Well, I can tell you my answer to that, but it's just my answer, and I'm biased and don't really know. Others' answers may be just as good or better. Every HN user has their mental model of the site and these models disagree wildly. Users tend to be passionate about their model, which is more important.

My view is that HN has managed to stave that fate off so far. If so, that's not by accident—it was one of the founding intentions of the site to stave off Eternal September, which was just as well-known an issue amongst internet forum types in 2006 as it is today. You can see this in the second paragraph of https://news.ycombinator.com/newswelcome.html. (Also https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...)

If that's true, how have we been able to escape entropy? The same way anyone does: by spending a lot of energy. In particular: a lot of worry and hard work, patient engagement with particular users about specific issues, stubborn focus on optimizing for one thing (https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...), the ability to modify the software to adapt to problems, and the fact that growth has been linear over 10 years and that we're not under pressure to make it faster, which is a historical accident of HN being part of YC—sort of a felicitous one-off, like Iceland having a mild climate.

I do think there are trends in an entropic, Septemberesque direction, but there are also counter-trends in more organized directions, with the net result being that the site has changed but not yet become uninteresting. Other things I think are that there's a cap on how good HN can ever get (boo), but that there are gradual things we can do to make it better (yay). They just need to be slow.