Ethics
Latest Column -- Testing training: Disturbing behaviors of students
Submitted by sbarber on Fri, 07/11/2008 - 20:17. ethics | events | general software testing | industry recognition | other online resources | patterns | people issues | perspectives | test managementMy latest column...
Drive-by training. Never heard of it? It is exactly what it sounds like. You drive to a training facility (or an instructor drives to you), for a day or three the instructor delivers the pre-packaged training class, then everyone drives back home. It's not the best training model ever invented. There is generally no student assessment, and the only instructor/course provider accountability is reputation. Even so, many good ideas can be shared and lots of students come away feeling that it was well worth "the drive."
As it turns out, I've been delivering a lot of drive-by training to software testers this fall. That in itself isn't particularly noteworthy -- end-of-the-budget year is a popular time for drive-by training -- but something that is noteworthy is that I have noticed a rise in some disturbing behaviors among the individuals and organizations that select and attend drive-by training.
At first, I thought it was just me. But after an informal poll (and some lively discussions) with my employees and trainer friends in the testing realm, I became increasingly convinced that the behaviors I'm noticing are not exclusive to me and that I'm not the only one who thinks they are on the rise.
Read the rest of the column.
- --
- Scott Barber
- President & Chief Technologist, PerfTestPlus, Inc.
- Executive Director, Association for Software Testing
- Co-Author, Performance Testing Guidance for Web Applications
- "If you can see it in your mind...
- you will find it in your life."
When information degrades to data
Submitted by Erik Petersen on Tue, 28/10/2008 - 02:40. books | ethics | people issues | perspectivesLawrence Winkler has two libraries. One is a traditional one, a library of information with new books and some more than 100 years old. The other library is now just a library of data, which was 3 books he had paid $100 for. What happened? Technological upgrades have degraded his books into unintelligible digital goobledegook. Now he is trying to highlight the dangers of storing information only in a digital form. In May, he had the 69th comment on a post about ebooks and copyright. In October, he had the 10th comment on a post about Amazon pushing their e-book device.
German Bank burns $500 million in automated transfer, and no one blames the computer?
Submitted by Erik Petersen on Tue, 21/10/2008 - 10:41. ethical hacking | ethics | general software testing | industry recognition | people issuesThe German media branded them “Germany’s dumbest bankers” but two Board members and a senior risk controller of the state owned KfW Bank were suspended then sacked two weeks later. They were blamed for a transfer of 350 million euros to the US Lehman Brothers, 2 hours before Lehman’s declared bankruptcy despite saturation media reporting of Lehman’s troubles.
Homeland Security’s “unwelcome visitors": tales of many bug types
Submitted by Erik Petersen on Sat, 16/02/2008 - 04:46. acceptance testing | ethics | general software testing | metaphors | perspectivesThe American Homeland Security megadepartment has been in the news recently, covering a whole taxonomy of bug types, relating to people seen as unwelcome visitors and unwelcome visitors seen as raindrops! Firstly, we have the sad story of honest Debbie Williams Arthur who tried to renew her Virginia driver’s license over the net, only to have it rejected with a message saying her car stickers had to be picked up in person. Her husband went to pick them up, only to be told it had to be Debbie, and she had to call a toll free number in the meantime. When she rang the number she was told that Homeland Security had identified her as a fugitive called Debbie K Williams from another part of the U.S, and to bring her ID in to the office. When she got there, they said the problem would be fixed in 15 minutes, then she was told after an hour there were complications. These included the fact that the system didn’t think she existed and the Homeland Security override would not allow her records to be set up. After 3 ½ hours she left, after being advised to walk home because any police check would result in her arrest as a fugitive! She ended up with a hand written note (!) for the police in case she was stopped. Despite being told the issue would be fixed and she’d be given phone updates, none came. Finally after another visit to the office, she got her car stickers (at which point she probably was a fugitive driving with expired car stickers!). This is a production bug, relating to data. The system as a whole works, but missing data and locks on the existing data from the external homeland security system presumably needed software fixes to rectify the bug.
ET thoughts: Prediction and the all players all-in poker bug
Submitted by Erik Petersen on Mon, 04/02/2008 - 12:05. ethics | exploratory testing | perspectivesPushing boundaries at airports
Submitted by Erik Petersen on Fri, 18/01/2008 - 13:19. ethics | heuristics | perspectivesNot all bugs are created equally. One of the most common bugs is all about equality, too much or not enough. It is caused by code being “off by one”, where a boundary is meant to be at a particular point but is one value too high or too low, for example >=5 (5 or greater) instead of >5 (greater than 5).
Two incidents at airports had me inadvertently testing for boundary bugs. I had dropped off someone at the airport, waited for their delayed boarding and returned to the car park. Parking was $10 for up to an hour, then $18 dollars for up to 2 hours. I paid my ticket at a machine at 61 minutes, and paid $18! Damn. That boundary was spot on.
That wasn’t as painful as another time, getting into the checkin queue a few minutes before the half hour baggage cutoff. I waited a minute or so, and the attendant started to check me in then apologized that the system was preventing her completing it because the half hour boundary had just been crossed. I ended up bumped to the next flight leaving in another 2 hours. Even worse, she explained that if I had checked in at a kiosk instead of queuing, she could have checked my bags in and got me on that flight. At another time I might have been interested in my inclusive interpretations of their exclusive boundary (queue before cutoff versus be at counter before cutoff vs. be checked in manually by cutoff vs. kiosk checkin). That was beyond the bounds of common decency, but the program overruled any staff sympathy, though there was empathy. I just wandered off into a fog of unbounded apathy, which had no equal. I found no bugs but both situations certainly bugged me! [grin]
Read a great article about boundary testing here
Good Advice
Submitted by Alexander Podelko on Sun, 06/01/2008 - 05:42. ethics | non-functional testing | people issues | performance testingPrivacy, PR and the unpardonable "problematic" breach
Submitted by Erik Petersen on Sat, 29/12/2007 - 09:28. ethics | general software testing | non-functional testing | people issues | perspectivesWhile these are offences under the law, the fact is that many of them do not directly impact the people involved. A new privacy breach that is unforgivable has been occuring in Ohio in the U.S, Some adopted children visiting hospital are being called out with their birth names, and having their details displayed with their birth parents records, not their adopted parents. Some of these children may not even realize they are adopted. And of course, there is some PR as well, a government official said, “Obviously, when information like this is being inadvertently spit out by a computer, it is problematic”. I guess if there was an award for the understatement of the year, that may qualify!
