Stanford’s war against its own students

There are quite a few sites whose content is mostly skewed but which occasionally publish good articles that, for whatever reason, wouldn't get published by less tendentious sites. Our approach is to downweight such sites but not to ban them, and to turn the penalty off when an article seems good for HN—that is, intellectually interesting and not too repetitive of what has been discussed recently. Needless to say (I hope?), we do this irrespective of ideological orientation, so you'll find articles appearing here from all kinds of sources.

The longer we've had this policy, the better I think it's held up. For a community organized around intellectual curiosity, it's super important not to miss interesting outlier articles. Finding something interesting doesn't imply that one agrees with it, and turning off the penalty on a particular submission doesn't imply that one endorses everything else on a site.

Guilt by association has pros and cons—it can be a useful shortcut, but fails badly when it does fail, and when applied repeatedly it has a slippery-slope downside that can get pretty bad. Still, I don't think your argument is wrong; I just think it's a tradeoff and the question is which side of the tradeoff one should take.

In HN's case, many tradeoffs are resolvable by remembering that there's only one thing we're trying to optimize for*: intellectual curiosity. I think this principle (of evaluating articles rather than sites) follows pretty easily from that.

* https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...