I am skeptical too. But I know clients won't pay. So, we will need to market these for users directly, clearly demonstrating why their WoT graph is better than any other algo in Nostr.
To me it always boils down to interests and the ability for users to filter their own graph (of users and events) weighted by their specific interests.
Language, for instance, is a big one. Posts (or users) in my language should be weighted higher than in any other language.
Why do you think clients won't pay? Traditionally apps are made by companies and those companies pay for the services they use.
What makes you say nostr is different in this regard?
Well, I have a client and won't pay. It doesn't make sense for me to pay for any service on behalf of my users.
On Nostr, most clients are too small and will never have the coding structure AND the regulatory structure to make passthroughs of cash. Damus and Primal are the only ones that can do it today. All the other 200 clients can't.
Lots of other service providers wanted Clients to pay. This is not a new idea. It just has never worked. Open-source is just not good for that.
We agree with the fact that Nostr is too small right now. But email was small too, but the it grew and I imagine gmail and co. use some paid services or host then themselves internally.
It's so much easier for a user to relate with Amethyst, the client they use, rather than with a service provider they don't know it exists, don't know or understand what it does. Or rather, it does things that are so fundamental they never had to pay for them.
I wasn't saying that Nostr is small. I think it is a good size for any service provider to play the game. But the apps themselves are way too small for payment integrations. And they will always be small.
Apps on Nostr are like Watch faces. They don't have the time to build the supporting infrastructure to redirect cash.
On the second point, yes, people trust Amethyst more than a service they never heard about, but that is fixed when Amethyst recommends services in a settings page. We pass the trust we have to providers. If we just code a WoT settings screen where you are one of the providers, users will click on it and create an account with you. The better the integration, the better the capture.
That's also why I think we need to have WoT-visualization clients whose only job is to "sell" how web of trust is better than everything else. Then users just need to pick a provider and use in all apps.
> “I am skeptical whether the end user would actually pay for such low-level primitives that every social media app has built-in”
nostr:npub176p7sup477k5738qhxx0hk2n0cty2k5je5uvalzvkvwmw4tltmeqw7vgup
> “Well, I have a client and won't pay. It doesn't make sense for me to pay for any service on behalf of my users.”
nostr:npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z
I think you’re both missing the WoT value prop … but in different ways.
Users will pay to access WoT powered recommendations and discovery for the same reason that they pay for AI credits … but don’t pay for Google.
Clients will pay for WoT services (on behalf of their users) NOT by representing the user’s npub directly but by providing their own npub’s (value added?) take on “content and user curation” … for users who DONT wish to “bring their own WoT provider” (or don’t have one).
Why? Because FULLY realized WoT powered content and user discovery WILL BLOW PEOPLES MINDS … and everybody will want a piece of it.
I mean … that’s just my optimistic take.
Traditionally clients have the money to pay for services because they’re monetizing our data and raking in billions of dollars. Our goal is to destroy this business model and replace it with a new one. If anyone monetizes our data, it will be the end user, not some big tech platform.
And WoT will be how we monetize our data. If you want to leave a review for a really good restaurant, the fiat way is you leave your review on Yelp for free and they monetize your review. The way we’re going to pioneer is that you have the option to give your review away for free, but an even better option is that you sell your review for a few sats. And I’m going to be willing to pay you a few sats for your review because my WoT tells me that you’re probably not a bot, you’re probably a person, you probably live in the same town as the restaurant, you probably meet whatever requirements I decide are important to me.
idk if our #wotathon will make it this far, but one of the applications we envision for our Decentralized Lists NIP is that I can start a new decentralized list of something I want — example: best steak restaurants in some town I plan to visit next week. To do this, I publish a kind 9998 event, per the NIP, and it includes a description with whatever details I desire. I then place a bounty of, say, 5000 sats, with automatic payout of 500 sats to anyone in my WoT who submits an item to this list (kind 9999 event) until the bounty runs out. High enough bounty means your desired list gets populated quickly.
How will the bounty be managed? We’re not sure. It might require some sophisticated middleware, like maybe nostr:npub1q6mcr8tlr3l4gus3sfnw6772s7zae6hqncmw5wj27ejud5wcxf7q0nx7d5‘s #safebox, but we haven’t thought it all the way through. We are, however, confident that a tractable solution exists and someone will build it. Can someone build it in the next 6 months, in time for the hackathon? Decentralized Lists with a secure bounty system? If someone does it, and does it well, it will be a contender for the win. We will see!
All of the above is covered in this article:
nostr:naddr1qqnxgetrv4h8gunpd35h5ety943h2unpw35k7m3ddanz6umfd4cxcefdd35hxarnqythwumn8ghj7erpwe5kgtnwdaehgu339e3k7mf0qgsw2feday2t6vqh2hzrnwywd9v6g0yayejgx8cf83g7n3ue594pqtcrqsqqqa28klas25
This is a great line of thought to come back to.
Pip and Vitor are right to question whether users will pay for personalized trust metrics. The initial WoT use case is to get rid of the super obvious bad actors, and for this use case, most users will be content to use trust metrics personalized to someone else: an influencer, one of their friends, etc. At this stage, “personalization” of trust metrics would mean adjusting the parameters to case a wider or a narrower net. Brainstorm already does this. But will users pay for that? Maybe a few, but it’s not a must-have.
Once the bots are gone, we turn our attention to contextual trust. This is where personalization gets more interesting. Suppose Alice is interested in sports, Bob in the latest developments in AI. They’ll want their personalized Brainstorm to keep track of everything having to do with that field — including an ever-changing ontology — and to stay up to date, which will be a complicated matter. Will users be willing to pay for this? I think some will. But what if I can piggyback on my friend’s personalized WoT for free for topics like these? Might still be a tricky sell.
And then we come to the next level of complexity: privacy. This might be the hook that starts to bring in a lot of users. Maybe I want personalization of interests, AND privacy regarding who/what I’m following and why, AND I want *inviolable control over the curation* because otherwise I won’t trust the slop content being fed to me.
The trick is going to be to make personalization a must-have. And not for ideologues, but for normies. A combo of personalization of interests, inviolable control that keeps out the slop, and privacy may be required to make personalized trust a must have. Which is not an intractable problem, but it will take a lot of work.
And how can I forget: personalized WoT turns the legacy monetization model on its head. No longer will you give your data away to big tech for free. You can give it away if you want, OR you can sell it for micropayments. I’ll be willing to pay you a few says for your restaurant review if my grapevine tells me with a high degree of confidence that you’re not a bot, if your previous restaurants have been helpful and in good faith, etc.
But I’m not going to pay anyone for anything unless I am ***VERY*** confident in the information my WoT is feeding to me. And I probably won’t have the requisite confidence without the features in my previous note: personalization of preferences, privacy, and provably inviolable control over the trust calculations. Brainstorm will provide all of these.
