nicodemus

I think the war for individual users is already over for them.. just a matter of time before streaming expert models allow frontier-level AI on typical consumer hardware.
But for businesses... tougher call. They will definitely lose on the harness front. Businesses are already discovering the power of a custom agent tailored to their way of doing things. But the models? Things like Bedrock make it very attractive for them over managing their own shared models with rented GPUs.
But many businesses that I've talked to are really concerned about per-token costs. So we will see.

2026-09-30 22:56:08 GMT

A game of chicken.
I think OpenAI, Anthropic, etc hold more leverage than any given state. What do you think would happen if they said, "fine, we just won't serve your state."
The state would fold so fast...
Only problem is that closed-weight companies need federal stimulusand regulations to survive long term against open weight models. So they will likely comply.
#nevent1q…449g

2026-09-28 17:06:48 GMT

- reply

I keep waiting for marshal law to kick in. They just need a good enough narrative for it. Then tyranny is fully unleashed.

2026-09-25 23:44:55 GMT

- reply

Dynamic UI generation by agents. We're pretty close, I think.

2026-09-16 15:56:22 GMT

- reply

As always, the more money you have, the easier it is to make money.

2026-09-16 15:45:09 GMT

- reply

I dunno - the weak-minded and delusional?
It's all gonna burn. It has no value.

2026-09-06 01:16:04 GMT

- reply

Once it's packaged to HLS, it becomes an issue of how well the backing blossom server(s) can manage a 1MB (avg 1080p 10s segment size in h.264) download every 10s for however many users.
If HEVC is used, the server(s) should be able to handle even more.
So the difference between a 10s clip and a 100 hour live stream is simply a question of how many concurrent downloads can be supported by the served.
The client/player doesn't care, it typically only ever needs to hold 1 windows' worth of segments.

2026-09-05 19:24:39 GMT

- reply

Gotcha. If you get a chance, take a moment to run what I said above through an AI. You have a chance to do this on-device, forever free and sovereign and no 3rd party.. once its in HLS, there's all kinds of cool things that can be done. Like NostrTV.
Source: over a decade working on streaming video codec tech.

2026-09-05 19:15:52 GMT

- reply

Will you be transcoding on the device or somewhere external? Or segmenting as-is?
You could segment the video based on IDR frames, alone. This avoids a transcode. I think ffmpeg will do this.
The downside is that you can get some really long segments that players dislike. But, for the most part, I think it should work across players/devices. Plus, you can parallelize the uploads/downloads to take strain off of the device.

2026-09-05 17:01:36 GMT

- reply

Historically, miners are reactionary. I think they're just following momentum, right now.

2026-09-05 16:25:52 GMT

- reply

There is a bit of idolotry with bitcoin.
Bitcoin is a gift, and we should be thankful for it and be good stewards. But we shouldn't forget where it came from.
All good things originate from our Lord.
All that said, its been really comforting seeing the amount of faith among the people of sovereign-focused communities. I see more faith here than any other community.

2026-09-04 16:25:39 GMT

- reply

I probably still have it along with other comics from my youth in storage. Kinda miss the newsprint smell of old comic shops.

2026-09-04 16:18:10 GMT

- reply

I have an old copy from the early 90s as one big graphic novel, which is easier to come by.
Its super interesting. There's a whole notes section at the end that reads like an encyclopedia and predicts not just kinds of tech we now have today, but the evolution of that tech up to this point and beyond. For example, laptops -> mobile handhelds -> wearables -> subdermal devices -> wetware.
And this was written by one guy in the mid 80s.

2026-09-03 14:43:44 GMT

- reply

I know people want nostr to grow.. but I am pretty happy that its not a cash cow for influencers and scammers. It means the bulk of what you run into are real posts by real people giving really hot takes.
As influencers take over, my usage of nostr for already-limited social media intake will just evaporate.

2026-09-02 18:39:36 GMT

- reply

The way narcissism works is that they gaslight and gaslight until success. If caught, you get the fake apology followed by love bombing the accuser. Repeat until success or the mark leaves. If the mark leaves, find a new mark.
I put all influencers in this bucket by default. Few earn their way out. Those that do have my respect.
This has worked wonders.
Whats sad is watching everyone else get gaslit. Poor fools.
Even more infuriating when one manages to catch you off guard and make you the fool.

2026-09-01 00:18:01 GMT

- reply

Gresham's law completely lost on most folks.

2026-08-30 01:40:56 GMT

This is awesome.
#nevent1q…vqze

2026-08-29 19:31:32 GMT

- reply

Don't?

