[stevenicr]

9 out 10 times never read the article - comments often shine the light that there is a better X that is open source, or that the thing I wanted when reading the title is Mac or IOS only - save me the click and the time.

I do try to read the article / thing if I am going to comment - but that's not many.

Many times the comments let me know that other needed thing is not released yet - or that no pricing info is published - love those saved clicks.

[stackbutterflow]

9 out 10 times never read the article

I also do that. Most non-news articles aren't better than the average two paragraphs HN comment, but they are 10 times as long. Time is precious and at some point who has the time for yet another "[strong opinion about some technology/pattern/that's how we do at our company]". Almost 100% of the time the comments contain better information on the topic tackled by the article and with higher information density.

If it's a news article there's always a few comments quoting important parts of the article which is usually enough to understand what's being reported.

The articles that I tend to prefer and to read are the ones that highlight a feature that I wasn't aware of about a language that I'm interested in.

[xnorswap]

I do this, but it's a dangerous trap to assume that the commenters are more reliable or a better authority than the articles themselves.

The articles have a filter that someone has gone to the "effort" of writing up an article, and it takes a lot of upvotes to get to the front page. The commentary has no such filter.

It's very easy to shoot down ideas on HN in the comments and get rewarded for it, especially if it's to dismiss a product with alternatives. Not just the now notorious dropbox comment, but I'm often seeing comments here suggesting alternatives that if you're in the know you'll realise just don't solve the problems in a pain free way.

[stevenicr]

I agree with your statements here, and certainly what I do mostly is not what I do all the time. Especially if it was like health info I'd dig deeper for example. I think part of my method is also what threads I click or don't click - so the experience is going to vary with that selection biases as well.

when people mention mac only (big thanks to those who do this!) - I tend to just believe and move on.

with some projects I open tabs for the article and all the alternatives people mention bookmark to dig into later.

So my methods are not always this or that - but as a general 'way of HN' for me, these methods are what makes HN a great time saver in general, but not always.