The strange case of invisible ink at the polling place
Submitted by Erik Petersen on Wed, 06/02/2008 - 12:11.
metaphors | people issues | perspectives
Models are typically the basis of all testing. We either model behavior based on specifications, etc and create scripts that are run once the software is developed, or we explore models interactively with exploratory testing. Models can create strange misunderstandings.
I worked in and ran polling places in Australia for many years. One of the challenges is getting polling staff familiar with their tasks and responsibilities before polling starts, so every voter can be processed correctly. This was evidently not the case at a polling place in Chicago, during the recent American primaries.
Different polling places use different mixes of technology. In Chicago, some apparently use touch screens with a stylus to "write" on the screen, and some use paper ballots that are actually written on. Polling staff at the 42nd precinct of the 49th ward appear to have been wrongly given paper ballots with styluses. When voters complained that there was no ink in the stylus, staff (evidently very security conscious) explained it must be invisible ink! The blank ballots were later rejected by ballot machine readers, after 20 papers had been issued.
The head of the Chicago Board of Elections, Neal Langdon, went on TV on election day, holding a stylus saying, "This is not a pen. It has no ink. It has no lead." He then went on to explain what happened. He also said that after an investigation "we are convinced that it was just utter stupidity". A quarter of the 20 affected voters had voted again, and others were being contacted.
So is this a case of the infamous id10t code (a.k.a blame the user), or is there a deeper cause? Evidently, the person who issued the ballots had absolute faith that they had been given pens that could mark the ballots. When they were told there was no visible ink, the only explanation that would fit their model was that the ink was invisible. This was clearly a mistake, but could be a training issue. In Australia, the supervisor in charge and their deputy set up the entire polling place before the staff arrive. If papers were issued with styluses, this would be their fault. The supervisor should always spend some time standing in the background watching that staff are following processes correctly. Again, this could be a training issue. In Australia, there are also roving supervisors responsible for groups of polling places. It is their responsibility to make sure that polling places are running correctly, and they also spend time watching in the background in polling places with less experienced staff or helping out. They report back to managers who oversee that all polling places work efficiently. The ultimate responsibility lies with the organizational head, in this case Mr Neal. There appears to have been a reactive process, sending out investigators after the fact, rather than more pro-active actions like better training and supervision. Many testers and developers will recognize the blame culture that Mr Neal seems to preside over. So while this was a case of human error that occurred at the bottom, in my opinion the responsibility flows upwards. This is all based on my model of Australian polling places of course, but in this case I don't think there are any strange misunderstandings! Read the original report here.
I worked in and ran polling places in Australia for many years. One of the challenges is getting polling staff familiar with their tasks and responsibilities before polling starts, so every voter can be processed correctly. This was evidently not the case at a polling place in Chicago, during the recent American primaries.
Different polling places use different mixes of technology. In Chicago, some apparently use touch screens with a stylus to "write" on the screen, and some use paper ballots that are actually written on. Polling staff at the 42nd precinct of the 49th ward appear to have been wrongly given paper ballots with styluses. When voters complained that there was no ink in the stylus, staff (evidently very security conscious) explained it must be invisible ink! The blank ballots were later rejected by ballot machine readers, after 20 papers had been issued.
The head of the Chicago Board of Elections, Neal Langdon, went on TV on election day, holding a stylus saying, "This is not a pen. It has no ink. It has no lead." He then went on to explain what happened. He also said that after an investigation "we are convinced that it was just utter stupidity". A quarter of the 20 affected voters had voted again, and others were being contacted.
So is this a case of the infamous id10t code (a.k.a blame the user), or is there a deeper cause? Evidently, the person who issued the ballots had absolute faith that they had been given pens that could mark the ballots. When they were told there was no visible ink, the only explanation that would fit their model was that the ink was invisible. This was clearly a mistake, but could be a training issue. In Australia, the supervisor in charge and their deputy set up the entire polling place before the staff arrive. If papers were issued with styluses, this would be their fault. The supervisor should always spend some time standing in the background watching that staff are following processes correctly. Again, this could be a training issue. In Australia, there are also roving supervisors responsible for groups of polling places. It is their responsibility to make sure that polling places are running correctly, and they also spend time watching in the background in polling places with less experienced staff or helping out. They report back to managers who oversee that all polling places work efficiently. The ultimate responsibility lies with the organizational head, in this case Mr Neal. There appears to have been a reactive process, sending out investigators after the fact, rather than more pro-active actions like better training and supervision. Many testers and developers will recognize the blame culture that Mr Neal seems to preside over. So while this was a case of human error that occurred at the bottom, in my opinion the responsibility flows upwards. This is all based on my model of Australian polling places of course, but in this case I don't think there are any strange misunderstandings! Read the original report here.