My EE teacher solved the problem of drunks pissing on the wall of his house by running a wire along it, and connecting it through a current-limiting capacitor (couple nanofarads, X2-rated) to a safe but unpleasant value.
A suitable value of a “leakage current” could be a good deterrent.
your EE teacher is a liar, a stream of urine can’t conduct electricity.
If it is fragmented, true. If it is continuous, not separated to droplets, false.
Edit: Consider a short distance of the actual pee bridge, with the rest a continuous wet trace on the wall. Fairly different scenario than the usually debunked peeing-on-third-rail myths.
Edit2: Even the fragmented stream can be a danger if the voltage is high enough to jump across the air gaps. Not the case here, but high voltages can jump counterintuively far; many a hapless tech ended his career as crunchy bacon after standing up at a wrong place.
Citation Needed
Funny, we have metal light poles here too. Some probably older than anything in San Fran. Wind blows a lot harder a lot more often here, too, and they are often hit by lightning. Dogs pee on them all day. Have never seen one fall. Why is it always the dog’s fault?
Perhaps the presence of near daily salt-water laden rust-driving fog and constant humidity has something to do with it? Maybe they have been hit one too many times by any one of the 7M people in the area? Perhaps non-metal light poles are in order in such an environment, and this is less about nature than it is about bad human decision making?
Not true at all. Ask any farm kid who has peed into the weeds and hit a live electric fence.
The Mythbusters covered it a few years back. IIRC, a normal human’s pee stream would break apart depending on the distance/intensity, but it’s not impossible.
You live in an urban area.
Oh?
3rd rail-distance is much different from electric-fence/electrified-house-distance.
I mean, really – you city folk believe everything you see on TV? No wonder we keep ending up with these Republicans in office…
Exactly. Been there done that…
I imagine that since little kids seem to be able to shoot a stream about 20 feet, there are some slightly different stream characteristics as compared to an OLD.
I busted their half-arsed test when I was 8.
Perhaps this would be a good application for hydrophobic paint.
I 
Finland! 
There’s even a painless way to do the test.
Involves a LED and a high-gain transistor. Use a low voltage (5-12V) and a suitably high resistor (100k…1M) for the transistor’s base to limit the current that goes through the body. If the stream connects, the LED lights up. (Possibly use a buzzer for acoustic indication.)
Aren’t those being voted in rather by the rural or semirural/suburban contingent?
I would not depend on a low-voltage test to tell me about the behavior of a high-voltage system. Air ionizes with high voltage. Granted, very high voltage, but still…
People like to say that, but the rural numbers just ain’t there.
Or maybe we’ve all peed on the fence one too many times…
True that. However 5V DC is a good substitute for 230V AC for a simple low-current conductivity test.
Not all conductivity measures have to be hipot ones. 
Related thought… How could we test/estimate a high-voltage capacitor breakdown voltage without damaging the dielectric? Monitoring the leakage current vs voltage?
Watched our dog do it once.
