I'm close to 42 and I've been here for 15 years now (as a reader more than as a contributor); when I joined I certainly didn't expect to be here for that long. Testament to the quality of discussion, although it has gotten a bit partisan of late. I was on Slashdot prior to that for about 10 years. What was most appealing about HN was the information density, and the discouragement of jokes. I guess if you distil a UI down to the basics (HN just being pure text) there is a certain timelessness to it. Slashdot soon felt bloated and stuffy and I never went back.
Having said that, the fact that the 16-20 and 21-25 is so under-represented doesn't bode well for the future of the website.
Edit: I think 36-45 was the largest group when I started writing the comment. It isn't now.
Reading HN comments and stories at that time felt fresh: There were startups making money with software, there was people that valued different things and could have a good discussion. I think it was me naturally maturing from my 'communist' youth to 'capitalist' adulthood.
Funnily as time passes it seems theres more pro-open source feeling in here. Even some of the Destop Linux discussions that happen here remind me of old /. discussions. And it's hilarious that I'm often on the opposite side of the argument from when I blindly defended Linux 20 years ago.
I don't think it necessarily says all that much; at that age many are probably busy doing a lot of other things: going out, dating, studying, getting your career off the ground, whatnot. Things like children that you may have later in life also take up a lot of time, but the patterns of that are different (e.g. more "quiet evenings" at home).
Reddit stats are had to come by, but I wonder if the users who use it for mostly news and text rather than interesting images/videos/memes (would old.reddit.com users be a good proxy?) has aged the same way.
I think "faster" and "flashier" content wins out in social media, so for example the text interaction people have left for twitter and Substack. Heck, compare pg's amount of public comments here vs all his content on Twitter, sama when he was still with YC, the other YC people, etc. (I mean they probably have alts but still).
I expect that certain ways HN was designed are becoming less attractive as a platform for the population at large, like the 80-character title limit so that it could fit a title in a single terminal line (although I personally love this more than I hate it, keeps things to-the-point).
I wonder what the TikTok version of the HN-type community is like?
There's no reason to have one category for the 5 years 16-20, and the most common age group of Internet savvy working age adults to be lumped in two 10 year "26-35" and "36-45" ranges.
Oh well. It's just for fun I guess.
For example 30-40 age range would group together people with relatively similar life experiences, but a 10-20 age range would not.
To be honest, I don't understand what "they develop faster" means either. One year of age is objectively the same length whether you're 10 or 90. What has growth speed and brain development got to do with anything?
A 20yo is 100% older than a 10yo. But a 90yo is only 12.5% older than an 80yo.
The older we get, the closer in age we become.
Maybe there’s a better way of phrasing it, but I mean the rate of change is faster when you’re younger.
A 10yo will have very different preferences, life experiences, brain development, etc. compared to a 20yo (10y apart). Yet a 30yo and a 40yo (also 10y apart) will be fairly similar in these categories on average.
Because at 10yo the body and brain are changing more rapidly than they would be at middle age.
So if you were doing any kind of marketing or social science report based primarily on age, you’d typically want narrower age bands at lower age ranges and higher age bands at middle and older ages in order to create the most homogeneous groups.
While the demographics certainly skew in that direction, one of the nice things about HN is that it's not just software people. While I certainly comment more on software-related matters because that's where my expertise is, the largest value I get out of HN is from non-software/IT stories and the comments on those.
So I would strongly disagree with "this is a forum for entrepreneurs and software engineers". I'd probably stop coming here if that was the case.
It's a constant tug of war that. I would prefer if there were more deeply technical posts than general news posts one could read on Ars Technica, but I appreciate the anarchic "if people vote for it, it's good enough" policy this forum has.
Most surveys are not done with any such skill - they’re done to get the result desired.
Every introductory textbook on empirical methods has a chapter on surveys and questionnaire design, and there are also specialized books just on that.
Tech would sometimes really do well in listening to people outside its field.
I worked for 4 years with a team of mainly social scientists (agricultural economists, rural development researchers, etc) for a huge EU project that included plenty of surveys for 6 locations in different countries/languages. The effort to design the necessary surveys right and in different languages/cultures really amazed me. Some of the teams even got (good) papers out of it. There are so many biases that you can have with the wrong questions, groupings or even question orders.
One example I remember is that while in Germany (Altmark region) you could go straight to the people and ask them about their use of policy programme money, what had worked and what hadn't. You could not do the same in Croatia or in Bulgaria (we had study sites in both places), but had to layout questions differently, to get useful answers.
> 2 points
Why am I not surprised.
Hacker News is pretty serious compared to most social media, and I love that. Which makes some rare harmless comments or posts like this one actually good to encounter once in a while.
The seriousness-to-joke ratio of this site is just perfect in my opinion.
It's just that a lot of joke answers early on in the life of a post may make other users who may otherwise seriously respond, look at the responses and go "Aah it's a joke post/it's filled with joke responses, better not waste my time". Consequently the post may not see many upvotes and die out quickly.
In comparison, a post like this has the potential to easily get a few 100 upvotes with good discussions on the (changing/constant etc) demographics. But if a lot of initial responses are jokey in nature we may lose out on that.
To do a second upvote.
Thanks for this, that's a good point.
The fact that the ranges include ages > 105 and that there are many answers to it just makes the post some kind of spam.
- Boaty McBoat Face
> If you had a million years to do it in, you couldn’t rub out even half the “Fuck you” signs in the world. It’s impossible.
> It was so nice and peaceful. Then, all of a sudden, you’d never guess what I saw on the wall. Another “Fuck you.” It was written with a red crayon or something, right under the glass part of the wall, under the stones. That’s the whole trouble. You can’t ever find a place that’s nice and peaceful, because there isn’t any. You may think there is, but once you get there, when you’re not looking, somebody’ll sneak up and write “Fuck you” right under your nose.
I think that's also the point that J. D. Salinger may have been trying to make here.
Peace is a good, well constructed, true and correct data set. There is no peace. One must always correct for the lack of peace.
https://slatestarcodex.com/2013/04/12/noisy-poll-results-and...
edit: as of right now 105-255 is sitting at 3.6%, which is pretty close to the constant of 4%.
That's the "you insensitive clod" option.
INT_MAX 2**31 - 1