Oddbean new post about | logout
 @ed709062 I think its biggest downside is that it's not designed for simplicity and low maintenance 
 @cedfe296 From a product standpoint, this might be unfair, but I have some skepticism (in terms of if I'd like the software, not its odds of success) due to how overconfident the developers are, such as in https://mastodon.social/@danluu/109374630690202023 (and I've seen other examples like this on FB, I just don't have a handy reference).

I think overconfident people often produce successful companies and products, but it's pretty rare that they produce a product that I personally like. 
 @cedfe296 I don't think I can state a rigorous reason for this link but, as a hand wave-y reason, if someone looks at Twitter and thinks the entire thing is so simple that it can be done by a smaller team than ForumMagnum because Twitter has less functionality than ForumMagnum, what are the odds that they'll correctly think through the relevant design and technical decisions for their own software? That feels to me like it should be a fairly negative signal. 
 @cedfe296 [sorry, mid-message pause due to IRL thing]

If I were King of Twitter, there are at least three rankers where I'd want a team of ten people minimum (there might be more — that's just what I know of without ever having looked into ranking at Twitter) and that's not including anything that would've been in the ad exchange that Twitter sold for $1B, which is surely at least 10x the hypothetical valuation of ForumMagnum if it IPO'd today and at least 100x the complexity, maybe much more. 
 @ed709062 I agree that the posts you're complaining about were silly and wrong.

Someone developing forum software for a few thousand users, though, can be totally wrong on what is like to develop for hundreds of millions of users and still be doing a good job on their own work.

My experience using the ForumMagnum deployments as an end user has generally been pretty good. 
 @cedfe296 Fair enough. If people find that it works for them, isn't flaky, etc., then that's more reliable information than the dev's opinions on some random topic that isn't directly related to their work.