ArclightJan 26, 2023

I was compelled to explain in a large training session that Excel and MathCAD are not, Not, NOT acceptable tools for performing QL-1 nuclear safety calculations.

I absolutely hate that I have to keep explaining this to degreed engineers doing safety-critical work.

Show threadArclightJan 26, 2023
I am going to die on this hill whether I want to or not.
Show threadArclightJan 26, 2023

This just makes me want to cry.

Excel is suitable only for tasks less important than Girl Scout cookie sales tracking. Do not bet anything more valuable than a box of Thin Mints on Excel.

Show threadmhoyeJan 28, 2023

@arclight
Excel is not a spreadsheet. Excel is a full-featured virtual machine running a smalltalk-inspired REPL whose display layer happens to resemble a spreadsheet.

Something like a third of the world’s money goes through Excel every single day, and the reason you don’t think Excel is a Real Programming Language is because if we admitted that, we’d have to admit that most of the most important software in the world was written by underpaid women in pink collar jobs, and we can’t have that.

Show threadArclightJan 28, 2023
@mhoye I think we're on the same page here. Excel is the world's most popular functional programming system. The problem is that it's sold and viewed as an easy-to-use tool for "non-programmers". The problem is that it is in fact a programming system, spreadsheets are software, and should be treated with the same rigorous review process as software written in traditional languages. Hiding the programming aspect of the system has a secondary effect of making them virtually impossible to verify.
Show threadGraydon

@arclight @mhoye This.

There was, awhile ago now, a report out of Wall Street to the effect that any non-trivial spreadsheet is buggy in some important way, and while they had had every motivation and a lot of money to throw at the problem they just could not fix it.

Code is only as good as your tests, and it's nigh-impossible to test your Excel code.

Jan 30, 2023 at 12:57amWeb
Show threadArclightJan 30, 2023
@graydon That is the context of this thread. If you take your work seriously, you will use tools that support good, basic software development practices (e.g. testability, auditability, configuration mgmt). The initial post in the thread noted that the problem domain was engineering, specifically nuclear safety analysis. The argument applies to any work considered by its authors or their organizations to be important.