Diagnostics

I find it difficult to believe, but apparently too much knowledge, or the availability of too much knowledge, is not an unabashedly good thing.

Take this:
"The connection to the server has failed" "Socket Error: 10061. Error Number 0x800CCC0E"

Starting about four months ago, I began to get this message routinely on my Outlook Express connection. I found through trial and error that only a restart of the operating system would make it go away.

Doing some reading, on the assumption that there is One Known Error, leads me to believe that it is another example of an error message trying to be helpful by not speaking in gibberish, and so instead it speaks too broadly. Error messages have that as a problem when they want to translate something that means something to the person who understands the code into something that will mean something to the person who does not. I would bet that the actual, original error is the Socket Error and Error Number; further, I would bet that what they actually mean is 'I tried to make a connection within a certain amount of time, and it didn't happen. No specific error was received, it just didn't happen, so I quit."

The assumption in what I read is that it is a transient condition which can be corrected by fixing where the transient error originates. Aye, but there's the rub. Where is the error? So far as I can see, there are three possibilities --

a) the antivirus software is intercepting the message
b) the mail package itself is corrupted in some way
c) the logical connection between the PC and the server has failed.

There is a great reluctance on my part to turn off the antivirus software 'just to see'. What if it works -- does that mean I'm going to turn it off every time I want to receive mail and get this error? Of course not. As for the corrupted files, it's possible that a file is corrupted, but if so then it is a file that gets rebuilt when the operating system gets restarted. Not impossible. And the third, which an additional message suggests as the probable problem ('server timeout') seems the most unlikely, unless it turns out that it really does time out, and in so doing leaves something set up that can only be corrected by refreshing the environment.

I thought of an article I recently read in the Times about people who have to search for medical information on their own because either they have no one who can make recommendations for them, either because they literally have no one or because the people they have either are overwhelmed and simply cannot give the time to them that such assistance would need, or because their support people simply don't know -- there's too much undigested information out there, too much changes too quickly, no gold standards exist.

I have been moderately interested for some time on how good diagnosticians become good -- is it nature, nurture, or what -- but up to now, I've always thought of that as purely a medical question. Now, it appears, it's more.