2026-08-28 21:59:16 GMT

- reply

Thats one attack vector. There's also Embrace Extend Extinguish, which has been leveraged by suits against FOSS almost as long as its existed.
Omarchy has smells all over it.

2026-08-26 14:53:01 GMT

- reply

OH, LOL. I totally misunderstood. Good luck, then, sir!

2026-08-26 03:02:21 GMT

- reply

Cracked, sure. That's not the same as spending someone else's money. I was just asking about your intentions if you found one.

2026-08-26 02:59:51 GMT

- reply

Why does this domain show up on malware lists (Dandelion Sprout's Anti-malware list)?

2026-08-26 02:36:13 GMT

- reply

Crawling for wallets? If you found a key that belongs to a utxo, would you spend it?

2026-08-23 23:06:13 GMT

- reply

You said it. The only relief we have.

2026-08-17 16:53:22 GMT

- reply

Exactly. Thats been my take on the entire performance in Iran. Definitely the endgame for the US empire. Dollar can outlive the US, but not USTs.

2026-08-17 16:45:08 GMT

- reply

I don't get the "80% abstained" argument. Wouldn't "nay" votes take the form of sticking with the status quo and doing nothing? Isn't that the path of least resistance? You either signal for 110, or you don't. Or, is the argument all based on sentiment and not actual signaling?

2026-08-16 19:35:46 GMT

- reply

Totally irrelevant. Also, incorrect. Crime rates have fallen YoY, at least since 2024. A simple city data search confirms that empiraclly.
But even so, you're of the belief that mass surveillance will magically fix the crime rate, eh? Likely not. We haven't seen the crime rates meaningfully.
In fact, the data is so mixed that, even where flock cameras were used directly in investigations, all that could be show statistically is that there was a directional consistency with aid (ie, surveilled area vs improvements in investigation leads) (see:
https://link.springer.com/article/10.1007/s41887-025-00099-y and https://www.researchwithrowan.com/en/publications/an-evaluation-of-a-major-expansion-in-automated-license-plate-rea/).
There is so little evidence that even the costs are being questioned, ethics be damned (see above springer link and https://www.icpsr.umich.edu/web/NACJD/studies/37049).
I don't know how you can make any kind of claim of ALPR systems, in general, being a benefit to crime reduction let alone prevention. When even the municipality leaders are having to question the cost/benefits (likely after wining and dining from companies like flock), and the best case they can possibly make is "it has helped recover stolen cars", your argument lacks any kind of footing.
There is no signal here at all, only noise.
Meanwhile, the continuous encroachment on civil liberties continue with these kinds of rollout. All in the name of money.
We are just going to have to disagree. I am not interested in trying to change your mind so much as demonstrate to others that your argument is completely without merit.
And, worse, it signals how little you understand the historical precedence that this kind of surveillance leads to.

2026-08-15 16:16:13 GMT

- reply

Who said I believe that these cameras are a deterrent? I don't believe that! I believe they are used to surveil all people. And there is plenty of anecdotal evidence, since you wait to include that, to support the claim that mass distribution of cameras lead to mass surveillance of innocent people and loss of privacy and agency.
Regarding your studies, follow Tue money on them. They are sponsored or ran by flock. These cameras aren't deployed with the good graces of the people's will in those jurisdictions - instead, it smells more like traditional greased palms. You need only to search of the uproar these cameras have caused and, in some cases, removed by local governments after the will of the people is actually made known. Austin, TX is a beautiful example of all of this.

2026-08-14 21:15:13 GMT

- reply

So you believe these flock cameras to be an effective deterrent to crime? If so, do you have any evidence to support this?

2026-08-14 17:18:11 GMT

- reply

I hear you. But that's not what people think of when they read "alternative to flock".
You're asking a bigger question. One that flock doesn't even really address any better than a neighborhood watch does. The data flock gives is extremely cherry-picked and manipulated. Consider the scale of their rollout. If it was really effective at all, you'd be seeing crime dropping throughout all participating jurisdictions. But we don't.. I haven't seen one ounce of data that starts to suggest any kind of efficacy.
So, you see, flock is a farce. It is surveillance masked as crime prevention. It's up there with "for the children".
So there is no viable alternative. It was never about crime.

2026-08-14 16:32:50 GMT

- reply

There is no "viable" alternative to mass surveillance. This is a scourge that needs to be excised from "public safety".

2026-08-13 03:14:55 GMT

- reply

Hodlnaut still has my respect. His heart seems to be in the right place.
I think a lot of people will come around when they realize they just got wrapped up in the drama.

