Ask HN: Should we plan for a life after HN?

53 points by hliyan ↗ HN
While reading about the death of Omegle, I came across this comment and it sent a chill down my spine: https://news.ycombinator.com/item?id=38201035 ("Omegles come and go, but "[how is it] hacker news?" is here forever.")

Although the comment was not necessarily about the longevity of HN itself, it made me realise that HN, like any other online forum, has a finite lifespan. At some point it will end (although that seems like a distant possibility right now).

Should we plan for this? I.e. building some sort of mirror platform with the same profiles, same data and as much of the algorithm as possible, with the agreement that we meet there should HN unexpectedly meets its demise.

I understand that a huge part of why HN works is dang. I'm unsure how we mirror him though.

74 comments

[ 3.2 ms ] story [ 159 ms ] thread
I worry about this myself. I feel like HN is the last refuge of what the old Internet was supposed to be like.

I'm not sure it can be replicated. The future of the Internet seems to me to be small private "hidden" communities and mega platforms filled with garbage. No room for a middle

Maybe a paid platform?

Either a private, paid platform, or a Wikipedia-like foundation that relies on donations.
I'm also tempted to start by thinking about the moneys — after all it's hard to be sustainable without revenue. But that's only part of the problem and money is by no means a solution. Indeed the revenue streams discovered by Google and Meta — while responsible for the financial success of those companies — are also responsible for disinformation and degrading quality from user perspective.

If the thing that makes HN enjoyable is the community, maybe we should see it more as a commons and worry about how it is/can/should be governed. Taking Wikipedia as an example, if the worst that comes out of the money problem is an occasional desperate plea from Jimmy Wales begging for a small donation, I'm perfectly happy.

Would it be possible to foster a similar community using a platform like Mastodon? If, yes — it might already exist... Anyone have a pointer?

> the last refuge of what the old Internet was supposed to be like.

What, how? HN is one of the most squeaky clean - corpo - heavily moderated sites with a extreme bubble centered around SV tech.

Its basically the exact opposite of the old internet.

It’s primarily filled with intelligent people interested in technology. Before smartphones, that was the primary population of many (most?) websites.
> intelligent people [...] was the primary population of many (most?) websites.

Oh sweet summer child.

GP post is correct in the general sense that the Internet used to be a thing only accessible to researchers at major universities; and therefore a place where everyone could be assumed to have some basic level of technical knowledge (if not common sense), and to consider themselves "an intellectual" of some kind or another.

GP post is technically incorrect, though, in that during this time, the web didn't exist yet. The tech folks were on Usenet and IRC, not on "websites." (In fact, the creation of the web and "Eternal September" basically coincided: the web was the killer app that got regular people to sign up for Internet service. Despite HTML being designed for scientific-research use-cases, the web that actually formed in practice was "for normals" from the very beginning.)

I came online, completely accidentally with the Eternal September crowd, we beta tested NCSA Mosaic mostly as a Usenet reader 'cos there were almost no websites yet. Still it was not only researchers, "Eternal" September is a reaction to earlier Septembers with a large influx of newbies. And even then there flame wars were a thing.
Ha, agreed fully that the "old internet" was far more of a cesspool than most care to remember. We look back with rose-colored glasses. We remember the good and forget the bad.

USENET was not only every bit as bad as the worst of 4chan -- it was worse, because there was no effective moderation.

That said, there was also a large "signal" component in the "old internet" that is now lacking in most of the New Internet. Places like HN remain as lonely outposts of that lost world, where people wrote in entire paragraphs, to convey complex and meaningful ideas for mutual discussion -- rather than just firing punchy little 140 character epithets and 10 second insta-videos off into the void and then forgetting about them, moving on to the next flash in the goldfish attention span pan.

That level of sustained, complex actually seems like it was more common on the "old Internet," even recognizing fully that the "old Internet" was also overflowing with dross and garbage.

As much as I like my Internet free as in beer, I do think that in order to transition from ad-supported content platforms, some form of micropayment or subscription model works best to support creators and bastions of intelligent discourse. I don't mind chipping in for content, (ie independent reporters on substack, local NPR station, Wikipedia), the value exchange is positive for me.
Don't worry, existing entrenched tech giants will kill this idea before it could take root: this week Facebook started offering a subscription in the EU instead of using your data for ads. They however ask for 10 EUR (and 5000 HUF in Hungary which is 13 EUR) which is far , far more than what anyone sane wants to pay for FB. So they can go back to the regulators and say, see, no one wants to pay for this.
(comment deleted)
> The future of the Internet seems to me to be small private "hidden" communities and mega platforms filled with garbage.

