- From: Danny Ayers <danny.ayers@gmail.com>
- Date: Sat, 31 Jul 2010 08:06:27 +0200
- To: Mischa Tuffield <mischa.tuffield@garlik.com>
- Cc: Public SWXG <public-xg-socialweb@w3.org>
On 31 July 2010 02:24, Mischa Tuffield <mischa.tuffield@garlik.com> wrote: > Hello All, > Many have been banging on about issues surrounding the Facebook Like button > and its proliferation on the web. My concerns are that services which are > hosting this "social" component are not being transparent wrt to what are > the capabilities of the Like button, and that they are not being upfront > with their users about the "opt-out" nature of the button. > I have described this in detail on my blog : > http://mmt.me.uk/blog/2010/07/30/the-facebook-like-button/ I'm not sure the onus on transparency should be on services which host the component, rather that Facebook should be more upfront to its users that by signing up with them, they are agreeing to be tracked. I don't know what Facebook tell folks that use the component, but wouldn't be surprised to find that there wasn't much mention of the potential "covert surveillance" aspect. But for a large proportion of Web users (e.g. those for whom the Web = Internet Explorer) such warnings would likely be ignored, the perceived benefits of getting more utility out of Facebook outweighing any malware concerns. Whatever, it does seem like the horse has already bolted on this one. While it's reassuring to know that it is possible to dodge the tracking (with AdBlocker Plus or whatever), I can't see much mass appeal in such an approach. Even if such functionality was built into the browser, it wouldn't necessarily get used. Just a thought on raising awareness of the issue: an iFrame that looks like Facebook's Like button, but actually calls a proxying server which in turn does the call that Facebook expects, but also presents the person browsing with details of what they just told Facebook about themselves (and perhaps a link to http://panopticlick.eff.org/). Incidentally, not long ago I put together some semwebbish analytics code, and in the process reviewed approaches to cross-domain Ajax - notes here: http://blogs.talis.com/n2/archives/770 Cheers, Danny. -- http://danny.ayers.name
Received on Saturday, 31 July 2010 06:07:01 UTC