> ...platform directly under the bridge so there shouldn't have been any sideways force.
No net sideways force. Any +X force on the left wall will have a -X force on the right wall. If there's a tension T in the ropes across the top they'll be pulling the walls inward with a force of T causing a torque of T*height.
> If there's a tension T in the ropes across the top they'll be pulling the walls inward with a force of T causing a torque of T*height.
Almost, you'd still have to subtract the friction component across the 90 degree angle where the cable hangs down from the edge near the porthole but that's probably relatively small compared to the total (depends on cable material and any kind of padding they put in between the cable and the concrete).
It would have been a lot better to sling those cables through the portholes on both sides and to make it loop the bridge, one person would have to rappel down to get that started but you'd have zero load on the walls.
The longer I think about this the more I think those guys and girls were extremely lucky and I hope that this will not inspire any copycats to try this with other bridges or even the same one. It's a long way down.
I prefer to live in a frictionless world :) It looks like it was rope on metal (http://jaccuzzi.ch/photos/700/1722_264_03.jpg) with a coef. of friction of 0.2. My gut says the friction force will be at most half horizontal. That puts friction an order of magnitude lower than T, though that is more than I expected. (edit: damn, it rounds the corner from vertical to past horizontal so my symmetry argument doesn't quite hold.)
Why they went over the top of the walls I don't know, I think they wanted to measure the ropes exactly instead of adjusting it the day of. Like, that was where they drew the line of "unacceptable risk"? I agree a loop through the drain holes would have made their lives easier in a bunch of ways, except the tripping hazard would have made it a very risky project...
No net sideways force. Any +X force on the left wall will have a -X force on the right wall. If there's a tension T in the ropes across the top they'll be pulling the walls inward with a force of T causing a torque of T*height.