Anyone who blogs has been there — the temptation to say something just slightly negative about someone you know. Then the corresponding inhibition as you remember that this is a public forum, and that you can’t just go talking trash about people in a public forum. So you come to the slightly devious compromise that you’ll say the negative thing, but without naming any names.
I’m here to tell you that this is the worst of both worlds, not because the message will not reach its target, but because of collateral damage (countless innocents will be wounded in the process).
Extended hypothetical: Imagine that your friend Jerry brings his child Beelzebub over for tea, and Beelzebub literally (not figuratively) kills your cat, then breaks your crystal, then dials 911 (bringing the police out for a visit), _then_ starts a fire (bringing the fire department out for a visit). You might be tempted to blog somewhat darkly, naming no names and giving no specifics, about friends who bring over underdisciplined underage guests. The only problem with this is that Jerry will probably never read your blog, but _every_ other parent who has ever brought a child to your house will be sure that you’re talking about them, and will glumly reflect on every instance when Donald or Daisy was so badly behaved that s/he had to be prompted twice before s/he said thank you to the nice man.
As an example, there’s a rant I wrote about engineering management and product management. My intent was to be somewhat wry about the professions in general, and reflective about myself. But I named no names, and so I guess it’s not surprising that my manager at the time seemed to think I was talking about him (which had never occurred to me). (Luckily he has a sense of humor, so the only result was that I got occasional text messages on my phone which read (in their entirety) “Venga venga venga!!”)
As another example, I was once taking questions in a panel discussion, when someone asked what kinds of research in this area we _didn’t_ like. Taken aback, and not ready for the question, I nonethless mumbled something truthful but anonymized. It wasn’t until several days later, at the very end of the conference, that a researcher I very much respect rolled up on me and said: “So I assume that you were talking about my paper with Green, Brown, and Black?” As I thought about it, it was actually not a paranoid thing to think, but quite plausible given what I’d said. I was glad he approached me, because we got to straighten it out. But what’s the ratio of instances like this that you get a chance to correct to instances that you’re blissfully unaware of?
You could draw a couple of different lessons from this. One would be some variant of: if you can’t say something nice, then don’t say anything at all. But that’s really boring, and I must therefore reject it out of hand. The alternate lesson is: if you can’t say something nice, then make sure the right party pays. Names may not be enough — you might need addresses, social security numbers, GPS coordinates to remove all ambiguity. Don’t be shy – remember that you are just protecting the innocent!