Show HN: Visit the front page of Hacker News on a random day (randomhackernews.com)

267 points by nickipedia ↗ HN
Hi HN. I was surprised that there wasn't a feature here that lets you go back in time to the front page of Hacker News on a random day...so I made one. http://randomhackernews.com is a simple HTML page that navigates you to the front page of HN on a random day between today and February 19, 2007 (the oldest date I could find with content).

I made this for myself, but figured others may find some interest in it.

78 comments

[ 2.5 ms ] story [ 145 ms ] thread
Feb 19, 2007 was indeed the day Hacker News came to be - https://news.ycombinator.com/front?day=2007-02-19
Haha, the core of this website really hasn't changed, has it?

> Why Startups Don't Condense in Europe

> DabbleDB -- web-based DBMS coded in Smalltalk

dead comment: > Fedora is far better than ubuntu

Dissing Ubuntu in 2007 is a bit prescient: it was one of the best distros around, and Fedora (Core) a buggy mess.
Wow, gave a strange feeling to see DabbleDB mentioned. I had launched a product in the same category. It didn't get traction, and I lost interest in it for some time, but I kept it running and it's still available and has just been rewritten in F#. My problem now is the same as nearly 20 years ago, find its niche. I'm thinking of making a Ask HN about that.
Where there any success stories that came out of Stanford start up school? Or was the success Stanford monetizing the process?
I counted 14 out of 30 links on the front page back then still work and appear to refer to the same article / something relevant to the original submission.

Link rot is always disappointing, but I actually expected it to be far worse for links from 17 years ago.

It was running in stealth from https://news.ycombinator.com/front?day=2006-10-09 thru https://news.ycombinator.com/front?day=2006-10-22, with a few gaps. Then https://news.ycombinator.com/front?day=2006-12-14 with one story: "Productivity Surges During Reddit Downtime" (not much has changed, as PhilipRoman said...)

And then https://news.ycombinator.com/front?day=2007-02-19, as you say, which was the public launch. Let me dig up the link...here it is: https://news.ycombinator.com/announcingnews.html

And then https://news.ycombinator.com/hackernews.html 6 months later.

I posted links to Startup News pages the other day: https://news.ycombinator.com/item?id=39237746.

We should make the top left say "Startup News" for /front?day=2007-08-13 and earlier!

Please do not hijack the back button.

Your website should send a 307 Temporary Redirect to HN, so when I go back, I return to where I was, instead of being sent to another random day.

If it is done on the client, look into `window.location.replace`: https://developer.mozilla.org/en-US/docs/Web/API/Location/re...

It's incredibly annoying when this isn't taken care of. Alt + Left Arrow becomes impossible to use and I have to resort to using my mouse.

Don't Microsoft's support pages [1] do this as well? I just checked with a random support page [2] and it looks like the middle redirection url is the same. Is that a different thing?

[1] support.microsoft.com

[2] https://support.microsoft.com/en-us/windows/fix-file-explore...

(Firefox if that matters. Currently on Linux, has happened on Windows)

Edit: I just checked on Ungoogled Chromium and it looks like this (the microsoft support) isn't an issue. It's only on Firefox. That's bizarre.

No issue on my end. Is it already fixed, or does Android Chrome just overcome poor implementations?
Looks like it's a firefox thing. I tested in Ungoogled Chromium and it works fine for me too.
Can confirm this behavior on iPhone 15 pro max safari ios 17.3
Here's a bookmarklet that does the same thing:

    javascript:(()=>{d=new Date(1171843200000+Math.random()*(Date.now()-1171843200000));window.location.href=`https://news.ycombinator.com/front?day=${d.getFullYear()}-${('0'+(d.getMonth()+1)).slice(-2)}-${('0'+d.getDay()).slice(-2)}`;})()
Please make it open in a new tab, that'll make more sense (I suppose).

It's not pleasing to go back to use it again.

Not usually a good idea to mess up with the default, and you can always middle-click or ctrl+click or cmd+click or long press. If you open in new tab by default you risk breaking user expectations, disorient them and you break users wanting to open in the same tab.
> If you open in new tab by default you risk breaking user expectations, disorient them and you break users wanting to open in the same tab.

This feels backward to me. I don't do user research, but is that a real pain point? Just... close the old tab?

I won't be as good as the person behind css-tricks [1] and definitely not an UX expert, only interested in the topic, but let's try:

- I'm tech savvy enough and want to open in the same page: you are preventing me from doing it easily, while it's easy for you to open it in a new tab if you want it this way.

- I'm not tech savvy. I barely know to use a browser. The link opened in a new tab and I didn't notice. The back button I know to use is greyed out. I'm confused.