What you're describing are two attractor-points in community design space, for communities that don't put any effort into fixing their position elsewhere in the design space.

Managed-membership communities don't grow by default, because by default they don't advertise / recruit members. So they have strong cultures, but they're "hidden."

Unmanaged-membership communities start off small and nice, sort of like closed-membership communities; but then they grow "by adsorption" — diluting the original spirit of the community with greater and greater numbers of people who are less and less committed to the premise of the community. Until the community is a generalized nothing, a random sampling of all of humanity.

A managed-membership community that advertises / actively recruits could be nice, but it would have to first also avoid the third natural attractor-pole in community design: becoming a cult-of-personality-driven money-making scheme.

The tildeverse? SDF? Mastodon? Slashdot? The chans? Lobsters? Gemini? Superhighway84 (not sure if this one's still alive?) Undernet or Libera on IRC?

I guess it depends on exactly how picky you are about pattern matching to "the old Internet," but a lot of that stuff feels like a throwback to the early Internet or even BBS era... none of it is mega platforms and most of it isn't really hidden... all of it is pretty cool... HN fills a great niche to be sure

Hold your nose and throw in a few of the better run subreddits until Lemmy takes off...

And WordPress is now paying a guy to work full-time on ActivityPub integration, hold on to your hats ...!

> it made me realise that HN, like any other online forum, has a finite lifespan

Everything we touch has a finite lifespan, and I don't think that's the important thing to think about.

I would further comment, but the guidelines[0] last point says not to.

[0] https://news.ycombinator.com/newsguidelines.html

> Everything we touch has a finite lifespan

Indeed. Nothing is permanent. The only constant is change.

We'll just follow dang wherever he goes :P
YC could train a dangGPT to auto-moderate. I don't see HN disappearing. It is niche enough and YC likely uses it.
Maybe dang already did that and is now sipping Margaritas on a beach somewhere
I can't imagine doing what he does and not needing a margarita or several. I don't know how he survives it.
He gets a paycheck.
"ChatGPT42, generate a working copy of the old Hacker News site for me to read based on today's news and the probable reactions to it. Have it respond to my posts accordingly."
Now what's left besides such corners is the "commercial internet" run by A/B testers, KPIs and OKRs.

Ironically, that's what Venture Capital is all about. Mostly.

assume nothing will exist "forever"

HN will one day go the way of the dodo bird

Whether that is in a month, year, decade, or century is up for realization

But like every other website/service in existence ... it will have its time, and it will fade/die eventually

Should we plan for a life after HN?

There is no need to wait for such an event. In my opinion what makes this site unique is how it is run rather than the tech. One could make note(s) of how Daniel manages this site and try their hand at mimicking the patterns on existing sites. People could attempt to replicate something close to his methodology, guidelines and levels of engagement on their own self hosted Postmill, Mastodon, Forums, Chat sites, etc... In my opinion there is no harm in having multiple great sites.

"People could attempt to replicate something close to his methodology, guidelines and levels of engagement"

I wonder how close are we to getting an AI to do this kind of work? Pretty far I would guess?

Managing a community requires the uncountable efforts of the human soul. As of the current date, large language models do not possess one.
That's rather eloquently romantic, but doesn't actually seem to mean anything beyond "it requires some ineffable quality of vague goodness that just FEELS right to me, I know it when I see it."

What can be done besides "emotional feedback subjective gut check on how it makes me feel" when we propose to analyze the scenario?

Based on the commentariat’s enthusiasm for Section 230 reform then, the community wills its own destruction.
Calling up Dang actually points out the one major weakness of this site from the outside. I don't know what his succession plan looks like, but if he gets hit by a bus, this site could change dramatically (for the worse) and almost immediately.

When one person is your linchpin, you have one major, and guaranteed source of failures unless you take steps to mitigate. So it would be interesting to know what kind of succession plan there is, I guess.

Calling up Dang actually points out the one major weakness of this site

Semi-related fun side note: In my former workplace a coworker coined the term "Single point of success" after management said he was being negative by calling things a single point of failure. Such bottlenecks can't be solved by one person. It takes a village.

That's why I suggest everyone here could take notes and at least try to do the same things on their own sites. We can't change the big platforms but those here that tinker offering up their own time and discipline to host small communities could try to make the internet a slightly better place. I've tried my hand at this and burnt out after a decade. That's why I appreciate this site as I can just be a participant.

