i-sako.com

In my previous post, I mentioned that I have started using XFN when I make links to people I know and think highly of. The idea (that you can identify your relationships with friends and colleagues in your hyperlinks) is interesting enough, and it’s easy to do, so why not? (Well, actually, I can think of a few reasons not to, but I’ll get to those later.)

XFN in brief
In my previous post, I identified Zachary as an acquaintance, someone I know by name and have had a few conversations with. I also indicated that we have never met in person and that we work in the same field. I didn’t explicitly say any these things, however. I simply embedded them in the link to his site (as I have done again in the link above). My tech-savvy readers are probably already aware of this technique and how it is done, but it involves adding specific attributes to the XHTML <a> tag. The XFN specification provides a number of different attributes for identifying friends, sweethearts, relatives, coworkers, neighbors, “muses,” and just about anyone else you might have a positive relationship with. Different relationships get different attributes, which can change over time as the relationship evolves (i.e., that cutie at the office might have been identified as your date last year, but this year she’s your spouse). Although this information is transparent to the casual reader, it can be read by search engines and other things that can parse XHTML (like geeks who view the source of their friends’ sites—not that I am, er, that I know anyone like that or anything).

It seems kind of intriguing to think of the possibilities that could result if this becomes a widespread practice. Years down the road, people could do a Google search for information about their genealogy, for example. Or, as a slightly more near-term example, companies might be able to better tailor product pitches to your unique social situation (and they won’t have to pester you for information about your friends and family, because they would already know). Although the current XFN specification doesn’t touch on this specifically, it would be an interesting experiment to see if regular use of XFN could cause, for example, a Google Images search for, say, “Sako’s coworkers” to actually turn up images of the people I work with. (See, there I did it again!)

The Hazards of Friendship
On the other hand, I know from personal experience that making assumptions about relationships can be tricky. In fact, the title of this entry comes from one such experience. There was a girl I knew in high school (we’ll call her “RM”; if blogs had existed back then, she would have qualified as a “crush” in mine, but I would not in hers—an example of an XFN asymmetrical relationship) who really wanted to set me up with a friend of hers who had just recently enrolled at UC Berkeley (we’ll call her “NB”; her relationship with me was similar to mine with RM). I hung out with NB sometimes, and we even flirted a bit—okay, more than a bit, but always in jest, or so I thought—but my real aim was always to spend time with RM. Anyway, one day I made the mistake of referring to NB as an “acquaintance,” a label that really seemed to hurt her deeply, which I felt really bad about. Our once-playful friendship never recovered, which also caused my relationship with RM to suffer. “Hell hath no fury...” and all that. 

You assume too much!
That experience made me a little leery of using that particular term, but the reverse is also possible. Have you ever met anyone who become way too friendly way too soon? Around the same time as the RM/NB fiasco, I had a neighbor who was irrepressibly gay. I knew this because he asked me if I would like to sleep with him less than an hour after we first met. He was quite charming about it, in his own way, but I’m just not that kind of guy—really, just ask my wife. That can be uncomfortable when it happens in person, but if it happens online, now all of a sudden you’ve got search engines caching it and what not. Messy.

Also, what are the implications (both ethical and technical) of implying that a relationship that does not, in fact, exist? I’m not sure at this point what, if anything, would come of such behavior, but it’s kind of interesting to think about. 

Anyway, please don’t be offended if I assume too much (or make to little) of any relationship we may have, dear reader. In return, I promise not to flip out if you don’t return my crush. 

Incidentally, XFN is partly the work of Eric Meyer, after whom we middle-named our son (so obviously Eric qualifies as a muse). Hmm...Eric and I both do Web-related work, does that mean I should also list him as a colleague? Choices, choices, choices.