As with others, "poorly", though I've made some attempts.
I keep my principle mobile device (an e-ink tablet) free of virtually all authenticated services. (An articles-clipping app is the one exception). I can read HN, but cannot engage from that device specifically.
I'll also try to turn off WiFi at least when possible.
HN as an article-filtering tool works well. Combining that with Algolia Search's time-bounded search, you can filter by the most popular stories of the past week, month, or year, which can be an interesting further restriction by entering a blank search:
Week: <https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=fal...>
Month: <https://hn.algolia.com/?dateRange=pastMonth&page=0&prefix=fa...>
Year: <https://hn.algolia.com/?dateRange=pastYear&page=0&prefix=fal...>
As always, "popular" isn't a synonym for "best", though it is a reasonable proxy.
You can also of course search for specific topics or people of interest, though here story search is often less useful than comment search, for which matches tend to be more likely and relevant. Note that comment votes were removed from publicly-exposed data ... a long time ago? Apparently 2011: <https://news.ycombinator.com/item?id=2595605>, though I thought it was more recent.
You can also visit "past" which will show previous days' posts, after the voting has largely settled. Again, this tends to filter out lower-quality content.
Otherwise:
- I'm trying to read or at least skim the article before reading comments, though that's one of those "observed in the breach" pledges...
- Recognise that comment quality on HN, whilst higher than most other significant online sites ... is still fairly poor relative to more curated sources, and there's a great deal of partisanship and ignorance (myself included) in discussion.
- "You don't have to attend every fight you're invited to", a/k/a "Someone is wrong on the Internet" <https://xkcd.com/386>.
- The HN Hivemind itself tends to glom on to specific topics and ideas more readily than others. I'm reminded of this every time I review my own submissions --- the stuff I'd really like to see take off rarely does, though there are occasional exceptions. It's usually either throwaway or predictable submissions which do get attention. (The submission queue is fickle, there is the Second-Chance Pool ("Pool": <https://news.ycombinator.com/pool>) to which you can both submit and peruse usually higher-quality / less immediately catnippy articles.)
- Time-boxing is underappreciated. Recognise what role HN does play in your life, and admit it at times but only in a limited fashion. You have 24 hours in a day, and probably about 2--4 hours per day which you can tightly focus on intellectual content. You're going to need to allocate that to work, personal matters, and general distraction. "Eat the frog first", put your most important tasks earlier in the day. (Much of this comes from David Allen's Getting Things Done.)
- I've also tried to apply a "BOTI" principle to what I encounter. That's "best of the interval", where "interval" is day, week, month, year, etc. I'll try to capture what seem both at the time and on later reflection to be the most significant articles, books, ideas, etc., I've encountered. This is partially reflected in the Algolia tips above, though I'll also keep a local list (bookmarks folder for "BOTI-Jan-2023", say, or a save-to-ePub EinkBro compilation on my e-book reader). There's the question then of whether or not that material actually was as compelling as thought at the time, and of course, the challenge of actually reading what I've archived. But it's at least an effort at organisation.
That said: the struggle is real and constant.