[Edit] Based on dang's comment there are other moderators and that makes sense given he would have to sleep at some point, otherwise people would be saying "Mods are asleep" and the chaos would ensue.

This problem has recurred throughout history.

The issue with the "study this person who does it well, take notes, it takes a village!" approach is that that quicky ossifies into a bureaucratic morass of petty tyrants who don't actually understand (or care about) the principles behind whatever that person did. They instead engage in endless rules lawyering about technical minutiae. Some do it simply to feel powerful; others genuinely feel that rules lawyering is helpful.

This becomes self-amplifying as those who actually care about the underlying principles and themes, as opposed to petty legalism, tend to leave the project permanently and go elsewhere. Meanwhile, those who care about and prioritize bureaucratic bickering are attracted to the project, eventually attain a larger share of the powerful positions, and build more of the same.

See the evolution of Wikipedia for a great example of this process in action.

approach is that that quicky ossifies into a bureaucratic morass of petty tyrants

Agreed, I see that happen all the time, especially with chat platforms for some reason. That's why I also said discipline. Having run large forums the temptation to make rules and power-trip is human nature. The biggest battle is with ones self and not letting others help moderate unless they share the same discipline. Finding people that share the same values and are not at risk of power-tripping is hard. Assistant mods would have to know up front they would be booted as soon as they exhibit such behavior.

I would suggest people first start up some little just-for-fun thing and learn the hard way what the biggest struggles are before creating something they truly are passionate about.

The ones who are power-tripping are bad. The ones who genuinely believe that their rules-lawyering is actually good and beneficial for the whole community are even worse.

“Of all tyrannies, a tyranny sincerely exercised for the good of its victims may be the most oppressive. It would be better to live under robber barons than under omnipotent moral busybodies. The robber baron's cruelty may sometimes sleep, his cupidity may at some point be satiated; but those who torment us for our own good will torment us without end for they do so with the approval of their own conscience." - C.S. Lewis

Dang is good at a lot of moderator things, but his best attribute is that he seems to be capable of disentangling his own personal views on a topic from enforcement of objective norms of decency and respect. Less than 1% of people can do that; they prefer to bludgeon people with technicalities in order to enforce their own views -- and most believe that they are doing good when they do it.

his best attribute is that he seems to be capable of disentangling his own personal views on a topic from enforcement of objective norms of decency and respect

I could not have said it better myself. That is a perfect description of what makes this a unique site.

This is why I think paid moderators like dang are better.
I'm not the only moderator who knows how HN works. Just the only one who posts publicly these days.
I'm certain you already have the dataset necessary to create an AI replacement of yourself as the moderator, even with the existing available models.

If this isn't already in the works, it should be as a key-man insurance policy.

I'm sure the other moderators would be fine, but they couldn't ever carry that dang weight.

I think we might find that human, paid moderators is one of the factors that make HN succeed where others fail
That's fair. And thank you for putting my little mind at ease.
[flagged]
Because he's consistent and has a real nose for actual low-quality posts. Plus the massive amount of moderation we don't see. Also the whole thing where he's a real person you can meet, not a "community standards team".
Like I'd be happy to contribute financially to a HN elsewhere if the need arose.

We'd just have to work out the details, I think the underlying software HN runs on is open source, so it wouldnt be too hard to do as a community project.

As long as Y Combinator is doing fine, HN will probably be fine too.
It would be very strange (and rather sad) for you to remain the same person exactly and HN to be exactly as it is now in perpetuity. Perhaps the next thing will be better? For me HN > /.
> Should we plan for this?

No. Most things don't end abruptly. There are signs it's ending. People will migrate to something else as they always do.

> some sort of mirror platform with the same profiles, same data and as much of the algorithm as possible

No offense, but you’ve just distilled the blindspot behind a lot of social startup ideas. Good communities are not a product of technology or data.

Technology and data facilitate different cost and scaling opportunities, but are just the bones upon which a community might be built.

This community is good because of excellent human involvement through dang (and others at YC) and through the luck of having the right, responsible users around so far. There’s no recreating it.

Sometimes, you just have to enjoy things while they exist and then let go of them once they don’t.

Just take the login page. It’s obvious that what makes this place special isn’t fancy technology.
If HN goes away it's because it's replaced by something else.

Something that meets users needs better.

What's there to plan for, except if you intend to make that new thing to supplant HN?

> If HN goes away it's because it's replaced by something else

Why are you so sure of this? The internet is chock full of counterexamples to the idea.

