Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This doesn't change the crux of your story--which is that Senior DBAs who aren't familiar with ACID are probably not good hires--but that said, "consistency" is actually a somewhat ill-defined guarantee [1].

And if you throw in distributed databases, there is yet another understanding of "consistency" (vis-a-vis the CAP theorem), which really means "linearizability".

Sorry, didn't mean to be super-pedantic but consistency as a concept may seem simple to the uninitiated but is easy to get tripped up on, especially for folks who are aware there is a deeper layer of understanding associated with the concept, but can't articulate it right off the bat.

All this is to say filtering people truly is a hard thing. The extremes are easy to tell apart, but people who hover around the average are much less differentiated.

[1] https://en.wikipedia.org/wiki/Consistency_(database_systems)...



It filters out certain groups of people, but you will also lose out on good hires that don't focus on your given question. ACID has had almost no impact on my career. Only evaluating lies of ACID from certain db vendors is the only time I think I can remember it coming up meaningfully.


Yeah, I would have been okay with pretty much anything close, and would have given massive bonus points if they could give me a couple of relevant definitions. I wasn't looking for textbook, just for a reasonable discussion.

And I have similar anecdotes related to candidates who claimed years of distributed systems architecture experience, and were yet unfamiliar with CAP (and when explained, could not tell me even at a high level whether, in the event of a partition, their system was CP or AP, let alone the specifics of how that actually exhibited itself).




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: