One node of mine has a reputation of 6 right now. Another has a reputation of 14. One has a reputation of 39 and another 81.

What can you learn from that?

That none of the nodes was so controversial, or so obviously wrong that it didnt get downvoted into oblivion. And it also suggests that it won't be a total waste of time reading them (presuming one is interested in the subject matter.)

I find all these arguments about how rep means nothing to be pretty bogus. Positive noderep by itself only really suggests that a node doesnt have many characteristics that would lead it to be downvoted. But i think that a negative noderep is highly indicative of various things. It suggests the node for one reason or another is controversial, it suggests that care should be taken when relying on the information in the node, it suggests that a fair of number of presumably reasonable people thought it was worth receiving negative feedback.

I have written nodes that contained 100% accurate information, detailed code and analaysis but that was presented in an agressive, rude, or otherwise unsocial way. Unsurprisingly these nodes have been downvoted into the negative. And you know what, when I go back and review those nodes I generally tend to wince and think the downvoting was deserved.

So in short, I think that there is a point that postive noderep is probably not a very useful metric, but negative noderep in my experience almost always is deserved.

What can you learn from that?

Admittedly, little to nothing, and I pretty much agree with your post. For nodes that have only a few responses, you might have a rough, and arguably inaccurate idea of how others felt about the various responses, and that could add or subtract weight to someone's comments. Keep in mind, my take is flawed because it's based more on my own personal habits when it comes to voting than anything else.

I rarely, if ever, vote negatively, and typicically when I vote, it's more a way of saying, "Yeah.. I found that interesting or worthwhile. I may not even necessarily agree with it, though more often than not it's a vote of general agreement. It's more in line with other sites that allow others to say that they found someone else's comments helpful ("There were X others who found this review useful..")

Is it scientific? No. Is it statistically sound? Again, no. Keeping in mind the numerous and possibly random way in which some people vote, I'm amazed at how often reputation -- most often on some of the smaller threads seems to fairly accurately (IMO) reflect the quality of a node within a thread. I don't think it's ever been the only reason why I've ever looked at a node, but then again, I have looked at some nodes that I never would have seen if they hadn't appeared in the "best/worst" lists, so I guess - yes, in a sense I've occasionally found it useful. That probably doesn't reflect very positively on my judgement, but there you go...

My question to you would be why show this information at all to anyone but the person who wrote the node, if it's so useless? What does the act of me voting positively or negatively on a node have to do with making the information any more or less worthy of being viewed? Just curious, and trying (but failing) to see the downsides of seeing the reputation of a node prior to voting.

If it were up to me alone and if Perl Monks had never displayed node reputation to anyone but the owner, I wouldn't show it at all. I think it's useful within the context of a thread for arranging nodes in order from highest reputation to lowest and within the context of PM as a whole for showing, to some degree, the collective value of an individual's participation. That's about it though.