Said differently: you can always open normal links in new tabs. I do it all the time, to the point I almost never understand someone requesting a link to open in a new tab: help yourself, it's so easy. Don't rely on the page author to make the call for you because they can't please everyone. However, target=_blank links are harder to open in place, while it's reasonable to want and expect a link to behave normally. I get a bit mad when for some reason a new tab was opened and I didn't want to nor I expected it. Just leave me alone, you know. If I don't want to come back to the previous page, or if I want to do it using the familiar back button (or gesture), you should not force me to close the previous tab and break my history buttons.

Moreover, nothing usually tells you that a link is going to behave differently as you expected (principle of least surprise), and even if it does, I shouldn't have to install an extension or fiddle with the web dev tools to work around this and get the default and expected behavior, that's terrible UX for obvious reasons.

Deviations from the default behavior are sometimes better / necessary but you always risk causing surprise and thus confusion, which you should also avoid for good UX. Even someone who want to open the link in a new tab probably expects a link to open in place. It's usually impossible to know in advance if users will prefer opening a particular link in a new tab, especially since different users will have different preferences in the general case. As a user who will sometimes prefer a new tab, it's also uncomfortable to rely on the author of the page to make the call so you may as well just help yourself.

So, for good UX, the normal behavior is supposed to be the best option, unless in specific cases like those discussed in [1], basically when the confusion or frustration caused by the normal behavior is greater than for the new tab behavior, like loss of work or disrupted playback.

I don't know if there are proper studies about links specifically, but I'd expect existing research to prove the principle of least surprise, generally accepted as basic UX principle. I'm a bit lazy to find such work, on a phone during a small insomnia, I hope you'll forgive me. A "(opens in a new tab)" tag partially fixes the issue but doesn't fully address it.

[1] https://css-tricks.com/use-target_blank/

> I was surprised that there wasn't a feature here that lets you go back in time to the front page of Hacker News on a random day

May I ask why were you surprised? Which other website has this feature?

This sort of thing is a feature I often feel is missing, sites have "last week", "last month", etc., and you can see what stories have been top. But you can't then go back to last month and see what the top 'stories' for the preceding month were.
Not exactly the same, but in the same vein, Reddit has a "Random" button that takes you to a random subreddit. Google has "I'm Feeling Lucky" which gives you a random search. A "random" button seems like a relatively common little easter egg.
Surprised by the surprise too but why not, but to answer, xkcd.com has it.
What a gem 2010 was.
Thanks for the feedback everyone! I've updated the website to open HN in a new tab
I have one more suggestion, make that title more catchy, you can try such as

1. HN Lucky Dip

2. HN Roll

Does anybody want to give it a try?

I'm happy to see people looking at /front?day=yyyy-mm-dd because it's a trove of great stories and threads. It's linked from "past" in the top bar, if anyone didn't know.

Please realize that it isn't an exact representation of the front page at any particular time. For that you'd want a point, not an interval, and certainly not a 24 hour interval, which is what /front?day=yyyy-mm-dd is showing.

What's it called when you make a single photograph by overexposing a series of time-lapsed frames into one? It's like that. I found it trickier than I expected to write the code to do this, but I think the result serves its purpose: to give people a way to find frontpage stories they may have missed.

We do this via a combination of votes and frontpage time, but only back to 2014-11-11 which is when we started logging the lists of stories on the front page. Those are the 'frames' in the above analogy. Before 2014-11-11, we only rank by votes because it's all the data we have.

If you want to look at actual snapshots from the past, archive.org has a lot of them, e.g. https://web.archive.org/web/20160620131548/https://news.ycom.... Perhaps OP could add random links to those as well!

> What's it called when you make a single photograph by overexposing a series of time-lapsed frames into one?

Long exposure

But not quite that either right?

That’s what you have if you mechanically keep the shutter open for a prolonged time, exposing either film or a sensor to light.

But from what dang was saying, he seemed to describe emulating that effect in software on discretely captured frames.

Sounds almost closer to onion skinning, and in particular the part of this article https://en.wikipedia.org/wiki/Onion_skinning where they say:

> This effect can also be used to create motion blurs, as seen in The Matrix when characters dodge bullets.

My favourite date is Erlang Day, after a request from pg to have more technical content: https://news.ycombinator.com/front?day=2009-03-11
There was an Erlang Day sequel later that summer: https://news.ycombinator.com/front?day=2009-08-20, which I remember because I unintentionally started it and pg got mad at me.

