- Michael Loukides talks about Social Privacy, how we can make sure our website meant to be not shared with the public stays secret or how in general we are to the mercy of any web application to not publish our data.
- As you might have heard there is some discussion about the usability of OpenID. It might be only suitable for geeks right now because many people do not think that a URL is their identification. They might think more like it’s their email address. And thus there have been some attempts to come up with a mapping from E-Mail address to an URL and the latest one came from Brad Fitzpatrick with this proposal.
- Brad Fitzpatrick was earlier talking about the launch of the Google SocialGraph API and this also provoked some reactions regarding privacy. Here I discuss with others over Tim O’Reilly’s post on the O’Reilly Radar about the launch of the new API. He continues in another post about a potential „pain reflex“ in now suddenly seeing how much of our data is public. There is also another post by Joshua Porter about the same problem. My take on this is below this list.
- I also blogged about FastCompany.com now being a social network sort of website and Brian Oberkirch posted about how he thinks it’s a bad idea if now every company wants to be their own social network.
So the main discussion seems to be about the Social Graph API. For quite a while now service providers have been told to use Microformats such as XFN to markup their data for better machine readability. This seems to be somewhat successful, Twitter et al. are doing it.
Now the SocialGraph API comes along and suddenly makes all these relationships visible. So people get conscious about this problem. Of course one can always say that this information always has been public on these sites (just go and visit http://twitter.com/mrtopf to see my public list). But it’s a new quality of that information which is now available.
So who is to blame?
Probably nobody at this stage. You cannot blame Google because it seems that this is exactly what Microformats wanted to enable. You can also not blame Twitter as they are just doing what seemed to be the right thing.
What is needed though is a discussion about this and what the best practices in these cases should be. One option would be the opt-in for such XFN-markup but of course those links would still be available although not that easily extractable. Making it more private would probably harm the site’s usefulness. Removing them might also harm social services such as dopplr which make good use of them to find your contacts on dopplr.
And how decides which links to show or not to show? The one linking to people or the people who got linked to?
So it’s many questions to be answered and I wonder what a good process might be to answer them. Surely there is the around but of course they cannot just publish guidelines and the problem is solved. A broader discussion is needed and for a start maybe an exact description of the problem and coming from these some proposals to solve it.