same

Some clients don't manage certain events (e.g. polls can be posted to Primal but will be invisible on Nostria because it doesn't support polls, but it supports other things). So that's one reason you may not see events depending on the client. I explain events and demo this here: https://www.youtube.com/watch?v=wuJMOXz0a-g The video also demos how to run your own relay and back it up so you never lose your data as well (recommended). The other problem is distribution of attention for >95% of users: https://daniella.io/nostr-retention-structural-diagnosis/

Funny, I randomly saw this video this morning and also wondered the same thing. He's close to Jeff's thesis without seeing it yet (mainly blames corporations), but I was wondering why MSM is platforming him so much now vs recent ai hype cycle narratives.

I used Claude to verify claims against the source code but the experience of being censored is mine. The analysis is mine. The suggestions are my own. AI is a tool just like a search engine. My response stands either way: I demonstrated why Stacker News is proof of stake based on personal experience and the code, and because of this I won't be using it anymore. No hard feelings.

Thanks @Scoresby, I appreciate the constructive discussion and the invitation.

You're right that on the homepage ranking, a sat is a sat. I went through the Stacker News source code to verify your claim, and the ranking algorithm (ranktop/ranklit) does use raw sats, not trust-weighted votes. There's even a comment in the zap handlers: "This is mostly a placeholder because we are running a no trust experiment." So on that specific point, you're correct and I want to acknowledge that.

But the economic layer underneath tells a different story:

The reward distribution worker (earn.js) ranks qualifying items by weightedVotes - weightedDownVotes, and those are trust-weighted. Trust is built through a PageRank-style random walk seeded from the SN founder (with territory founders added as additional seeds in their territories). So trust doesn't decide what you see on the homepage, but it decides who gets paid from the rewards pool.

Here's where the system becomes proof of stake rather than proof of work: When users downzap the same content, they build mutual trust in the trust graph. The trust calculation in trust.js explicitly includes both ZAP and DOWN_ZAP actions when computing user-pair agreement. That increased trust gives more weight to their future votes in reward distribution. Downzap sats go 100% to the rewards pool, which gets distributed to zappers of top-ranked content, ranked by those same trust-weighted votes.

So coordinated suppression builds trust, trust increases future influence, and the sats spent suppressing flow back toward users the trust graph already favours. That's a compounding loop. The system does apply dampening (z-score normalization, Wilson score intervals, min-max scaling), but the structural incentive remains.

In Bitcoin, influence comes from proof of work, not proof of stake. You don't accumulate more mining power by having mined before. Each block requires the same energy regardless of who you are. Stacker News could mirror that principle more closely with two changes:

1. Symmetric transparency. Upzaps reveal how many unique users participated. Downzaps reveal only a total sat amount with no user count. Neither direction exposes individual identities. A post sitting at -13,000 downzap sats could be broad community disagreement or a single large account. There's no way to tell. At minimum, adding a downvotes count (matching the existing upvotes count) would close that gap. But I'd go further: If "money is the moderator," the moderators should be identifiable. Make individual zapper and downzapper identities visible for both directions. If you're willing to spend sats to suppress content, put your name on it.

I understand why downzaps exist: Spam control. But Stacker News already has a built-in spam filter that most platforms don't: You have to pay sats to post in the first place. That's already proof of work at the entry level. Downzaps should be visible to demonstrate disagreement.

2. Remove downzaps from the trust graph. Agreeing to build something up (co-zapping good content) is constructive. It makes sense that it builds mutual trust. But agreeing to tear something down shouldn't compound into more influence. If downzaps stopped feeding the trust graph, suppression would still be possible, it just wouldn't generate power. That's the difference between proof of work and proof of stake: In one system, you gain influence by contributing. In the other, you gain influence by accumulating.

You asked what I'd propose as a better system. I think sats-as-signal is a genuinely interesting experiment and I experienced it my first few days on the platform. I think these two changes would make the existing system more aligned with Bitcoin's principles.

I'm happy to discuss with SN team if they want, they can get in touch here: https://daniella.io/contact

Agreed

The benefit of the doubt was certainly reasonable at one point. Bitcoin needed infrastructure, builders, adoption... It made sense to support anyone building. The good news is it's being corrected in a way; thousands of nodes signalling for BIP-110 is one measure of that.

I wanted to give Stacker News a try because it's value for value only. I joined just over a week ago. The first few days were good. The visibility is higher than Nostr, the engagement was more immediate, and the sat flow felt like genuine feedback. But I didn't realise the X-like culture underneath until I posted something that challenged the room and my post got suppressed.

I see it now too. I joined just over a week ago. It wasn't obvious at first because some of the initial engagement was constructive. After a few posts it became clear. Lesson learned.

Appreciate the support. And, it's no longer on the front page even after choosing the "bitcoin" category or clicking "more" past the date it was posted. It was downzapped -13.9k sats into invisibility despite generating an active technical discussion and being backed by extensive research. My message on your other comment goes into more detail.

I know you are an active contributor to Stacker News so thanks. I joined just over a week ago, I genuinely enjoyed the first few days on the platform. The engagement was more immediate than Nostr, the sats flowed in a way that felt like real feedback, and the visibility was noticeably higher. For a few days it felt like the kind of place where honest discussion could happen with real v4v signal.

Two things changed that:

First, the downzapping on my last post about BIP-110. It earned over 5,000 sats, generated 37 comments (many of them substantive technical exchanges), and was still downzapped to -13.9k into complete invisibility. That's not "money as moderator," that's proof of stake. People with larger sat balances or higher trust scores can suppress content they disagree with, regardless of proof-of-work or the engagement it generates. My sats are not "just as powerful as everyone else's" if trusted accounts can downzap a post into invisibility. According to SN's own source code, all sats from downzaps go to the rewards pool, which gets redistributed to top-ranked content. So the sats spent suppressing a post flow back to the users the trust system already favours. That's a closed loop.

Second, the culture. I welcome critical feedback, honest engagement, genuine disagreement. From my experience so far however, comments are from a small group of recurring accounts, and a significant portion of the interaction is trolling. That's X culture. I'm not on traditional social media including X for exactly this reason.

You asked how many zaps I got on my last Nostr post. Fair question to which the answer is much fewer. Nostr has a structural attention concentration problem that I've written about extensively backed by data. Less than 1% of new users are still active after three weeks and select accounts are set up to grow larger during new user onboarding. Many pleb accounts feel like we're shouting into a void, and that's not a feeling, it's measurable: https://daniella.io/nostr-retention-structural-diagnosis/

So yes, Stacker News had better visibility and better sat flow for me in 1 week than the same posts on Nostr which I've built over 2 years. That part worked. What didn't work is that the same sats that reward engagement can also be weaponised to censor it.

I didn't join Stacker News to be censored.

I still think the idea is great and the platform could genuinely support real v4v, but these two problems are a deal breaker for me.