"(Likewise, your apparently misinformed opinion of condoms is also irrelevent.)"
Not sure what his misconception was, madg, but I use this reference for HIV/condom effectiveness. The study was for male-female contact and did not get into gay sex in any detail:
"Among participants who reported always using condoms, the summary estimate of HIV/AIDS incidence from the twelve studies was 0.9 seroconversion per 100 person years. Among those who reported never using condoms, the summary estimate of HIV/AIDS incidence from the seven studies was 6.7 seroconversions per 100 person years. Overall, Davis and Weller estimated that condoms provided an 85% reduction in HIV/AIDS transmission risk when infection rates were compared in always versus never users."
http://www.niaid.nih.gov/dmid/stds/condomreport.pdf
This is the report CDC uses when asked about condom effectiveness. Assuming the same protection rate among gays, a 15% risk (the flip side of 85% effective) is pretty risky when your life is on the line. The more partners or the more contact with an infected partner and the risk goes up.
In Russian Roulette, having only one cylinder loaded is 5/6, or 83% effective in not getting shot, or 16% dangerous (1/6), depending.
When I read the NIAID study it made me rethink risks!