3 ms·

Rather than just believe in an unproven hypothesis why don't you test it out? Find the interesting-code projects you actually use or have heard about (maybe Ap

by smokinn 16y ago

Rather than just believe in an unproven hypothesis why don't you test it out? Find the interesting-code projects you actually use or have heard about (maybe Apache, Cassandra, Google's Bigtable, etc), find out who did the majority of the coding and then try to find their blog. How many people work on Apache? How many of them have blogs? Do you even care that they blog? Wouldn't you hire any of them anyway? Avinash Lakshman and Prashant Malik were the guys that wrote Cassandra when it was still an internal Facebook project. I can't find blogs for either of them. (Though they did write one blog post on the Facebook Engineering blog.) Doug Cutting wrote a lot of great code for Hadoop. He has a blog here: http://cutting.wordpress.com/ http://cutting.wordpress.com/ but hasn't touched it in nearly a year. Jeff Dean has done some amazing work at Google ( http://research.google.com/people/jeff/index.html http://research.google.com/people/jeff/index.html Now THAT is one hell of a resume) and I can't find a blog for him either. To me, it seems mostly random whether the person has a blog or not. It would be an interesting project to catalog and analyze how many do and don't. Then you'd have a real answer rather than a wild guess you believe religiously rather than scientifically.

barrkel 16y ago
You have it backwards. You are looking at the cream of open source developers, and then asking whether they have blogs. But the question isn't whether blogging is correlated with coding excellence (i.e. whether top tier of devs blog or not); it's whether coding ability is correlated with blogging (i.e. is blogging more associated with above median devs or not). There are over two million programmers in the US alone. My proposition would be true if more people in the top half of that distribution blogged about coding than in the bottom half, and I don't think that's an unlikely suggestion. Even if you think whether someone blogs or not is random, the subject of the blog is not. People's estimation of their ability is correlated with their ability (Dunning-Kruger effect notwithstanding, check the paper's graph - perception is compressed but not inverted), and I think it's conservative to say that people who think they know more about a subject are more likely to talk about it. To make it clear, I'm talking about blogging as a signal of ability, not ability as a signal of propensity to blog (which is what you suggest measuring). Since there are a lot more people coding than blogging about coding, it would be unlikely in any case to find a high proportion of bloggers at any level of the ability scale.

by smokinn 16y ago

barrkel 16y ago

You have it backwards. You are looking at the cream of open source developers, and then asking whether they have blogs. But the question isn't whether blogging is correlated with coding excellence (i.e. whether top tier of devs blog or not); it's whether coding ability is correlated with blogging (i.e. is blogging more associated with above median devs or not). There are over two million programmers in the US alone. My proposition would be true if more people in the top half of that distribution blogged about coding than in the bottom half, and I don't think that's an unlikely suggestion. Even if you think whether someone blogs or not is random, the subject of the blog is not. People's estimation of their ability is correlated with their ability (Dunning-Kruger effect notwithstanding, check the paper's graph - perception is compressed but not inverted), and I think it's conservative to say that people who think they know more about a subject are more likely to talk about it. To make it clear, I'm talking about blogging as a signal of ability, not ability as a signal of propensity to blog (which is what you suggest measuring). Since there are a lot more people coding than blogging about coding, it would be unlikely in any case to find a high proportion of bloggers at any level of the ability scale.