The Internet - Intro Notes - Back to Index

Going Social


After reading Kevin Bankston's assessment of Facebook's new privacy changes I had a look at my own settings and decided to tighten them up considerably. Almost everything is now restricted to "Only Friends" and only "Friends of Friends" can invite me to be a Friend. I don't use Applications but apparently if a Friend uses an Application then it can access your information as if it weere a friend. I tried to prevent that as far as possible using the "Applications and Websites" privacy settings.

I am transitioning from Delicious to Pinboard for bookmarking.


Back in June I mentioned that my OpenID provider GetOpenID didn't work with Facebook because it didn't support OpenID 2.0. I emailed GetOpenID about it but have not had a response. Most websites I log in to either do not use OpenID or they only accept their own OpenID, so there were only two sites where I was actually using it. One was a defunct wiki, which I could safely ignore but the other was, which I do log in to. So a few weeks ago I decided to abandon OpenID (at least for now). I logged in to and removed my OpenID, reverting to username/password login, and then removed the delegation links from my Zenatode site. There didn't seem to be a way to delete my GetOpenID account so that still exists but I will not be using it.


Good article by Andy Oram on why social networks need to be decentralized.


Great article by Brent Simmons on Weblog Software.


ITV has sold Friends Reunited for £25m, despite having agreed to pay a total of £175m for it in 2005


From a recent Ars Technica article by Ryan Paul:

Facebook is launching support for OpenID and will act as a "relying" party, which means that users will be able to register Facebook accounts and log into the service using an OpenID identity provided by third-party web services such as GMail.

I just tried to set this up in Facebook but it didn't like my GetOpenID OpenID, saying "Service type is not supported". It seems Facebook requires OpenID 2.0 and it thinks GetOpenID doesn't support it. I have emailed GetOpenID to see what they say about it.


I highly recommend reading an article called The Elements of Social Architecture by Christina Wodtke. Amongst other thing it is a good overview/comparison of some of the most successful social networking sites in use today.


I have removed the link to the FoaF Explorer view of my FoaF file from my home page since it wasn't really adding anything of use.

I just had another quick look at the Google Social Graph API that I mentioned nearly a year ago on this page. In particular I tried out the "My Connections" example application - entering my homepage URL and clicking "Find Connections" brought up lots of stuff gleaned from my FoaF file and XFN values.


I just read a good article about Getting OpenID Into the Browser by David Recordon.


UserVoice is designed to "harness the wisdom of your crowd".


Twitter and the Micro-Messaging Revolution - a report by Sarah Milstein.


Back in August last year, Paul Gearon blogged about best practice for URIs that identify people.


I have finally given up and registered on Facebook! I was trying to avoid getting involved with any more social networking sites but too many people I know are on it.


MySpace is the latest big name to announce so called OpenID support which is actually nothing of the sort.


Good review of a new URL shortening service with open stats and semantic and geospatial analysis.


Last week at Open Tech someone mentioned something called Weave which is a new project withing Mozilla Labs. Here is their overview of the basic idea:

  1. browser metadata is pushed into the cloud (e.g. bookmarks, history, customizations, etc.)
  2. this metadata is transparently reflected everywhere an individual gets online
  3. we provide a basic framework for easily sharing and delegating access to this metadata to friends, family and third-parties
  4. we build tools and APIs to extend this framework and to provide new user experiences

Note that they are proposing the use of strong encryption to make sure that you stay in full control of your metadata (even people running servers on which it is stored have no access to it).


When I went to log in to my SourceForge account today I noticed that you can now get in using any OpenID provider. I tried it and it worked, though I did of course then have to associate my OpenID with my existing SourceForge account - here are the instructions and FAQ.


Mike Loukides, in his Building Better Silos blog post, has pointed out what I knew already. That although Yahoo! say they are supporting OpenID they are actually doing nothing of the sort because you can't log in to Yahoo! account using your existing OpenID, you have to create a Yahoo! one, which totally defeats the object.


I have just read a bit about APML, the attention profiling mark-up language. I am too busy with other stuff to think much about it now but it seems like something to keep an eye on.


