fromderinside
Mazzie Daius
- Joined
- Oct 6, 2008
- Messages
- 15,945
- Basic Beliefs
- optimist
Ever try to debug a program without observing it malfunctioning, forming a hypothesis about why it's doing what it's doing, using the hypothesis to predict what will happen if you change the program or its input in a certain way, running it again to test your prediction, and discarding or modifying the hypothesis if the program doesn't match the prediction?I'm just saying that at core engineering isn't science in thought or practice.
Eyup. Designed and implemented language addition for DEC's FOCAL for my dissertation in ought '73. Obviously had to validate it before I ran people through it's routines. Also headed up an EA-6bB Software Support Team at Pt Mugu at PMTC's EA-6B SSSA. That's what US SSSAs (System Software Support Activities) do, have been doing since the late sixties. My last attempt to validate a system's software with for the EA 6B Prowler back in ought '84. A tiny patch of code in a IBM 360 generation box with a bit of real time control and sensory connections with a tiny interface for up to four operators. (two seats two rows of ECMOs and a pilot). Based approach on boundary and limit testing mostly along some methodology for with rooting out remaining 'efficient' go to type programming attempts. We actually had fairly good diagnostics back in the day.
Way beyond that today obviously.
Point is I was a scientist on an engineering team. It was my job to develop the hypotheses based on the evolving system's capabilities. I only wrote code as examples for engineers based on concepts I developed for evaluating such as limits, boundaries, and capacities, computer and human for making up shit. Referred to Krantz, Cosmides, and others a lot. Structure and statistics yano. My whole bag back in the day was coming up with neat ways to extract information efficiently from samples.