Just did some tweaking to my algo for measuring usage. Here's the top ten relays by zap activity.

It's crude but it takes the latest 100 zap receipt events and measures the time between the latest and earliest one and ranks by the difference between those numbers.

If there aren't 100 zap receipts they don't get a ranking for "usage"

Doing this in python is pretty terrible so I'm sure it's buggy and I'm open to suggestions on ways to do this better. It was at least a fun exercise in how bad I am mat async operations in python.

It just dawned on me that keeping media delivery separate is the key.

clients could incorporate external media delivery services like nostr.build, and those services would bear the brunt of dealing with illegal content. If people then used external links to bypass media delivery systems, this is something outside of the control of nostr, and outside of client developers responsibility. That is the purview of the Feds.

If then, a media delivery service got weird and started censoring for reasons outside of illegality, they could be replaced, or bypassed by posting external links, preventing full centralized censorship.

If a client starts moving past this, and centralizes censorship for arbitrary reasons, they can be replaced.

Larger relays will need AI filters. Your Blocklist idea isn't gonna work when I can generate a new id for each post. Smaller relays can moderate themselves just like the fediverse does. I'm not saying we can't make shared Blocklist but it's gonna do jack shit to stop illegal material.

I think the easiest first step is a Reddit-esque voting system, where rather the downvotes are reports of abuse/copyright/etc and when they hit a certain threshold, relays can have auto delete settings to block or delete. That way it’s user curated, but decentralized by relay and that operators desired settings.

daemon + freedom books by daniel suarez are a fun take on a potential nostr + bitcoin future

here’s the thing…

even if child sex trafficking wasn’t a multi billion dollar industry (which makes me want to go on a Rambo outing), the FBI would plant child porn on the network at some point. Or “terrorist communications” to justify whatever they wanted to do.

Nostr is a sitting duck until it solves for these things. Unfortunately, short sighted soy devs who also want to censor the world for their fefes, make up the majority of the “moderation” crowd. So thinking through how to deal with this in a new way has to happen.

Either easy way of dealing with this, censorship or removal of anonymity, are the things the corrupt State wants. And as soon as the door is open, the Feds will come in.

The problem comes from dealing with it using centralized power. So the question becomes how can it be dealt with on the individual level without changing relay’s simplicity, or making all clients censorship tools of a corrupt state, and groups of crying soy devs?

How does one censor content without censoring content?

It’s a real question…

The United States rolls over its debt on a scale of a few years, not every day. So the "run on the dollar" would play out over a year or two rather than overnight. Furthermore, I have described for clarity a sudden one-time loss of confidence. The actual process of running from the dollar, however, is likely to take more time, much as the European debt crisis has trundled along for more than a year. In addition, because prices tend to change relatively slowly, measured inflation can take a year or two to build up after a debt crisis.

Like all runs, this one would be unpredictable. After all, if people could predict that a run would happen tomorrow, then they would run today. Investors do not run when they see very bad news, but when they get the sense that everyone else is about to run. That's why there is often so little news sparking a crisis, why policymakers are likely to blame "speculators" or "contagion," why academic commentators blame "irrational" markets and "animal spirits," and why the Fed is likely to bemoan a mysterious "loss of anchoring" of "inflation expectations."

There's one thing that intrigued me, why do you think that at some point relays will be forced to moderate content?

Okay, looks like we've come to an agreement that it's not possible to stop any kind of content in #nostr

If the intention is to help with a cleaner feed for new users who don't want or can't for whatever reason be their own algorithm, we have to look at filter.nostr.wine, I use it to access global, it brings me notes of who I I follow and who they follow, great service that keeps my global feed clean. I always talk about this service to help those who are arriving. What we need is more options for this type of service. Not blocks, simply because it won't work.

Not our duty. There are agencies we pay to do that. Do you actively check people in the streets to see if they are talking about something illegal? Why would you do that here? All we have to do here is not interact with your “bad actors” and in exceptional cases notify the authorities.

I respect your point of view, but it dangerously adheres to the political doctrine that we all feel as a yoke.

I hope we never get censorship or any form of blocking at the relay level. I will be more than happy to pay for relays (wherever they may be) that DO NOT censor or block any content. All blocking, muting, censorship should be done at the client level. If the task becomes too heavy for a client-side app, we could have more beefy "personal" private relays that do that for us and our client-side app gets its data only from that personal relay (which would then rebroadcast our notes). Same goes for email : I choose to use servers that do not block any spam. I can do that for myself in my email client. I don't want someone else deciding for me.

All moderation/censorship lists should be private lists and I would stay away from any list from any of the current major social media players. If I wanted that censorship, I'd be on those platforms.

I would simply recommend that it be easy to share private lists between nostr users.

Kids-friendly moderation scares me if it's done by the likes of youtube. They actually *target* kids. I've seen it in action.

Small kids probably shouldn't be on social media, period. But if one wants them to be on nostr, I would like to have a personal, private, whitelist of people they can follow (family and family friends) and the client app limits what they can see to only what those few people share. Even then, I really don't think young developping minds should be putting out their every thought for all to see, for all eternity, with no possibility of taking any of it back.

There might be a case for only allowing a watch-only login (npub, no private key) for young kids. It might be a useful education tool which would provide different (and timely) topics of conversation with our kids.

Just my two sats.

Couple things

1. Agreed there is no way to stop the content. That’s not the point of the proposal, this is about helping users to get it off their feed at scale, using lists that they choose (or choose not to use any)

2. I really think there’s an order to this. Users should do this primarily. Relays may have to because of their legal liability and they should do it as little as possible. And I hope clients never have to use blocklists. I believe the primary responsibility is on users to choose the kind of content they want (and want filtered out). But relays will also be forced to content moderate at some point.

I just haven’t heard a proposed solution that will help relay operators to not get taken down. Nostr can not be mainstream unless we have many relays in many jurisdictions. But forcing a user-only moderation strategy will have consequences at scale.

As you said here...

"The beautiful part about Nostr is that if you want Wild West clients there is nothing stopping you from that."

There is no way to stop this type of content.

Censoring, in my opinion, is a futile effort. Education is the way.

In time, today there are already clients who do moderation, zbd and primal.

You sound like you want to be the arbiter of truth. Who will be deciding what is to be seen?

My thoughts: Legal content that isn’t abusively spamming the system should never be blocked by the system. But if individuals want to block someone… yes, they should be able to do that. As for setting up to auto block anyone that three of my follows/ftiends have reported - well that would suck because i often follow spinmeisters & other shills so I can expose their lies. And these bad players often gang together to get truthful & good content &accounts banned. So please don’t make it easy for them. On the contrary how about making an algo to temporarily take away reporting privileges when they’re obviously mass /spam reporting and/or they’ve reported X amount of content that turned out to not be in violation?

If you want to solve social problems by hiding them, you’re in for a bad surprise.