Interestingly, this story actually helped me once. I was debugging an issue where some people in the multi-national company I work for (many offices) could connect to the server and some people could not, and it was thanks to this story that I seriously entertained the possibility that it had something to do with timing, due to the geographical distribution of the problem reports. Subsequent investigation revealed that for various uninteresting technical reasons, the server was setting a 200ms timeout on the SSL negotiation part of the connection, which closely matched the ping times of the participants, including the people who could sometimes connect and sometimes not. And I discovered this far faster by knowing I should be poking around on timing issues, rather than searching out something deeper.

(I am deliberately eliding other details not relevant to this aspect of the story. My point is merely that I am glad I had encountered it. Could easily have saved me a day or two.)