zlacker

Looking at this comment page it's clear that there is an absolute deluge of excellent ideas waiting to be implemented. The bottleneck here is pg. pg doesn't scale. As far as I can tell, he's the one who does the vast majority of work on Hacker News, and as the site grows -- and as YCombinator grows -- pg's (already huge) workload is only going to increase. This is, of course, similar to the "Linus doesn't scale" problem faced by the Linux kernel, to which the solution was git. [1] I expect a similar distributed solution is needed for Hacker News.

Re-writing the software in a language more people understand (e.g. Python) could be a good first step here. But I don't know if pg is willing to give up on his silver bullet (arc).

Turning Hacker News into a business might help. Create a situation where exceptional people can make lots money by figuring out how to make HN great and let market forces do the rest. Although figuring out how to make money off of content could be a pretty tough problem.

More generally, I think pg should be thinking less, "How can I improve Hacker News?" and more, "How can I create an environment where other people can improve Hacker News?"

I mean... investing in startups is a full-time job, running a high traffic website is a full-time job, building a programming language is a full-time job, raising a child is a full-time job... trying to do all four at once probably isn't going to work.

[1] http://lkml.indiana.edu/hypermail/linux/kernel/9809.3/0957.h...