2026-08-12 23:34:45 GMT

- reply

Total god complex.

2026-08-12 16:11:54 GMT

- reply

Yes. And lose all hashrate just as fast as miners go elsewhere. You make it sound like they're stuck on that pool.
The moment a pool is captured it becomes a threat of multiple kinds, to all parties involved. Game theory / incentives kick in even harder, then.
The flip side is real, too. Pools enable distributed coordination. This cuts both ways and, I believe, will balance itself out.

2026-08-12 15:48:19 GMT

- reply

Exactly. And one can see how easily those miners in the pool redistributed when China had it's last big ban, and miners fled the country.
When you zoom out, I see mining becoming more and more decentralized, not centralized. Home mining was barely a thing in the last cycle. Now it is not just reachable but its getting to a point where any consumer can just plug it in and walk away,
If people were that worried about mining, they'd push for DATUM, home mining, and simplification of nodes (more appliances).
All of this rhetoric is just rationalizing their drama.

2026-08-09 01:15:54 GMT

- reply

I should add that I've never been more petrified for all bitcoiners. The above comment is not stated out of arrogance, but reality. It's scary as hell, and I imagine we will see a lot of divestment.

2026-08-09 01:12:29 GMT

- reply

Written by someone who is still unwilling to take accountability for their own custody.
The coldcard catastrophe is all on coinkite. But the accountability of each and every person's coin is on them and no one else. Thats me and you.
People can and should be pissed at coinkite. But they should also be pissed at their own ignorance.
Bitcoin is brutal to own. That's my biggest takeaway. As much care as I've put in, I need to put in even more.
We don't get to own the most pristine bearer instrument ever created by man without the responsibility that comes with it.

2026-08-09 00:54:24 GMT

- reply

Peace.

2026-08-09 00:45:24 GMT

- reply

Awesome observation.
The masses are only reactive, too. Consider, instead, the scale of the catalyst. We are talking worse than COVID lockdowns.
I'm thinking something along the lines of massive bail-ins or CBDCs evaporating two weeks or a month of people's wages as "punishment" for something like disagreeing, socially, with a government policy.
The alternative is a slow creep until competition kicks in at a national level for people's time and effort (assuming people can still move around somewhat freely).
Or another alternative is that there is no revolution. Bitcoin adoption only goes so far. And so citadels become much more of a thing. - a concentration of wealth.
"We are still early" cuts both ways.
I do think bitcoin usage for international trade is inevitable. Simple war will make that happen. - it's already started.

2026-08-07 00:29:44 GMT

- reply

Of course, I wasn't implying otherwise. It read to me like an appeal for the one product, when there are other products (and non-products) that already do what you ask for - and its wise to support those doing the right thing.
But if your aim is to just chat with jack, by all means. Mea culpa.

2026-08-06 20:43:42 GMT

- reply

I mean... you can just not support devices that are closed and let them figure out why...

2026-08-06 18:09:26 GMT

This.
#nevent1q…yrsa

2026-08-06 17:59:16 GMT

- reply

Right. The hard part of every cycle's bear market victims is accountability. Bitcoin is the ultimate exercise of property accountability. It's hard, and it's so easy to miss what you don't know. You really, really have to put in the effort to know what you own.

2026-08-06 14:52:57 GMT

- reply

No. And I made the same comparison to the wife. Every cycle, we have a lot of people lose a lot of coin due to misplaced trust. You can't trust much of anything in bitcoin. Even yourself. People gotta learn to think in layers of security.
This one DOES hit a little different, though, because a lot of people *thought* they were doing the right thing, following best practices. These people just lost out of ignorance, not greed or gambling.

2026-08-06 14:49:08 GMT

- reply

Just use an old computer that's been air-gapped. Sparrow wallet is great, open source, and well reviewed. Look into using TailsOS (runs off of a USB, does not store data) for increased security.
Store your seed(s) on a steel plate.
If multisig is your bag, store your xpubs + derivation path in an encrypted file somewhere in the cloud (password manager note, or Dropbox, or icloud, etc). Sparrow just hands you a single PDF, I believe. Even if someone stole the file AND managed to decrypt it, the most they get is your address space.
This setup leaves the computer at rest with nothing on it, your seeds are backed up (and can be distributed for added security). You can still create PSBTs to sign and send transactions. You don't need a 3rd party multsig provider.
Seeds can be hand-rolled. In fact, a lot of things can be done on paper, which is a good exercise to just learn what happens by the software.
Snag some testnet coins and try it out. The above will teach you more about bitcoin in one or two afternoons than most people will ever learn. And its not hard.

