3 ms·

why does it seem like most side projects are done by people for whom the predominance of their work appears to be side projects? It's easy to see why most side

why does it seem like most side projects are done by people for whom the predominance of their work appears to be side projects? It's easy to see why most side projects are done by people for whom the predominance of their work appears to be side projects: This is a trivial consequence of the fact that the vast majority of the software developed on this planet is "dark matter." It's developed inside companies, and even if you know I work for IBM on DB2, it's hard to see from the outside whether I'm a coding machine by day or whether I write emails from 8-5 and code side projects from 5-8. In the universe of software projects, we're mostly talking about side projects that are visible things like Github repositories, and mostly talking about day gigs that are invisible. If we want to drop the word "appears" from your first statement, we need to include side projects that aren't on Github (like startups being developed on nights and week-ends while the authors go to work "for the man" M-F 9-5). We then need to compare all of the side projects to all of the code these people write during their day gigs, whether it's highly visible stuff like Linux or "invisible" like the internals of an ATM locator for an online bank. If we don't include all of the side projects and we don't include all of the day gigs, it is very easy for our skewed sample to have many examples where the authors appear to do most of their work on their side projects. Please don't assume from this that I disagree with where you seem to be taking the conversation. If we see that things are not always as they seem, we also must accept the idea that even if people with side projects appear to be good programmers, it might be that the high-profile examples we find on HN or who blog are not representative of programmers as a whole. It is very dangerous to draw conclusions about the correlation between a side project and talent based on an incomplete sample. I think we agree on this here and in another thread elsewhere. As they say... The plural of anecdote is not "data."

kenjackson 15y ago
If we don't include all of the side projects and we don't include all of the day gigs, it is very easy for our skewed sample to have many examples where the authors appear to do most of their work on their side projects. Absolutely -- I think that's where I was trying to go, but not as eloquently. If you look at the thread, I define "side project" in, what I consider, an odd way. Basically something that is on GitHub, because that has become what so many mean by side project. In a roundabout way I was trying to say that people learn and grow in ways that are less visible. And while others may question your programming skills as a result -- you shouldn't.

kenjackson 15y ago

If we don't include all of the side projects and we don't include all of the day gigs, it is very easy for our skewed sample to have many examples where the authors appear to do most of their work on their side projects. Absolutely -- I think that's where I was trying to go, but not as eloquently. If you look at the thread, I define "side project" in, what I consider, an odd way. Basically something that is on GitHub, because that has become what so many mean by side project. In a roundabout way I was trying to say that people learn and grow in ways that are less visible. And while others may question your programming skills as a result -- you shouldn't.