> If HN goes away it's because it's replaced by something else.

Have you tested the site backups? How do you know that dang's backups work? when's the last time they were tested? An act of god could take out the servers that run this place, and their backups, and the site's history and contents would be gone. you'd go to http://news.ycombinator.com and get an error page that says "we're having trouble serving your request." Or worse - no response at all.

Would the site survive hours or days of downtime? would we come flooding back to an empty site and just start filling it back up again after our threads, our /comments, out /favorites, our /highlights were lost? It would be like a death in the family. We might take off, never to come back instead.

This is not to impugn dang's computer skills. I'm sure there's a good plan in place. But you might want your open copy of the pages you want to keep.

I think I joined the first month HN was introduced or around there. It really hasn't lost too much of its "feel" to me since then. Seems like it's pretty active with no plans to change, but who knows?
For HN we should have a full blown desktop client and server tools for the p2p distribution of the archives. Then build the stuff to resurrect the front page and recent posts when they become unavailable. Eventually the hn server does nothing but sign postings and count scores. Users who've accumulated to many points can sign their postings themselves until such privilege is revoked.
Technology does not create or kill communities, the people do. If you replace dang, you have a different website than hacker news.
> I understand that a huge part of why HN works is dang. I'm unsure how we mirror him though.

dang possibly has the most prolific and easily accessible structured data on his moderation decisions and replies we can have for modeling the domain specific behavior of a single person... surely we can train classifiers and language models for him

I suspect that enough of the y combinator crowd gets enough value out of it (and y combinator itself) that it is in no immediate danger - either financially or from yc deciding to shut it down

Besides - it runs on a single box last I checked

Change is inevitable, and going to extreme lengths to preserve the current status quo is almost never worth it. Consider the recent death of Twitter, which seemed so real that tweets were being routinely embedded in news articles. I personally spent a lot of time there, but when it became unusable for me I moved on, to discord communities and bluesky. These are not the same, they are not a twitter-replacement per se, but they fill similar niches in my life. This process is normal, and has existed for all of human history. People are adaptable.
Twitter is not dead; it rather caters to a different and broader audience segment now.

What happened is that it is no longer an institutionally governed "safe space" where only leftist progressive views are allowed to be voiced in public, and any dissenting views were censored.

The segment of people who think that applying that sort of leftist ideological purity filter on all public discourse is a good thing are leaving, while another (and ultimately much larger) segment of the public who highly value free speech are joining.

I don't want to hear your ideology of how newly empowered bigots harassing folks off the platform is good, actually. Twitter used to be good and useful; removing actual verification, and then tweetdeck, and now you can't view threads without logging in, these things have made the site worse.
Merely disagreeing with liberal or progressive viewpoints does not, in itself, render someone "a bigot."
Lemmy already exists
I’m old so I have seen many communities come and go. Usenet went from being the go to source for documentation and troubleshooting to a front end for IRC groups. Those IRC groups were great but then died all of a sudden.

It’s even happened with personal communities. For four straight years, my friends and I primarily stayed in contact and shared pictures of our latest adventures on Livejournal. Or hell, I had an entire long term relationship that largely happened because of Twitter.

As those services disappeared, I changed too. Sometimes I found replacements and other times, I found that I don’t need those particular dopamine hits.

But the one thing I do know is that whenever someone started planning a replacement, the world went off in a different direction. My LJ usage was a great example - there were a lot of conversations about replacing it with something else but none of those conversations predicted that an identity required platform like Facebook would replace it. We actually thought that it would be replaced by something even better for privacy. Whoops!!

If dang broke off and started a for-pay site he’d get rich. The numbers are sort of unavoidable. If a million people subscribe to HN and only 1% stayed with him to pay $5/month for the ad-free version to start, it pencils out nicely.

He could easily afford an apprentice, though finding a good one would be tough.

I would love to see that happen be dug before the burnout kills him or drives him batty.

If HN died then maybe lobste.rs would become more popular. I think Lobsters does some things better (tags you can filter) but posts currently get very few comments. However, the invitation requirement might be too much of a barrier for many people.
Just let it go, but keep the memories (perhaps save some interesting content somewhere). Every user generated content platform has a finite life and lives with no way to replicate it as-is elsewhere. People have attempted to replicate many communities with varying degrees of success, but it’s just not the same (that doesn’t mean it’s always worse when replicated…it’s just not the same anymore).

One of the hardest but useful lessons in life is to move on (and figuring out when and how to move on).