2026-08-03 19:23:27 GMT

- reply

Ok, where to begin?

2026-08-03 15:09:43 GMT

- reply

This is a solid analogy. And i can see an opt-in standard forming. Then let incentives do the work.

2026-08-02 16:14:50 GMT

- reply

Just roll the dice, look up the values in a chart, and just KNOW. Why cut corners?
Unless you're just experimenting, go the distance. It's your future that you're taking shortcuts against.

2026-08-02 16:13:00 GMT

- reply

Though I don't own one, the BitBox2 is fully open, I believe.
And an interesting new player is FrostSnap.

2026-07-30 14:46:22 GMT

- reply

I can't help but draw some parallels to what's been going on in Japan over the last 20 years. Men acting like boys, women like little girls, the love of childish things... and then look at marriage rates, working hours, etc. I suspect something systemic, and maybe related to consumerism.

2026-07-29 23:03:09 GMT

- reply

I agree. And I don't think people are asking for it, per se, but they sure are loving consuming it.
As a kid, I couldn't wait to be an adult and have freedom at the expense of responsibility. I yearned to escape control, even if it meant full accountability. And its been everything I hoped it would be. And I want more of it, not less.
The masses that consume this stuff are akin to the Hebrews selling themselves into slavery, only this famine was intentionally created by the state for this very purpose.

2026-07-28 15:00:28 GMT

- reply

One of my children going through the same, basically living on my couch and needing help for some basic movements. Its a tough recovery, amigo. Grumbling is very merited.

2026-07-28 14:51:48 GMT

- reply

Holy crap!

2026-07-25 17:17:26 GMT

- reply

Some optimism: this will further incentivize more and more implementations of the protocol.
It needs to be like Linux - so many implementations that it makes you question "why".

2026-07-23 22:56:59 GMT

- reply

Lol. They can *try* to enforce it...

2026-07-20 18:38:50 GMT

I don't get concerned by this, at all.
It paves the way for more amateur miners and home miners, both furthering decentralization.
Then, the AI road ahead of us is not paved with giant warehouses of compute. It's composed of smarter MoE models that can run on consumer grade hardware. Costs and security incentives guarantee this. This means these mining corps will either pivot back or die. Either way, bitcoin continues on.
#nevent1q…yjts

2026-07-18 15:48:33 GMT

- reply

Supposedly origin of the term "computer worm". The tech still operates on phones, but its pretty close to where we're headed.
Check it out.

2026-07-14 22:18:52 GMT

- reply

An oldie but goodie. Ever read The Shockwave Rider?

2026-07-08 06:03:18 GMT

- reply

Precisely.
Now do the math on the size of boomers vs other generations. Then add in the average number of homes they own.
Housing is going to be a horrible “investment”, even for the normies.

2026-07-04 19:33:40 GMT

- reply

Yes. And they’re all terrible. If you just want simple, markdown notes, you’re good. If you want it to require hours of setup and plugins to make it decent, there’s a couple there, too.
Otherwise, they’re all mostly doing the same simple things with poor UX.
That’s why the better ones require some kind of subscription. They know the free ones leave a lot to be desired.

2026-07-02 21:15:03 GMT

- reply

Same. Same.

2026-06-24 06:53:17 GMT

- reply

My kingdom for a good FOSS maps and simple, but featured notes app. There’s a ton of little stuff that Apple really gets right.

2026-06-24 06:50:44 GMT

- reply

…but you don’t? I didn’t. Got an eSIM to work on a raw unlocked phone flashed with a fresh install. I think it depends on the carrier more than graphene.

2026-06-24 06:42:39 GMT

- reply

I guess I’m in the minority. I’ve found it very easy to install (though the boot loader stuff is always a bit of fumbling on each android device).
The only thing I wish was smoother and clearer is how to install google play services from certain perspectives. IE, if you want another profile to keep things separate, and need something like a bank app running, here’s what you do…
Otherwise, FOSS all the way. Easy to use. There’s perhaps too many options and it’s easy to lose a bunch of time discovering which app stack you need to replace your old. Especially if coming from iOS.

2026-06-20 15:40:14 GMT

- reply

Congrats and well done. Keep putting out good content. Keep skewing further and further towards normies - they need the basics. Cheers!

2026-06-11 05:17:41 GMT

- reply

It’s tiny sick syndrome. Literally nothing to see here. We should all just move on and be happy.