I just checked at rubhub and found that it has now spidered my homepage. Here is my rubhub entry.


Ringside Networks have been working on an Open Source social application server:

Ringside Social Application Server is the first open source platform that enables website owners to build and deploy social applications that operate with existing website content and business applications while seamlessly integrating with social networks such as Facebook.


Yahoo! announces support for the OpenSocial standard and moves to form OpenSocial Foundation along with Google and MySpace.


I have spent the last couple of days looking at XFN, FoaF and OpenID. I had set up an OpenID some time ago at GetOpenID but the only place I have used it to autenticate is on the OpenID Wiki. Although Google, IBM, Microsoft, Verisign and Yahoo! are all now on the board of the OpenID Foundation they currently only seem to be paying lip service to the concept. Yahoo! for example made a big deal about the fact that they were enabling Yahoo! accounts for OpenID acceess, but that only means you can use your Yahoo! account to log in to other sites, not that you can log into your Yahoo! account using your own existing OpenID. Google have been a little better, and in December they announced that OpenID commenting was available for all Blogger blogs. Lets hope things start moving forward a bit quicker now.

Before I leave the subject of OpenID I should just mention how I got it to work for me. When you are asked to authenticate using OpenID you need to be able to give an OpenID Identifier, which is a URL. I use as my Identifier, and it works because I have put the following two lines in the head section of the index.html page:

<link rel="openid.server" href="">
<link rel="openid.delegate" href="">

Now I will move on to XFN (XHTML Friends Network). This is a simple way to represent human relationships using hyperlinks and I have used it on my homepage for two fairly distinct purposes. First I added rel="me" to each link in my list of other web pages of mine, which enables various automated systems to associate those pages with me. Then I added a list of websites of people I know and marked them up appropriately. For example, for a friend who I have met in person I used rel="met friend". To enable this stuff I just had to add profile="" to the head tag of the page.

Once the page was ready I submitted the URL to rubhub for spidering but that was four days ago and it doesn't seem to be showing up yet. Two days ago I submitted the URL to XHTML Friends and it crawled it on the same day, registering me as person 46024.

Finally I will say a little bit about FoaF. XFN and FoaF are both designed to address the same problem but they take quite different approaches - for a good comparison check out the XFN and Foaf article I mentioned back in December. I have had a FoaF file for some time but I had another look at it and have made some modifications including the addition of my OpenID Identifier. Another thing I did was add foaf:homepage tags for all my other pages listed on my main homepage. I guess they are not really homepages in the strict sense but that seemed the simplest way to refer to them - it certainly serves the purpose of indicating that those pages are controlled by me and at least in some sense "about" me. On my main homepage I have added a link to the Foaf Explorer view of my FoaF file.

Actually, while I am on a roll I will just say a little bit about the various social networking services I am using. The one I seem to be using most at the moment is Delicious. This is a very well know social bookmarking service that I started using in earnest last year and I now have 768 bookmarks in the system. I look at YouTube quite a lot and I have a channel there but I have never uploaded anything. I should really try to Digg more stuff but the problem is, when I find a story I want to Digg it takes a while to find out whether a separate report has already been Dugg. Last but not least there is twitter. I created a twitter account early this year but I have not made much use of it yet - like all these things it will only become really useful to me if lots of people I know start using it.


Social networking site Facebook has seen its first drop in UK users in January, new industry data indicates.


DataPortability is a project to promote technologies that are all important in the area of social networking.


Tim O'Reilly comments on Google's announcement of its Social Graph API.


I recently read an article called Battling Social Network Fatigue ... By Going Open in which David Recordon pontificates about moving beyond closed social networks. I commented about the fact that he had not mentioned FoaF (which I had already begun using) and he replied by asking:

How do you see using FoaF to describe different types of relationships? As far as I know it only has "foaf:knows" whereas XFN has a richer set of vocabulary.

I had not heard of XFN so of course I had to investigate, and found XFN and FoaF which compares and contrasts the two technologies. XFN does seem to have some definite advantages and I am thinking about using it. Before heading down that road though I asked myself another question. Is there an easy way of adding an XFN "relationship type" field to my Mac OS X AddressBook? I am not sure so I have just asked the question in an appropriate newsgroup.


