Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Favorite HN threads of all time
235 points by cellis on May 19, 2012 | hide | past | favorite | 51 comments
Have any Hacker News threads ever truly changed your perspective on life, or technology? Post them please.

This should be a good starting point:


There are boatloads of good content in there, plenty of which have the potential of changing your perspective on life, tech or just about anything else that is covered.

I feel like "ask hn" questions are very similar to a large portion of the initial content on quora and the initial communities were somewhat similar. I wonder if there is anything that YC could do to make hacker news better preserve the evergreen content on the site. Sometimes I check out a 2 year old thread and feel like commenting on it but then I realize that boat has already sailed.

Sometime in the last year someone made a comment in a thread on learning/insight that I have wanted to cite more than once -- the comment went something like this:

"When I struggle with something and finally figure it out, I write down the one thing/sentence that I could have told my future self that would have helped me understand it."

I have searched for this comment several times but haven't been able to find it.

Thanks. I've been on this site for a long, long time. I knew there had to be a curated list like this but Google turned up nothing with a cursory search.

I was so looking for that post frantically for the past couple of days but it just kept eluding me. Thanks a lot for posting it.

It makes me wonder how effective a well implemented human powered search engine could be. Its kinda overwhelming just thinking about it.

It's interested when you task reddit with finding something. It's like crowdsourced searching, with a few really resourceful people.

Thanks for sharing, I was searching for something like this. Usually I bookmark links which I like to re-read later on delicious. How often this list is updated ?

I've stopped updating it. It's at least a year stale.

This is the best HN comeback of all time: http://news.ycombinator.com/item?id=35079

Dunno. 5 years later and this pair of comments in that thread seems more interesting:



Dropbox is a huge business, with real revenue, which has helped millions of people. Tarsnap, despite cperciva's obvious talents, has not achieved anything like that level of success. dhouston wasn't flaming people in the thread, cperciva was.

Part of a really good comeback is substance. cperciva had the snappy retort. He really had won the Putnam. And on a much smaller scale, he won a comments thread on Hacker News. But the 20 posts there didn't create real world value. dhouston did. Maybe that's the real lesson of Hacker News. The best thread is the one you don't participate in, because you're working.

Tarsnap offers better security, and functions as more of a backup and less of a synchronization service. It's not really a head-on competitor to Dropbox, and doesn't especially favor non-technical users the way Dropbox does.

more of a backup and less of a synchronization service

I would say entirely a backup and not at all a synchronization service. If you're using Tarsnap for file synchronization, you're probably doing something wrong.

Tarsnap aims to be "good unix software" in the truest sense of the phrase: Pick one thing and do it well.

Is another difference that dhouston is building a huge corporation (with many employees etc), while you're building something you can run yourself? (Not a criticism - it's my aim too.)

No, I don't want Tarsnap to be something I run by myself. I want Tarsnap to be something which runs itself. I just spent two weeks at conferences and didn't log into any of the Tarsnap servers even once -- and when I spent 36 hours travelling from one to the other (Ottawa to Brisbane -- 23 hours in the air!) I was worried about whether my suitcase would make all the connections, not whether Tarsnap would break while I lacking internet connectivity.

But getting to what I think you were really trying to ask: Yes, Tarsnap is much smaller than Dropbox, and I'm happy that I haven't taken money from VCs (or any investors for that matter) who would push for faster growth. I'd rather have better product than more product; I will probably hire other people to help with Tarsnap at some point, but the question I'll have to answer is not "can this person do useful work" or even "can this person do task X better than I'm currently doing it"; rather, the question will be "can this person do task X sufficiently better to overcome the cost of my no longer understanding it".

Tarsnap is first and foremost about security. Security is about getting details right. And getting details right... well, that requires a level of understanding of how all the different pieces fit together which simply wouldn't be possible if I were hiring dozens of people and throwing them into teams to churn out new features every week.

Damn, you've caught my flair for understatement!

That would only work if dhouston wasn't also following and commenting on the thread :).

"But the 20 posts there didn't create real world value. dhouston did."

Do you believe that blog comments can create value in general?

I'm never going to escape that thread, am I?

Neither am I!

If it makes you feel any better, I now use it as my own, personal "don't be an ass" moment. If I feel the urge to say something snarky, recalling this interchange pretty much quashes it.

Allow me to join with PG in congratulating you on a great snappy comeback. Ignore the haterz, PG gets it worse and you wouldn't believe the ones I get.

In the same thread, one of the best comments (http://news.ycombinator.com/item?id=35139).

This is one of my favorite comments of all time, from mahmud. I've had it bookmarked for a while. The context isn't really important, because I think it was slightly OT comment. It's basically a suggested curriculum for studying Lisp implementation:


I have received more thank yous and mentions for that post than it got upvotes. It got 19 upvotes, but people still mention it.

(interesting to note that 2.5 years later, my opinion would be exactly the same, though now I would probably mention a few type-theoretic works; more foundational stuff, less petty bit-fiddling)

I've barely scratched the surface of that list, but I'd still be curious what you'd add to it. What type-theoretic work would you add? Pierce's TAPL, or is there something else that's more applicable to type systems in dynamic languages? What more foundational material would you recommend?

