A large amount of Show HNs are not startups; they are personal projects.<p>Relatedly, using "the startup is not dead" as a metric for <i>the startup alive</i> is a bad idea. People do not shut down everything when a Hacker News submission does not get upvotes (<i>especially</i> in the case of Show HN, where many are hosted on GitHub pages and are free to host, although you account for that in the regression).<p>The regression has a few issues:<p>> It's a classification problem (alive or dead) so OLS doesn't make sense.<p>> Score and Comments are multicolinear and cannot both be in the same model.<p>> You don't answer or give statistics toward "how well" the model predicts.<p>> Related to the comment earlier, you don't comment on the magnitude of the on_GitHub coefficients, which are <i>huge</i> and skew the entire result of the regression!<p>While I always appreciate analyses of HN data, the conclusions raise more questions than answers.
The dataset interpreted my site as dead because it 301 redirected to the SSL connection. This is probably quite common, so take the living/dead stats with a grain of salt.
This a really great overview.<p>Not complaining, just a note. Those counts are from different points in time of Hacker News. Which means that a chance to get point grows with userbase. Some extreamly popular thing in 2011 can't compete with something in 2016.<p>Probably it's hard to guess how many user HN got in those time points. But maybe adding section by year could help a bit.<p>Anyway, Anton great job.
Heh, I have 3 projects on that list and I'm still running two of the three. One was a flappy bird clone in Swift; the other is a Touch Visualizer for iOS and the third is <a href="https://www.gitignore.io" rel="nofollow">https://www.gitignore.io</a>. Pretty interesting analytics; it's funny because I'm in the process of migrating the touch visualizer to a Swift project and a new project owner.