Check out Joshua Porter's blog entry about Facebook's Brilliant but Evil Design.


Google and MySpace have announced they are collaborating in a deal which could shake up the social networking industry.


I have started using my account which I created back in March. I have done a couple of bulk imports of bookmarks, for which I had to get my URLs into something called Netscape Bookmark file format. When I tried searching for documentation of this format the first result that came up was at the Microsoft website. Apparently the MSIE Favorites file format is exactly the same as the Netscape Bookmark file format, and it is documented on the Microsoft site. The Wikipedia entry for Bookmark (computers) has a Reference for Netscape Bookmark File Format but it links to the specification at the Microsoft website. Since it seems to be such a widely used standard you would think it would be documented somewhere less, shall we say, partisan? But I guess that is putting the cart before the horse. The real question is why sites like does not support importing bookmarks from a properly specified open file standard? I have just done a bit more research into this.

XBEL is an XML bookmark exchange language which is being used as the native bookmark format in a number of modern browsers and it is specified by a formal DTD (the Netscape format documented at the Microsoft site is not even valid HTML - and if you do any obvious modifications to make it valid then does not recognise it). I would hope that is at least thinking about supporting bookmark import from XBEL files.

Before I started using I did not really use bookmarks as such. Sure I had about half a dozen sites bookmarked in Safari because they were sites I used very often. The main one was my own website (this very site) which is split into about three dozen broad subject categories. Within each category I used to have a page of links and those served as my online bookmarks. The trouble was that some URLs did not fit neatly into any category or clearly belonged in several. Even without the social aspect of, the model of just having a single list of tagged bookmarks has great advantages. Each bookmark can have any number of tags associated with it and you can choose to limit your view of the list to only those bookmarks with a certain tag (or tags). Once the social aspect is added this allows for the dynamic evolution of a shared folksonomy.

I have now realised that a site like is far more than a bookmark manager, and actually provdes a viable alternative to traditional search engines like Google. Google is great for certain purposes. For example, suppose I run application foo and it generates the error messages "Error 444: Could not grok flubfile". All I need to do is search for "foo" and that error message and if there are any reports of the same problem I am likely to find them. On the other hand, suppose I am searching for web pages about astronomy software for Mac computers? Yes I could search Google, but I might get better results by going to If nothing else I might get some more up to date results because it sometimes takes months for Google to index new sites but as soon as someone bookmarks a site using those tags it will become accessible.


I have never used Twitter myself and had not thought much about how it might be useful. Adam Engst was skeptical but now enthuses about it in Confessions of a Twitter Convert.


The Sun Babelfish Blog has some good stuff about social networks, OpenID, FoaF etc - see the Semweb Category.


With all these different social networking sites around it is hard to decide which one(s) to use. I do have a MySpace account but I use it as little as possible because it is owned by the "Dirty Digger", Rupert Murdoch. The one I settled on is Multiply, which has better features, looks about a million times better than the hideously designed MySpace, and generally seems to be less evil. I have recently been invited to join Facebook and I declined because I have too many Internet accounts already. I am glad I didn't because it seems that Facebook may be some sort of front for US intelligence services. See this short presentation which asks "Does what happens in the Facebook stay in the Facebook?".


Why do activists continue to make use of MySpace? - Calling for a new online social network.


AOL are getting onboard the OpenID bandwaggon as well now. It looks like it could really be taking off so I guess I should start using it myself now.


Microsoft have jumped on the OpenID bandwaggon.


About a week ago I mentioned OpenID phishing vulnerbilities. Well Simon Willison has implmented a couple of phishing protection features in his idproxy system.


A friend of mine has been playing with OpenID and I was thinking it might be an idea to start using it. I changed my mind after reading OpenID: Phishing Heaven on Ben Laurie's blog.


I got an email from plaxo today which asked me to "Please take a moment to update me with your latest contact info." The reason I got it was that someone on a Yahoo! group of which I am a member had started using plaxo. I was dubious about responding so I did a bit of research. Search for plaxo on The Social Software Weblog to see why people have a problem with it. Ian Gregory 2010