Webpack Doesn't Work on Monday (On Windows)

Rollup has a lot of momentum behind it.

I think that's an exaggerated reaction. Bugs happen all the time, is this one any different because it involved such unusual circumstance? It wasn't a security bug, and how many users did actually freshly install Webpack on Windows-based computer on monday? Not a large subset. So it's more amusing than harmful.

Whenever things like these happen, of course you should think if the effort ripping off Webpack is better than suffering possible losses of productivity in the long run. But for majority existing projects I'd bet the projected cost of lost hours vs the cost of changing your bundler would result in net-loss.

And as an observation, I feel programmers often have a habit of lamenting loudly about other people's mistakes while not saying a word if they make an error of their own. Maybe they have a psychological need of maintaining a feeling of superiority about your own abilities, I don't know. Just something I've noticed during my years.

There's the classic 'Wednesday bug': http://web.media.mit.edu/~lieber/Lieberary/Softviz/CACM-Debu... [Story 'B']

(I'm sure I've read a longer version of this one somewhere else on the internet.)

But unironically.

Sorry, due to French union demands, Typescript (tsc) does not work on weekends anymore (in the EU).

Call it `time-pad`

less bugs?

Basically, it’s a good chance to:

1.) Question whether that codepath should be gated to a particular day of the week.

2.) Assess the cost of that codepath failing (or failing to load). Does it have the potential to brick the project?

3.) How big could the bug be? If software tells me it’s “just another manic Monday” on Tuesday, or that I should “Walk like an Egyptian” on February 31, who cares? There is no incentive to unit test because honestly, who cares? On the other hand, if it has the potential to brick the project, go back to #1. Do you really need to test the day of the week? Are you bloody sure?? :)

4.) If you reach #4, it’s time to invest in unit tests. You wrote a condition where a day of the week test could result in software failure.

I assume the motivation for typing in code is for it to run to completion successfully every time. In this case, because of the bug, Webpack didn't print a message reminding you to donate money to them. You would think you'd want to keep that working, because those donations are what make continued improvement of the software possible. So you write a test for it, and then you can hack on the other parts of the code knowing, with confidence, that your donation reminder works. Even on Windows. Even on Monday.

Ok, let me up the game. I had one simple short-lived cron command that failed on even minutes before h:20 where 5AM < h < 8AM on Sat, Sun, Mon.

Cause was another job that hammered IO because reasons, so mine was being killed by some unreported IO timeout after 1min, and retried the next minute (so every other minute). The other one was scheduled to run on off business time (weekends and Monday before people get to work) and had some burstiness that made the IO go down enough every other minute which allowed my short-execution one to proceed.

top was silent because IO, took us a long time to even thing about iotop because the failure was non obviously tied to IO.

Link?

Rough. Good luck! ;-)

I just re-read Ra, which is amazing. Found it much easier to follow than the (equally amazing) Fine Structure.

https://qntm.org/ra

Sounds like a DST issue, tbh.