While reading these (1, 2) two pieces this morning which discuss respectively the potential for centralized or semi-normalized tagging and tagging folksonomies, I was reminded of a piece I wrote earlier on network applications and the browser. In it I discussed the innovative Amazon Light 4.0, which amalgamates network services in a single interface. Here’s the bit:
On the one hand, it’s just an alternative interface to Amazon.com. One that doesn’t do a lot of what Amazon does. On the other, it’s an intriguing amalgamation of lightweight web services (small W, though I won’t get into the current WS-* backlash implications here) that together form a different, and in my view, superior, user experience. Amazon Light 4.0 grafts together Amazon, Gmail, Blogger, del.icio.us, Netflix and iTunes in some weird, monstrous, super-webapp. It’s at once less and a great deal more than Amazon.com is alone.
Amazon Light, I would argue, is an excellent example of first order integration. Simple aggregation of services, with the merest hint of service integration. What does that mean? Well, Amazon is not actually integrated with Netflix, but because they’re both responding to the same search terms they may seem to be.
If the enterprise portal market has taught us anything, however, it’s clear that the next step for many of these services will be second order integration. That might mean tieing Amazon and Netflix together more tightly so that they’re both working not just off the same search term, but same customer. But the more likely approach from where I sit is the simplest – integration through metadata. Tieing del.icio.us to Flickr. Which brings us back to the two links above.
What they discuss and digest in part is the difficulty in relating tags from one system to another, one user to another. Despite the cited difficulties, however, I believe there is already an ability to relate tags from differing network services on several levels – voluntarily or otherwise. The differences between Flickr tags and del.icio.us tags is hardly insurmountable.
On first glance, this seems like a remarkably good thing. I could blend pictures of and links about, say, Petra. While there will certainly be hiccups and inconsistencies due to the human generated variation intrinsic to author based tagging, it would work as often as it didn’t, I think. But take that a logical step further, and think about linking users rather than tags.
This would be no more difficult technically than the basic tagging approach, I shouldn’t think, but has dramatically different implications – particularly for user privacy. In my case, it’s not difficult to envision someone linking my various Audioscrobbler, del.icio.us and Flickr identities. What would they gain? Quite a bit, I think. They’d be able to figure out that I’m a big Pearl Jam fan, visit Colorado often, and read a lot about Microsoft and open source. Hell, they could even figure out what kind of camera I had from the Flickr metadata or what media player I use from the Audioscrobbler plugin.
I don’t consider that information private (or I wouldn’t have shared it in the first place), and taken by itself, I have little concern it will come back to bite me. But begin to aggregate and integrate various services and a better picture of me and my habits begins to emerge. Again, not a particular problem. But what do you think that sort of information is worth to…say…Acxiom? They pay tons for access to spending habits, while I share a wealth of personal reading, listening and travel habits with the world…for free.
It’s free, of course, because it’s the sharing multiplied out several thousand or million times that both incents and rewards me – the two things every good social-application should do. But I do fear that it’s only a matter of time before big money marketing and customer profiling firms recognize this data for what it is – a potential gold mine – and set up shop, pulling information out as fast as I can put it in, building a much better picture of me than I want them to have. Second order integration and federation of the network services like del.icio.us or Flickr, unfortunately, would only facilitate this process by providing would be data voyeurs a central entrypoint.
The fault here, I should point out, does not lie in the services themselves. They are all completely voluntary, and some – like Flickr – provide a means for securing posted content to only a selected group of contacts.
I’m sure I’m not the first to throw the flag over the potential for merging and linking of our offline and online data. I also don’t mean to imply that this is all a doom & gloom scenario – it’s not. But it is important to remember that what you share publically could – and maybe will – be used for things you might not intend down the road. That fact is not likely to change my behavior anytime soon, as I enjoy the benefits of social-applications too much to easily give them up, but the inevitable talk of service integration will be setting off silent alarms for me. I don’t mind if all of you know these things about me, but I do mind if Acxiom does. And if the worst does come to pass, well, I want to be able to say I told you so.
Update: Someone asked for centralized tag search and presentation? Well, Technorati delivered it.
