Trust, control, and scraping Facebook
Judi Sohn’s got an good perspective on this week’s high-profile social networking kerfuffle:
This week’s blog storm centered around Robert Scoble. He temporarily lost his Facebook account because he got caught trying to scrape what he erroneously believed was “his†data into Plaxo using a script that violated Facebook’s terms of service….
It’s not about data portability. It’s about trust.
Well, it seems to me that it also is about data portability (Dare Obasanjo aka Carnage4Life and Nick Carr both have excellent posts on this). In addition to privacy concerns, one of the reasons I don’t put much data into Facebook or its applications is because I can’t take it with me. Visual Bookshelf, for example, is something that I’m continually tempted by; but if I’m going to go to the trouble of inputting all of my library, I want to be able to access it and display it from any site — and continue to use it if I decide to leave Facebook. Alas, social networking sites today generally regard the data as theirs, not their users; and so they don’t make it easy to move it.
Still, the trust aspects are paramount here, and there are at least a couple of interesting vectors:
- between Scoble and his friends. When I added him on Facebook, my intent was to give him access to my Facebook information and create a channel by which we could communicate; sure, he could proceed to scrape my information and use it for all kinds of purposes without asking me, but as Tara Hunt tweets “I’m appalled that someone can take my info 2 other networks w/o my permission. Rights belong 2 friends, too.†Did he violate our trust relationship by doing the scraping without asking for permission? He says he was just doing some testing and threw away the data; do I trust him not to have left any stray traces? Also, Scoble’s got a lot of friends, and Judi comments in her post that for the vast majority of them, he is the equivalent of a magazine publisher and the “friends” are his subscriber base/audience. These are very different connotations than we usually think of for friends; how does this affect the relationship?
- between Facebook and its users. Scoble used a script that he knows violated their rules of service; are they right to reinstate his account after his apology when his posts make clear that he still doesn’t think he really did anything wrong? On the other hand, Scoble points out that when there automated defenses flagged him as a potential security violation, they erased him without warning. Does this mean that all Facebook users are essentially living on sufferance?
Interesting issues. Shel Israel’s summary suggests that the overall impact is nil (although it helped the day go faster), and I tend to agree that it didn’t affect the reputations of Facebook (even though they looked good here), Plaxo (who’s already generally regarded as evil) or Scoble (who is known for doing stuff like this). Still, it’s a fascinating microcosm of a couple of the issues that are likely to come up again and again in the social networking space; so it’s well worth thinking about.
Update: Ed Felten’s It’s not about data ownership makes some very important points here and is well worth reading.