A percentage of the software industry has a bias against programmers who don't code in their spare time because it's extremely difficult to evaluate technical talent. We lean on quality filters like number of hours spent on side projects that have sparse evidence to support their efficacy. While I'm skeptical that a general set of metrics exist we can use to perfectly evaluate technical skill across the industry, we can certainly do better.
Is it actually harder to evaluate a programmer's talent compared to, say, a lawyer or a doctor's talent?
An institution that's hiring fresh lawyers or fresh doctors can piggyback on the extensive, rigorous and difficult evaluation (which costs a lot of time and effort both for the candidate and the evaluator) during the bar qualification, medical licensing examination, etc. For example, if a potential doctor is a board certified proctologist(colorectal surgeon), then you can just assume that they will be able to do proctology to a high standard, you don't need to (and likely even can't) verify that aspect of the candidate. (Though employment of doctors itself is quite different from how hiring normal employees work, they're often more like independent contractors or running their own practices)
If an institution is hiring a fresh software developer, then a CS diploma does not even reliably guarantee that they know the basics of software engineering (e.g. see yesterday's discussion in HN https://news.ycombinator.com/item?id=24740765 ) - so you have to do all the evaluation yourself, and neither you nor the candidate would be willing to spend as much time and effort on that evaluation as a lawyer or doctor does, so it definitely is trickier to evaluate since you don't (can't) have enough information.
In these scenarios they are going to get lot of candidates who will do far more fuzzing on their CVs
So you find out you have a semi-treatable carcinoma, how do you pick the right doctor? AM radio ads? Billboards about proton therapy? Internet search? Hearsay? Whoever your insurance company picks?
At least I know how to evaluate a programmer for life and death work, like avionics. That's just verifiably coding to a specification. Challenging, but relatively easy compared to cancer.
The professions that were mentioned in the grandparent were all things that can easily be tested in an testing process. I.e. _play_ the guitar. Cook a meal.
Testing a doctor or lawyer well in a limited time is difficult, just like a software engineer. The difference is that they aren't expected to do "side projects" to prove their worth.
They don't, but not because they don't want to - it's because it's irresponsible for them to. The occupation of a lawyer or a doctor professes some privileges but also creates legal liabilities. As a lawyer or a doctor, you're not going to risk losing your license over an advice or procedure given to someone for free, on a hobby project.
(Also, the work of lawyers and doctors isn't creative in the sense software is, but it's primarily a people-oriented service. Which means you need other people to do your primary type of work for/on. The kind of work lawyers/doctors usually do after hours is called research, or just learning. Meanwhile, in software development, I can do the exact same type of work for myself that I do at my job, and end up with a digital product I can enjoy using.)
These days many software developers have never delivered software by themselves, so when they say they were “part of a team that developed X” we can’t know if they really did a significant part of X, or if they were just along for the ride.