I would actually be interested in the foundational stuff if it might act as a primer for the heavier papers and books you listed there. I've read SICP and EOPL, but I found the chapter in Lisp in Small Pieces on denotational semantics almost incomprehensible, even though the later chapters on compilation were tremendous fun. I did take a stab at Semantics with Applications but couldn't quite get through it, although I probably didn't go after it with the gusto I should have.

Since then I've been getting into some more traditional compiler stuff (Appel's Modern Compiler Impl in ML) but have gotten hung up on trying to really understand parsing, which I have always skipped over like a smug Lisp weenie.

I am not ignoring this. I'm getting around to answering you. Perhaps out of band ..

Thanks, my email address is in my profile, but there's no rush of course. Your original comment provided quite a lot to indulge in.

Ask HN: What is your preferred Python stack for high traffic webservices? http://news.ycombinator.com/item?id=2910953

Introduced me to a lot of webapp stack pieces that were highly worth looking into.

It took me a long time to find this thread as I couldn't remember exactly what the discussion was about and parsing old emails and chats finally revealed it.

This remark by codyrobbins - discussing the 'I could care less' idiom - is perhaps my favorite HN comment of all time. http://news.ycombinator.com/item?id=854042

This thread (a couple parents up from that one) provides the appropriate context. http://news.ycombinator.com/item?id=853100

This is my go to for inspiration:


This one, about hacking hacker news, is pretty epic:


Ask HN: Is there something like HN for successful entrepreneurs?


See nirvana's second comment in the thread, especially:


What is your favorite programming language? https://news.ycombinator.com/item?id=3746692

This had some really neat discussion about the finer points of some languages.

I've used this thread as an example of who the members of the HN community are, and what startups they make:


I saw this one when I first started visiting HN: http://news.ycombinator.com/item?id=1924909

Why does time go faster as we grow older: http://news.ycombinator.com/item?id=713339

What's your favorite HN thread: http://news.ycombinator.com/item?id=2158116

A good place to look is at the HN lists page. Not exactly a curated list but you can usually find some of the more popular and 'hot' discussions here rather than the frontpage. Wadding through the top contributors usually lets you find some quality threads.


leaders Users with most karma.

best Highest voted recent links.

active Most active current discussions.

bestcomments Highest voted recent comments.

noobstories Submissions from new accounts.

noobcomments Comments from new accounts.

My least favorite, but the one that has probably influenced me the most:


I don't want to sound negative, but.. there is so much content produced every day on sites like HN that I feel a bit like the past discussions don't matter anymore at all.

I disagree strongly; there's a lot of gold in HN's archives. Some things don't change, or don't change that quickly, and there are some topics (even in the young field of CS) that are classic, and some great discussion has already taken place on HN about them. Why do you think that doesn't matter? HN isn't even that old.

What is the best way to find the gold? Looking for an efficient way to find the good content from days of yore.

EDIT: Other than Ask HN Archive, mentioned above.

I don't have a silver bullet, but if you have a topic you're interested in, the search provided by Octopart is very good (bottom of the page), and it's not that time consuming to dig through the results. That's how I found mahmud's comment on Lisp compilers, because I was looking for discussion on HN about it.

Obviously some topics are harder to search for (for instance if you wanted to read about what HNers have said about Techcrunch over the years) but for the good and classic stuff (programming, math, etc) I think this works reasonably well.

It would actually be really cool to see what HN threads are linked to the most from HN. I bet that would be a decent way to find gold, especially on stuff you weren't specifically looking for.

I disagree, older content can be much better, particualrly on somewhere like reddit where there has recently been a hge influx of users diluting the current content, the older threads such as those on askreddit were of a higher quality and had much better discussions.

Go here and sort by 'points': http://jflatow.github.com/popdots/

http://news.ycombinator.com/item?id=354539 (Mark Cuban is bullish on America)

The day after President Obama won the election and I lost over 60 points in that thread for calling him out, or rather the people for getting suckered into the race card. (I am not a US citizen and feel vindicated now, what a con job ... oops politics not good on HN)

Maybe it's something to do with your delivery of said feelings, rather than people behaving like you believe they are.

Why do you feel vindicated now?

Will only discuss offsite.. not going to martyr myself again pointing out the obvious.

add: for starters, read the top comment on that thread, then you tell me if it was justified being top: http://news.ycombinator.com/item?id=354589

Martyr yourself? Your comment basically says "Obama isn't black enough, and he's a Muslim, but not enough of one." It's so far removed from reality that it's hard to even think of a response other than "flag for craziness". You didn't get downvoted for saying something controversial, you got downvoted for stringing a bunch of incoherent half-thoughts together.

And yeah, the top comment is pretty good.

You seem very bitter about something that would not make any normal person bitter. It's been several years since that comment. Move on.

I'm getting good lulz right now, and we'll see who has the last laugh in all of this. I have moved on.. just that the OP asked about favorite threads.

Applications are open for YC Winter 2021

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact