It's funny isn't it. Recently in a job with "unlimited" vacation, because of a dubious message from one of my two bosses who was a bit of a dick, I was too scared to take a real vacation. Until Christmas. Then I decided I was going to take some. It had been a rough year, isolating from Covid, not enough money, and living in shitty circumstances. It was the first PTO I'd had in over a decade, as working as a freelancer/consultant often means no PTO, so I decided to savor it, come what may.
I took just under 3 weeks, like almost everyone else: there was a shared vacation calendar where I could see everyone else's Christmas break.
My reward when I got back? Low performance metrics "in December" were cited when laying me off. It wasn't just about December, but December was the month they decided to measure and "give me a chance". They didn't take into account the break, and the only way their "assessment" could be satisfied would have been to work through Christmas.
I then worked my ass off to ship a technically difficult, world-record-beating feature during my notice month, which they told me if I delivered it would surely be impressive, and turn it around. I did ship it, but not until the very end of the notice period, which was too late. If they had cared, they would have seen it was on track.
If they had kept me on, let me relax, and worked with me rather than their choice of how to assess work, they would now have a world-beating product. It's their choice of course, and I now don't think they were serious about trying to build a real product. I think it's a bit of a smoke-and-mirrors scheme to keep grant money flowing in. After all, in about 4 years nobody has ever run the product on the real data it is designed for, except me, and I had to pay for servers from my own pocket to run those tests. Even now, I believe I'm the only person ever to run it, or even be able to run it.
It's been interesting to watch how the product has stayed in the doldrums since I left, and how the folks working on it are now starting to implement things for which I have had working, high-performance functionality for months in my private fork since leaving. (It's open source.) It will be particularly interesting to see if their version is ever able to run on the real world data it was created for, or if their perpetual optimism will be forever misplaced.
Ironically, I'd say the company had the nicest, most helpful HR, legal and accounting teams I've ever seen at any company. There was a lot to like, and I'm sad to have had to leave. But I don't miss feeling constantly afraid there.
And, as a person who really enjoys creating things, I don't miss watching another team member shipping garbage commits that usually didn't work, and doing fine, while I was the only person on the project providing real functionality but not scoring well on the right metrics, because I spent too much time solving the product's blocker problems. To score well I'd have to ship garbage too. Oh well.