file unreadable by software. I think there must be some tools out there by now that make it easier to update a file. What it needs is a button you can click on in your browser that extracts appropriate data from a site belonging to someone you know and inserts that data into your FOAF file.

Another issue with this sort of data sharing is that it does not provide a way for you to limit exposure of certain data to selected people. I don't include things like my date of birth, address and phone number in my file as they could be misused by some of the less desirable abusers of the internet. I have thought that I could have something on my server that allows friends to access certain data using their email address as a key. Email addresses can be stored in the file in hashed form to make it easy to check for a match without giving too much away. I realise that it may be fairly simple to guess an email address if you know a person's name and web site, but I don't know if the bad guys are doing that. In any case you could email a security key back to the given address to permit access. I don't have enough experience of web programming to know exactly how this would work.

Obviously many people do not have their own web server where they can upload arbitrary data, although many may have some web space provided by their ISP. Ideally we would all have our own domains to prevent issues with details changing when you move to a different provider.

I just felt the need to get this stuff down after thinking about it today. I know there are some projects out there that are working on stuff like this, but I'm not aware of any that past the alpha stage. I'll be happy to be told otherwise. I still harbour a desire to do some programming in this area myself, but have not found the time. Foaflib looks like the Python library I wanted when I looked into this a while back. I'll try to have a look at it. It may allow me to achieve my aim of generating the root page of this site from my FOAF file.

I like to think that Facebook is not the future of the web. Is there hope or has Tim Berners-Lee's vision been forgotten in the gold-rush?

[22:16] | [/Internet] | Comments | G
blog comments powered by Disqus

About

Categories

Calendar
< September 2009 >
SuMoTuWeThFrSa
   1 2 3 4 5
6 7 8 9101112
13141516171819
20212223242526
27282930   

Blogging
Subscribe if you like
XFN Friendly

Adverts
Linux.org
Get Firefox!
Flying Spaghetti Monster

Hosted at VeloceSystems