(another rant about work)
I have a place to stand, but apparently I am am not allowed to have a lever. We have a deadline. The deadline is god.
The project team I am on has been working without the proper tools for several years now. Oh, they keep fancy and powerful PCs on our desks. We have the latest and greatest communication aids and the appropriate software development software. None of that does any good if we can not test the software to ensure that it works (yet, we must produce useless paperwork to insure the bureaucrats against its potential failure.)
We can not adequately test the software because they refuse to give us a development environment that has robust data. Without data we can not prove the software works, and we put test sites at risk because of it. This is not rocket science. This is common sense.
Let me explain. Say you are asked to develop a math formula that does a whizbang calculation. You work and come up with “a+b+c=47“. You think it is right. You can invent numbers for a, b and c that will make the equation correct. But is it? What are realistic, real world examples of the values for a, b and c? What if in the real world a and b are always greater than 24 and c is never negative? Well, the formula is bad, and you need to develop a new one. Big deal right? You send the software to a test site, they find the error, you fix it; no problems, eh?
Now, imagine that a is a blood pressure value, b is a potassium level, c is a white cell count and 47 represents a patient’s life. The software I write is meant to assist caregivers in helping treat patients. As such, if the caregiver is working from a bad formula, real harm can come to the patient. Let me say that again a little more clearly. Bad software can harm patients.
However, we are not permitted to spend the time to understand and then create realistic data to test our software with. You see, the deadline is god. There is a process to allow software to be released “with known defects” just so that the deadline can be met. And this causes a big enough concern that the bureaucrats have developed a policy to deal with software that contain “patient safety issues”. Who cares about the patient as long as there is a strategy to deal with blame?
What they haven’t done is to develop a strategy to deal with the fact that the vast majority of our software developers do not have realistic data available to them to properly test the software before it is installed at a test site. We have lots of paperwork to protect the pointy-haired bosses from blame, but none to protect the PATIENTS!
I get to bring this topic up again during a meeting today. Is it a sign of a broken spirit, or a firm grasp of reality, that makes me so sure that I will be handed another answer similar to “we are looking into the problem and if it is indeed a problem we will request funding for a study on how best to deal with the problem”?
Some day’s I hate my job.
I have a place to stand. All I need is a lever that can move the bureaucracy into action. I am afraid that Archimedes would have had an easier time moving the world.
Filed under: Devenbert, Tech/Science | Leave a comment »