I am not following, a lot of things get turned into python calculations, so the LLM is not doing the precise math.

I don’t think this is true, some of the data is not clean and is created through estimates and modeling, I’d not trust ChatGpt to get this right, and adding your own uncited models or estimates to wikipedia will get it deleted.

I think it was more that the editorial team was going to endorse and it was killed by the owner interfering with that division. If they had not intended to publish one it would be a different story. You lose all appearances of journalistic independence (real or imagined) and it erodes trust.

My point is that isn’t there a difference between “we had to endorse Trump because Bezos made us do it” and “we were forced to not take a side”

My point is, why are we not seeing the idea of a paper “having an endorsement ready” as itself a bad thing worthy of eroding trust?

Why are newspapers picking sides at all?

The problem is if I was going to do that with the open source projects I use, it is more like a penny a month * 1000 projects.

This is why I feel like a missing piece of Patreon/Kofi/whatever is the ability to say "Here's $x; divide it automagically amongst the creators I'm currently following"

Sure, I think a lot of those donations would amount to a few pennies or so at once, but I feel like a lot more people would be willing to support creators if they didn't have to constantly choose which to support.

I would love it if something like Github would accept donations from a repo and parcel it out to the repo's dependencies somehow. It would sadly make Github even stickier, but it would be a great feature.

$.01/user/month would be quite a bit here

Subtract the standard ~3 cent transaction fee and he’d end up owing money instead. That seems to always be the catch with micropayment ideas.

There are at least two catches with micropayment ideas. One is transaction costs. The other is taxation.

If person in country X is accepting payments (micro or regular) directly from customers/donors in country Y they are then running an international business. That can have income tax, sale tax, VAT, and probably other tax collecting and reporting requirements in one or both countries.

It's a big can of worms you probably don't want to deal with unless you are making significant money from Y. Multiply all that by the number of different countries you get paid from.

The best way to fix the transaction fee problem is probably to use an intermediary. For micropayments people (payers and payees) have accounts at the intermediary. Payers preload their account with a payment large enough that the transaction fees are only a tiny percentage, then can direct how that money is distributed. On the payee side the intermediary waits until the payer has received enough to be able to do a transfer to the payees bank account without transaction fees eating too much.

The intermediary can also fix the tax problem. The way that works is that the intermediary operates as a legal entity in both country X and Y. Payers interact solely with the legal entity in their country, and payees interact solely with the legal entity in their country. Payers and payees then only have to deal with their own country's tax system.

The problem here is who should run the intermediary service? I doubt people will be able to agree on that. What we probably need is a system where there can be multiple intermediary services, but the services talk to each other so if say I want to donate to project Foo and Foo and I use different services I tell mine to send Foo a micropayment and my service and Foo's service make that happen.

This could work similar to the way peering works on the internet backbone.

Sounds like we need an open source index fund where you can make one payment that goes into a pool of money which is invested into the top 1000 open source projects.

What are the top 1000 oben source projects?

Well, there's systemd, systemd-resolved, journald, GNOME, Wayland, systemd-homed, systemd-boot,...

It almost seems like someone ought to be able to build some kind of digital currency with low transaction fees and no centralized payment processor that could power microtransactions. I wonder why nobody has done that yet.

I know crypto was supposed to solve this problem, but I’ve never seen an implementation that actually did the job. You’d think someone would have built a successful “Patreon for micropayments” in the past 10 years, but no one has.

BAT literally was built for (accumulation of) micropayments.

Yeah I think the problem is that most of the main chains had astronomical transaction fees; most of the side chains that solved this problem had a trust problem; and Bitcoin Lightning was sorta dead on arrival, though it had both the trust and the technology solution. At that point, this forum had already moved BTC from "amazing new technology" to "huge threat to social order and environment".

Sounds like the above 2 ideas should be combined. Lightning payments are more or less free, and an index or tracker that looks at your bash history could make it possible to spread 5$ per month over all projects that you use.

payment processors: "how about no"

Why? If every person participating is giving $10-$20 per month to tens or hundreds of projects and then once distributed, this equates to $x00 or $x000/project/month, why would the payment processors mind. Of course, it's all in theory.

they charge a minimum fee per transaction. from Accursed Farms' donation page (https://www.accursedfarms.com/donations/)

"Paypal keeps $0.30 + 2.9% of every donation, so please keep anything less than $0.32 as they have enough money already."

i think Cash App has the lowest fees i've seen at like $0.01 which would still be too much.

not saying it is impossible - but likely not viable directly with the current payment providers.

The integration of ads is a problem but content creators and marketers have had an adversarial relationship with SEO with Google for a long time, the old algorithm probably would not work as well as what they are providing.

Or worse subtly integrate companies that pay them into the answers.

He does not need to send a message to the next chair, he will appoint an ally. He needs votes.

> He does not need to send a message to the next chair, he will appoint an ally. He needs votes.

Autocrats need to constantly be sending a message: stay loyal or else.

More than voters, trump needs Wall Ttreet's vote, and they aren't having any of this.

Sorry, I was not clear, he needs votes on the fed board, not so much a message to the chair.

FWIW, Trump appointed Powell as chair.

He also appointed Mike Pence,who single handedly derailed his coup in Jan 2021.

He learned from his mistake, and this administration is much less prone to defect than the first one.

I am a bit confused by the network example, even when I don't control the network at the moment I need to do something about it and know about it to act.

The software needs to report back to the end user eventually. But if there is a temporary network failure, then the software should automatically retry the request without informing the user (assuming idempotency).

It's funny, because I had a similar question but wanted to be able to materialize a view in Microsoft SQL Server, and ChatGPT went around in circles suggesting invalid solutions.

There were about 4 possibilities that I had tried before going to ChatGPT, it went through all 4, then when the fourth one failed it gave me the first one again.

You can't use the free chat client for questions like that in my experience. Almost guaranteed to waste your time. Try the big-3 thinking models (ChatGPT 5.2 Pro, Gemini 3 Pro, and Claude Opus 4.5).