the problem is that they built their "intuition" without ever looking at the code directly. They are always using second and information and those are not alway

the problem is that they built their "intuition" without ever looking at the code directly. They are always using second and information and those are not always good. My concrete design handbook in college had a corrections on at least 75% of the pages and it was the 3rd edition. We often look online and in popular textbooks for design examples to test our software, and we find a lot of errors in them too. >which might be more accurate than the horrible mess of a law that you describe I've had a senior engineer at one of our biggest client accuse us of having a bug in our software. He has dozens of people working for him and teaches other engineer how to design bridges. Turns out he used the arctan function in excel instead of cosine. Her mistake was roughly doubling the shear resistance of the main beams on the bridge. We ere the one to find the mistake in his calculation. clients often call us with question like "how can I use your software to design XYZ". When we talk with them on the phone, they often don't have a clue how to do it manually and they are hopping that the software will just do it for them. It's also very common for us for find critical mistakes in client's finite element models when they contact us for something completely different.