The problem with encouraging developers to speak up when they're stuck is that the developer that is stuck isn't really ever sure if it is something they really should know about the library, framework, or language or if it really is something hard. In some sense senior developers have a real leg up here because they know that they're capable of building systems for large companies.
I've seen many junior developers fired because "they require too much hand holding" and I don't think the advice should be "speak up" but rather "make at least one friend as soon as you possibly can that you can privately ask questions to so you don't look like an idiot" and I know this advice is better for the individual than the company, but it's the reality. The selection function for new, junior employees is "can they get the things done with minimal hand holding" not "do they ask questions" and I'm not ok with that, but it is the reality and we shouldn't push juniors into situations that would get them fired before they're up to speed.
Whatever the work is, the important thing is that it gets done, usually as efficiently and as good as possible. If it takes me explaining the same thing 3 times in a row to get something done properly the first time I'd rather do that than have to go back and fix things later.
And honestly, everybody starts somewhere, the best way to learn is by asking questions. If someone is asking me questions, it tells me they recognize they don't know everything, care enough to learn and are humble enough to ask for help.
Such qualities are rare in people and i've found every time, the people like that make the best employees and coworkers.
Until you get the person who questions everything, draining all your time. Then they become reliant on you, and stop thinking for themselves.
For coaching the questioner, I ask them how they approach finding information, and recommend they time box their exploration, and only ask their question if they've given it a good try. I I also will describe my personal workflow to find answers when answering their questions.
For coaching the rest of the team, I often suggest they ask "what have you tried so far?" or "where have you looked so far?" before just giving an answer. This can help show knowledge gaps, and oftentimes the questioner will take another look and discover the answer themselves.
And yes, if they never get away from this behavior even after getting feedback multiple times, then that turns into a different sort of conversation.
But I do think there's a difference. If a person generally asks a lot of questions, but there's clear signs of improvement and they're trying to pull their weight, then I really don't see it as a bad thing.
That said, if a junior brings up a question like "why does this error for some tests, but not others?"
records = some_orm_thing()
some_model_ids, other_ids = map(list, zip(*records))
I agree there's a difference between the kind of question like the one above vs. something like 'Why isn't there a check in there before unpacking a record?'
The first question is basically 'Teach me stuff I should know' the second is more along the lines of 'Why do we do it this way?'
And there does come a limit, if after a month they're still asking the same questions, not really seeming to have learned anything, it's probably not going to change.
I dunno, really I guess it usually comes down to the person and I hate to say it but, how annoying I find the questions and the person's behaviour.
I think my larger point was that smart juniors should seek to get clarification privately after exhausting their own skill, but you're absolutely right that they should raise issues as they come up in a thoughtful way to accelerate the team and their own growth.
Of course there is always a level of knowledge too low to participate constructively, but that’s another scope (fire your HR along with yet another zero-skills guy, etc).
If the answer is [in context] indeed "yes", then that could help tune their threshold for when learn-by-self vs learn-by-others will be appropriate [again, in that context].
a) do any research before asking the question,
b) trying to understand the answer given or
c) doesn't follow the answer given.
I'm not actually encouraging that. I'm encouraging them to not have imposter syndrome when they actually manage to fix something, regardless of their methods. The ones who ask stupid questions, in my experience, don't end up fixing the issue ever because they don't understand what they're doing enough to ask the right questions, so they wouldn't be at this point.
I've never even heard of a junior fired for "asking too many questions".
It's also important to know if it is a bug or incomplete implementation of a feature. Sometimes the code is over engineered or there are no docs to help understand its intent.
If junior devs get fired for asking questions, or because that makes them look like idiots, I'd take a hard look at whether they have the support needed to set them up for success. Maybe the solution is actually "don't hire Jr devs" because the organization isn't able to support them.
Thank you for saying this. I haven’t heard it being said anywhere else, but knowing people that you can DM and ask about things first is “socially” a better strategy than an ask tons of questions in a public thread. What’s also worked for me is to have a private channel meant only for developers (no manager, PM etc. Just devs, free to ask any question without judgement).
People instinctively assume the junior person asking questions is less competent.
I still remember the time trying to reproduce and fix a certain issue. We had a 3-node cluster mysql database setup with one main and two replicas. The replicas are read-only. If one replica goes down, reads shouldn't be affected at all. We had an application layer on top of this to coordinate and enforce these requirements. Every which way you look at the code, there's no reason why this shouldn't work. And yet, when one node went down in production under a very particular circumstance with the network, it took down the entire cluster - not that the cluster was not reachable - it wouldn't allow reads and just hang.
Tracing it took me down the hole of understanding the precise network issue (fading memory - I think the machine wasn't responding back and tcp kept retrying) + linux tcp setting (I think tcp_delack_min) + a java networking setting + how thread safety was setup in the application (do not use synchronized at the method level). It took a week to diagnose.
Recently, there was a minor issue with adapting a sample Spark UDF to work with our setup and it took the developers 3 weeks and they kept giving up after every try.
To me solving a problem like the one you describe here is one of the biggest joys of working with IT.
Though it seems to come at the expense of being able to design large systems. Not sure if there is a way to be good at both.
We didn’t just trick a rock into thinking - we beat on it until it was really thin, shoved a little bit of lightning inside of it, then tricked it into thinking.