It's only Monday and I'm already done with this week. I watched a work issue turn increasingly acromonious over either miscommunication, misunderstanding procedure, or being too wed to a tool to admit it's problematic. I drafted about half a sentence in chat, then stopped because I didn't think I was adding anything necessary and didn't want to make conflict any worse.

I'm increasingly convinced that Excel spreadsheet files should self-destruct or become permanently read-only after 90 days. That's plenty of time to migrate data into a proper database or to migrate calculations into a programming language amenable to auditing, verification, and change control. Excel should be treated as an attractive hazard with a time limit on how long an unverifiable and hostile-to-revision-control worksheet should be suffered to live.

Sadly, these two thoughts are related. Spreadsheets in their current form are too dangerous to be used for engineering work.

If you are questioned by a regulator about how your spreadsheet of engineering calculations/data complies with QA requirements on engineering software, I cannot help you. I'd love to be able to suggest a solution but there simply isn't one beyond migrating to a more appropriate, safer tool. And this isn't just about Excel; it applies to every calculational worksheet system (MathCAD, etc.) with similar characteristics of being effectively unauditable and unverifiable with no easy way to check versions or compare revisions. Spreadsheets are software.

I will never forgive Soace Karen for wrecking Twitter. That bastard broke a massive number of links I had to varied and sundry experts and professionals. One of the people I most need to reconnect with is Martin Nyx Brain who I believe was a researcher into provable(ish) computing. We had developed an intricate and very niche inside joke about a mythical startup we formed to make Excel verifiable complete with software products, consulting services. It went under the name of #JägerXL. It was a fine joke and about 40% serious which put it close to that gray zone of humor where nobody is quite sure if people are joking, including the participants.

People die because of bad spreadsheets. We've known about flaws in Excel's automatic type conversion for decades, silent limits on worksheet size, silent precision loss, the intractability of review and change detection/control. In every other programming environment cut-and-paste code is considered unmaintainable and bad implementation. Microsoft knows and they do nothing. Nobody can innovate in this space - it's business suicide to take on the 800-lb gorilla. With publish-or-perish being the law of academia, no progress can be made there - there's no prestige in tool development, it's not sufficiently novel, and novelty is what gets published and funded, not utility and need. Where does safer software come from? Nowhere. There's no way to create it outside a place like Microsoft and those organizations haven't shown an interest in it for at least the past three decades - why would anyone expect that to change now?

@arclight on the bright side, I've been seeing more and more students coming in the lab and using #R and documenting work in #RMarkdown from day 0, instead of using Excel. How did this happen? Well we taught them during their undergraduate years.
We're far from an ideal situation, but in a better place than we were 5 or 10 years ago.