If you've ever built a UI in Rails, you've probably noticed that views tend to get slower over time. Fortunately, Rails provides us with an easy-to-apply band-aid in the form of view caching.
I expected the performance characteristics moving to heroku “standard” dynos would be about the same as they are on our current infrastructure. But was surprised to see some degradation.
I was noticing half a dozen exceptions per page being silently raised and rescued on an app. Raising exceptions is pretty slow, much slower than just asking if the object responds or not.
Ignoring any benefit to the wider Ruby community, Sorbet has been a resounding success internally. In the beginning, it was just type enthusiasts who were adding type signatures to files. Now, having recognized the value they provide, it’s everyone.
There is a moment in time in incident analysis where you need to pivot from identifying timeframes in which the incident happened and machines that are affected to actual instances of the incident in question.
The fact this question continues to come up time and again after all these years prompted me to wonder why the matter hasn’t been settled by now. Thousands of people have tried their hand at pairing in a wide range of circumstances.