2026-06-09 05:04:14 GMT

- reply

So you think selling 32 btc is enough to move markets? Or that the narrative was enough to move markets?
Thus far, it’s a typical bear market (timing wise) and a pretty mild one (price wise). I don’t think the reasons even matter.

2026-06-07 17:44:21 GMT

- reply

Yep! It’s all unsustainable. But the genie is out of the lamp, now. People won’t go back. So this will drive innovation on smaller models that can dynamically load in more or less experts, allowing frontier-quality results on consumer grade hardware. At the same time, unified memory architecture will become more prevalent (like Apple’s machines and AMD’s Halo Strix line).
All of this unsustainable bullshit will just create great incentive for decentralization.
When AI is open and decentralized… well…it’s like handing encryption to the masses. There will be mass empowerment. Most won’t see it or know it, but it will be there.

2026-06-06 05:53:58 GMT

- reply

The 200w MA is the only real outlier. Everything else is pretty much a typical bear. We could have another 3-4 months of this.
I suspect we could go sideways, riding the 200w for 2-3 months, have a quick capitulation way below it followed by a near-instant recovery sometime in September. Something like that would be pretty typical bitcoin behavior and not really outlandish.
Having the bottom be now would be rather exceptional, timing-wise and drawdown-wise. So, for the time being, I’m thinking the former is more likely.

2026-06-06 00:55:28 GMT

- reply

Nice!

2026-06-06 00:55:00 GMT

- reply

All of my recipes (and maybe images, too) into a common format - or at least markdown and CSVs.
I’m hesitant to push data outside of my self-hosted environment.

2026-06-06 00:50:02 GMT

58k gang - NOW IS YOUR TIME. THIS IS YOUR MOMENT.
…until we go a little lower…
..and then back up…
…and then THIS IS YOUR SWAN SONG…
…and then it’s back to making Peter Schiff vomit.

2026-06-04 18:04:09 GMT

- reply

Thank you for being another voice of reason.
So far, this bear market is behaving exactly like all others.
Targets remain around $46k. Only thing that gives me pause is the 200w MA.

2026-05-27 15:24:15 GMT

- reply

Just get a mikrotik and quit fucking around. It’s basically a little OC with hardware acceleration and runs linux.

2026-05-20 18:29:52 GMT

Are we still winning, those of you who aren't apolitical (yet)?
#nevent1q…vpmf

2026-05-19 15:01:42 GMT

- reply

This almost feels like the beginnings of an Embrace Extend Extinguish operation.
Whatever, nostr will be fine. Maybe more fragmented, but fine.

2026-04-21 04:49:35 GMT

- reply

Any chance to import from Paprika? How about export support?

2026-04-14 17:19:10 GMT

- reply

Tell us about your rig running the model. Hardware and software stack. I saw that you're running vLLM?

2026-04-11 21:17:45 GMT

- reply

A right of passage.

2026-04-06 19:22:03 GMT

- reply

So one of those is enough to get you started. It is well supported by AMD and there's even guides out there for how to ccluster 4 of them together (definitely ad a later phase).
Stay away from Mac minis. Its a good toy, but you lose a good bit of memory to osx and your limited in config options. If you want a large-ish model, you're forced to cluster and that opens up a whole other can of worms.

2026-04-06 19:18:20 GMT

- reply

Agreed - 128GB is the only way to go. Running a 72B Q4 is definitely doable while still allowing a decent amount of headroom for context/kV cache.
Recommend checking out the latest gemma 4 offerings. You can get a lot done with the E4EB model handling tooling, routing, compaction, and other tasks. The 31B is also great for better reasoning.
I would NOT use this machine for anything besides inference. Save all memory for context (target 128k tokens). I really meant it when I said to treat it like an "inference appliance".
Offload everything else to whatever you have laying around, including openclaw. Keep it separate so you have a stable substrate.

2026-04-06 15:03:46 GMT

- reply

What’s your budget? Ryzen AI Max+ 395 APUs offer UMA, which you’ll need to a decent model. I like Framework’s desktop offering. A bit more expensive than some chinesium builds, but you’re going to get solid firmware and driver support in linux - and that is king.
Set 1 or more up as an inference “appliance” and that’s all it does. Have everything else run on a different machine.
Stick with Ubuntu Server to start with - just easier support. Go ROCm + llama.cpp first, then fall back to vulkan if there’s issues. Can go Ollama when things are looking good.
I aim to build a Nix port once it’s all stable, making rebuilds of these “appliances” simple.

