OK, now I know Caltrans has really lost it. Or never had it in the first place.
They’re now saying (as reported Monday by our buddy Jaxon) that they can’t find any proof that required testing was done on the questionable bolts. Nor can they find any proof that required inspections of the contractors making the bolts were made.
Caltrans Chief Deputy Director Richard Land is quoted as saying “The documentation problem is a concern.” I’ve got news for Mr. Land: It’s not a “concern”. It’s the core of the problem.
I’m not sure what world he’s in. In the real world, the way QA works is that testing is done and the result of the testing is documented. That documentation is used as the basis of a decision: proceed or don’t proceed. If the documentation doesn’t exist, then the testing wasn’t done. It doesn’t matter if the physical work of doing the test was performed or not, if the results weren’t documented, the work wasn’t completed.
The article references exactly one QA inspector who did her job: Mary Madere, who inspected bolts produced by a sub-contractor and reported that they passed her inspection (a review of the sub-contractor’s documentation and a visual inspection of the rods). And yet, the article implies that she was delinquent because she did a “random review” of the documentation. “Random” has a very specific meaning in the world of testing. It’s rarely possible to test every item in a set of physical objects or every input to a piece of software. So you work with a sub-set that is typical of the entire set. With physical objects, frequently that’s best done by random selection. Unless there was a requirement that all documentation had to be reviewed, Ms Madere is not delinquent here; she did her job.
On the other hand, the unnamed inspectors who were responsible for reviewing bolts produced by other sub-contractors, seem to be culpable. (I grant the possibility that all of their reports were misfiled; I’m speaking on the basis of the available information. Frankly, I find it extremely unlikely that it’s a filing problem. Based on the story, it seems like there is no record of the work having been done: no reports, no travel expenses, no complaints about late flights, no nothing.) Ditto for the resident engineer who was responsible for a final review of Ms Madere’s report. From a practical perspective, if there is no report, then the work wasn’t done. Somebody needs to start documenting who was responsible for testing what, who was responsible for reviewing what, and of those people, who didn’t do their jobs. Those are the people whose necks should be on the chopping block.
Unfortunately for Ms Madere, she has now been identified and associated with the Bolt Botch; if someone is scrambling to divert attention from their own failure (and you had better believe there’s a lot of that going on at Caltrans right now), that makes her an easy choice to be a scapegoat. Far easier than identifying and those who were truly responsible. Safer, too, as the chain of responsibility goes much higher in the organization chart than one lonely QA inspector.