IIRC, the front page before that was filled with stories about _why (they're split between https://news.ycombinator.com/front?day=2009-08-19 and https://news.ycombinator.com/front?day=2009-08-20 now), and people were complaining about that. I happened to have an Erlang article I'd already been meaning to post, so I posted it and it triggered a deluge, followed by complaints:

Ask NH: How annoying is Erlang day? - https://news.ycombinator.com/item?id=776789 - Aug 2009 (0 comments)

Ask HN: Why is all the news about Erlang today? - https://news.ycombinator.com/item?id=776622 - Aug 2009 (4 comments)

Are we celebrating Erlang's birthday or just that Erlang hackers cracked HN? - https://news.ycombinator.com/item?id=776519 - Aug 2009 (4 comments)

Poll: What do you think of Erlang on Hacker News? - https://news.ycombinator.com/item?id=776427 - Aug 2009 (3 comments)

Enough Erlang - https://news.ycombinator.com/item?id=776316 - Aug 2009 (9 comments)

Purge Complete: No more Erlang, _why, or Zed - https://news.ycombinator.com/item?id=776196 - Aug 2009 (25 comments)

I hate the Erlang days around here..... - https://news.ycombinator.com/item?id=776116 - Aug 2009 (9 comments)

Ask HN: Are we intentionally voting on Erlang articles? - https://news.ycombinator.com/item?id=776047 - Aug 2009 (6 comments)

... followed by pg, after a long day at Demo Day, manually killing all those threads at dinner using his iPhone. Nobody knew about that latter bit.

Ask HN: How did the Erlang articles disappear on HN yesterday? - https://news.ycombinator.com/item?id=778319 - Aug 2009 (11 comments)

I always wondered why I couldn't find any trace of this in the archives, as I distinctly remember the scolding I got! I've gone and unkilled them now, which should restore roughly what it looked like.

That's too funny.

Unfortunately, jumping across timelines reminds us of digital decay, a very real thing, when you try to visit some old stuff.

Whoever figured that caching will save HN's perpetually melting server is about to have a heart attack.
I was worried about that too, but it seems to be holding up ok!
Nice work. Slightly related question: say one hasn't visited HN in a few weeks and wants to see something like the top N posts of a given week. Does this exist?
Just yesterday I realized that post and comment ids are incremental and went back to look at the first year of posts!
It would be great if HN automatically added a "wayback machine" link for articles over, say, one year ago.
I also see value here for anyone looking to boost their credibility if needed.

"Yes Jim, my project did in fact hit the front page - take a look."

Oh yeah? Well I was Time magazine Person of the Year in 2006!
Spotty memory but the two posts that made me check HN daily were:

- How do you store data to last more than 50 years (diamonds and titanium were mentioned)

- The announcement that bitcoin reached 1 USD, a great party story (TL;DR I tried buying 100, had issues w/ payment, got pissed, bought 200 instead, lost wallet a few months later, would have sold it super early tho)

Clicked all the links on today's front page? Just go here for more.
I wonder if this would make a fun game, like geoguessr. You see the HN page from a random day, perhaps with some date scraping to remove super obvious clues, and you have to pick the year/month.
> Man saves wife’s sight by 3D printing her brain tumor

+

> [US president] calls for municipal broadband

+

> A Quick Comparison of [...] vs. Rust

+

> Show HN: I built an iOS remake of that old [...] game [...]

+

> [...] Tops [...] as World’s #1 [...] Platform

+

> Student [...] AI [Competition] [Year]

+

> Bitcoin crashes over [X]% in 24 hours, under $[Y]

+

> Frameworks are for hacks, libraries for seeds

(Of course for some of the above it would likely be trivial if you just copy-paste them in a search engine...)

> Youtube remains unprofitable 3 years after Google acquisition (dealbook.blogs.nytimes.com)

2009/July

First thing I did was relive a moment almost 10 years ago when I had a Show HN on the front page :-)
Really cool to jump in the past.

Thanks for making that!

How does this belong on HN frontpage? This is not just trivial, as another commenter 6 hours ago (!) posted: https://news.ycombinator.com/item?id=39262873

It's not hacker, it's not news, it's not innovative, it's simply boring, and would be cool if a 10 year old did this. Sorry for this rant, but I'm reading on and off HN over ~10 years and contributing under various user names, and I read comments like mine over and over again, and each time agreeing more.

Maybe everyone who definitely will downvote my post should read the first guideline "What to submit":

On-Topic: Anything that good hackers would find interesting. That includes more than hacking and startups. If you had to reduce it to a sentence, the answer might be: anything that gratifies one's intellectual curiosity.

A random() over the time HN exists, with a link to the generated date, which doesn't even handle back correctly is not something good hackers find interesting.

On the other hand, there are still lots of on-topic, curiosity inducing topics around here. Just getting scarcer every year. Oh my. Good night :)

What is interesting here is not the effort that went into the thing (although I am now planning a rewrite in Rust ;), but the output. Assuming it is true that HN standards for 'interesting' are falling, as you say, clicking on the link will tend to present you with a more interesting edition. Assuming it's not, it's still fascinating to look into a time capsule like this.