2026-03-20 19:23:46 GMT

- reply

Testes. Testes. 1, 2…3?

2026-03-20 02:33:54 GMT

- reply

Its called turning 40.

2026-03-18 17:16:20 GMT

- reply

When you’ve been an underfed mule for most of your life, I can understand how the pig’s life looks appealing.
I can’t fault people for being too tired to see the bigger picture.

2026-03-18 17:07:51 GMT

- reply

They don’t even know enough to realize that there are differences. To them, SQL is SQL, and relational databases just implement SQL. And that’s it. So it doesn’t matter, probably, except for highly specific scenarios.
They see the lower level stuff as a place for specialized optimization. They don’t see it as a foundation.
I see the same thing in all disciplines. The old ways are just that - old - and maybe are important on the edge. Otherwise, it’s not needed to learn the history of your craft.
Meanwhile I find I’m drifting further and further back in time in terms of seeking out new knowledge. And that’s because I see a lot of the conveniences of today becoming a hindrance for tomorrow in a world where money has value and so quality is king.

2026-03-18 15:26:14 GMT

- reply

I run into that everywhere. Even in software. The abstraction levels are making it to where the “kids” with 10 years of experience have no idea what makes, for example, one database different from another. This leads to grossly ignorant decisions that end up costing a lot later.
I’m finding it challenging as an old fuck with lots of experience trying to guide them toward the curiosity needed to be better.
But the fiat mindedness - that’s hard to break.
Reality will kick in soon enough.

2026-03-18 05:41:24 GMT

- reply

I mean, most are somehow in tech or tech-adjacent fields. There’s a lot of commonalities you could find.
The problem is that someone who’s labeled INTJ today may end up being an ESFP tomorrow. That’s the issue.

2026-03-18 05:28:51 GMT

Awww yeah!
#nevent1q…rnln

2026-03-17 15:10:43 GMT

- reply

Meyers-Briggs is all marketing. Huge foundation behind it. No one in psychology takes it seriously. OCEAN has actual studies backing up the claims and has no foundation.
Otherwise, right there with you.

2026-03-14 20:27:13 GMT

- reply

I do t think Proton ever said they wouldn’t work with their (Swiss) govt.
You don’t use Proton if your threat model involves hiding from nation states. In fact, you don’t use *email* if that’s the case.
You DO use Proton to prevent your data from being *sold*. Same with Tuta and other paid providers.

2026-03-14 20:24:49 GMT

- reply

But.. then I can’t make blanket comments and dunk on something and tell myself “I’m important”?!

2026-03-14 19:00:08 GMT

- reply

I would love this. Those that post a lot tend to have less meaningful things to say. It’s either shitposting (which can have its draw) or just blatant attention seeking.
This would be independent of replies, right?

2026-03-14 17:34:13 GMT

- reply

Librewolf is a fork of Firefox with all of this built-in and a very hardened set of defaults. It also strips out all of the Firefox bullshit like Pocket and other “features”. Highly recommended for a typical browser.

2026-03-14 17:30:12 GMT

- reply

Yes. It limits their creativity. This leads to less novel solutions. It also impedes learning. The more variety of models in your head, the easier it is to learn as you find similar bits of models to help speed up building a new model.

2026-03-12 03:13:08 GMT

- reply

Solid points. Thank you for the rational thinking!

2026-03-11 02:42:57 GMT

- reply

How am I outing myself as anything by questioning you? By suggesting you’re advocating for centralized algorithms?
Well let’s think through this…
Algorithms curate content for unique individuals. This means they need lots of data on the user in order to do so.
Rossman is advocating for no algorithms. You don’t like this. You go so far as to call this “sick”. Therefore, you clearly *want* algorithms.
If the algorithm is decentralized, then that means it’s distributed. So everyone’s individual data needs to distributed so that an algorithm can come to a consensus on what to show each individual..? This is insane. And retarded.
No one is going to want to process other people’s data. There’s no incentive. Not when the alternative is the algorithm running on the individual’s client. And that’s pretty much entirely within everyone’s control on nostr, YouTube, and any other tool that allows you to search, view tags, view followers, mute what you don’t like, etc.
So given that this is solved, and you’re still bitching, I assumed you were advocating for a centralized algorithm - the alternative to a distributed one.
So if you’re not, mea culpa - you’re just insane and retarded. My bad.

2026-03-09 23:25:55 GMT

- reply

Exactly.
Most people don’t understand the first thing about security OR they don’t want to take personal responsibility.
…or they just want to dunk on someone for fake internet beauty pageant points.