The field changes, it's changed significantly since I started 10 years ago. Yet there's a fairly large contingent of devs that aren't keeping up. I've interviewed old devs that are at the top of their game and old devs that are no better than the interns we have. I've worked with the same.
Another thing I've seen is people complaining about "old geezer not keeping up" is used to shut down conversation about whether $new_fad is really the best choice, and $old_thing may actually be better. Not saying that was your intent, but I've seen it used like this many times.
Not all that many years ago someone saying "no thanks, I'll stick with SQL instead of no-SQL" and even "no thanks, I'll stick with Java because I think the static typing is better" could be seen as "not keeping up".
This is why I tried not to classify this as a young v old thing. I don't think it's the case that old devs can't keep up. I think it's the case that some devs don't keep up regardless of age.
For young devs, it's hard to know if that's the case or if it's simply inexperience. For old devs, it's far more glaring when all their experience is with VB6. I think that's where a lot of the ageism in software hiring practices comes from.
> Not all that many years ago someone saying "no thanks, I'll stick with SQL instead of no-SQL" and even "no thanks, I'll stick with Java because I think the static typing is better" could be seen as "not keeping up".
Perhaps some saw it that way. I never did. I would say that if you were completely unfamiliar with no-SQL, you likely weren't keeping up. Just like if you aren't aware of k8s or docker today, you aren't keeping up. Whether or not you chose to use those techs is entirely different from simply outright rejecting them without evaluation.
They lack foundational software engineering knowledge/opinions
Use their IDEs as if they saw it for the first time
Lack of experience when it comes to architecture
I've seen people organize their data in a cache-pessimal way in order to squeeze out a couple of CPU cycles, for example. It becomes really tiring when you constantly have to whip up a benchmark just to settle debates on things that have been common knowledge for basically forever now.
You can argue everything with "because of performance reasons" because it requires way more effort to actually check it.
I always become suspicious when a software engineer has been working on a problem +10 years. The type that spend their time upgrading angularJS to angular or something like that.
[1]: https://news.ycombinator.com/item?id=4627373
[2]: https://ask.metafilter.com/62762/What-is-the-source-of-this-...