Replies

As a systems person who runs and tests systems and who has been a developer as well, I really think conceptually this was flawed, as well as the testing and implementation.

First, higher tech does not always mean better. To me, a smart phone is a really poor input device for most people.

I had a similar issue years ago, my employer had a light pen for media buyers to select dayparts on a screen, and as the system was on just one PC, it was a bottleneck and the light pen was a pain. I reprogrammed the process to produce paper reports that could be passed out, filled out and returned, and I entered in the results, far faster than they could, and they spent more of their time doing their real jobs.

A better idea would be to have scannable paper lists where a bubble is marked when a voter shows up - and have enough copies for them to be picked up hourly. Runners could pick them up, and you would have distributed points where someone is in a car with a laptop, a scanner and wifi. Easy to fill out, and if one car's laptop or scanner fails, run to the nearest one after that. Security is also improved this way - the campaign would control the access points. In the end, IMO this would have produced more speed of remote info than ORCA, and far better reliability.

Second, there were so many basic systems mistakes here it makes me wonder if the people running this had any experience actually managing systems. It was insane to think that they could give people access the day of the election and everyone would automatically master the interface within ten minutes.

A better plan would be to have a beta site and have a Saturday where everyone could log in and do a dry run. First, you would be sure their COULD access it. Second, it would make them familiar with the interface before the election. And third, it would provide close to real-world load testing.

Third, this was top-down. I would imagine you had some know-it-all MBA hotshot running this who had conference calls with suck-ups, none of whom had the balls to ask what the **** is going on with this crazy implementation plan. Someone who has no understanding of the **** happens factor, which is always there. I once had an MBA berate me about preaching that **** happens, and asked me to give him an example in his project plan of such happening. I told him that if I could do that, it wouldn't be **** happens and it would be in the project plan. Sure enough, it came in late and over budget because of unexpected problems. Seasoned pros with some humility use planning, testing, time and budget cushions, more testing, user acceptance, more testing and them rollouts rife with contingencies to deal with the inevitable problems that come up.

The real problem with ORCA was that they didn't even spot the obvious problems in their plan that anyone with two years of experience with running systems could tell them were coming at them like a freight train. But, then again, this is coming from the same bunch who thinks RINOs make the best presidential candidates, despite decades of evidence to the contrary.