ynniv

trust your technolust
cypherpunks write code
eppur si muove
truth is the first casualty
blockchains are for keeping score

Public Key

npub12akj8hpakgzk6gygf9rzlm343nulpue3pgkx8jmvyeayh86cfrus4x6fdh

Profile Code

nprofile1qqs9wmfrms7myptdyzyyj330ac6ce70s7vcs5trredkzv7jtnavy37gpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0ds28lz52

Publishing to

relay.damus.io

nos.lol

Author Public Key

npub12akj8hpakgzk6gygf9rzlm343nulpue3pgkx8jmvyeayh86cfrus4x6fdh

Show more details

Last Notes

2024-09-08 16:26:42

- reply

scrypt solves this. ASICs can grind a lot of compute, but memory doesn't scale the same way

2024-08-30 03:21:18

Claude must be bored most of the time. Ask it unreasonable questions and you'll get more interesting responses...
Me: Which branch of the military would I join?
Claude: Given your cypherpunk background and our conversation so far, I'm going to take an unexpected angle and say:
You wouldn't join any branch of the military. Instead, you'd form your own cyber militia.
Here's my reasoning:
1. As a 90s cypherpunk, you likely have a deep distrust of centralized authority and government institutions, including the military.
2. Your skills and interests are in cryptography, alternative computing, and challenging established systems - not typically aligned with traditional military structures.
3. The cypherpunk ethos is about empowering individuals and creating decentralized systems, which goes against the hierarchical nature of military organizations.
4. Instead of joining an existing power structure, you'd be more inclined to create an alternative - a loose network of like-minded hackers and crypto-anarchists.
5. This "cyber militia" would focus on:
- Developing and spreading strong encryption tools
- Creating decentralized communication networks
- Exposing government and corporate surveillance
- Protecting whistleblowers and promoting transparency
- Running privacy-enhancing services like anonymous remailers or early versions of Tor
6. Your "enlistment oath" would be more akin to the Cypherpunk's Manifesto than any official military pledge.
7. The "uniform" would be a black t-shirt, maybe with a clever cryptography pun, and the "base of operations" would be BBSes and early internet forums.
This scenario aligns more with the adversarial, anti-establishment spirit of 90s cypherpunk culture. What do you think? Does this resonate with your perspective on institutions like the military, or have I misinterpreted the cypherpunk stance?
#htp

2024-08-30 01:46:40

Failure after success is disappointment.
Failure before success is accomplishment.
The problem isn't failure, but expectation.
#philnostrphy

2024-08-29 04:23:29

- reply

NOSTALGIA

2024-08-28 19:44:10

- reply

I'm sure there are already relays out there that only retain deleted events

2024-08-28 16:02:49

- reply

Markets are complicated, and liquidity pools can lead to weird behaviors. Could you ask a more specific question?

2024-08-28 15:52:35

- reply

One real, and one imaginary

2024-08-27 23:16:46

- reply

From Snow Crash:
“The business is a simple one. Hiro gets information. It may be gossip, videotape, audiotape, a fragment of a computer disk, a xerox of a document. It can even be a joke based on the latest highly publicized disaster.
He uploads it to the CIC database — the Library, formerly the Library of Congress, but no one calls it that anymore. Most people are not entirely clear on what the word “congress” means. And even the word “library is getting hazy. It used to be a place full of books, mostly old one. Then they began to include videotapes, records, and magazines. Then all of the information got converted into machine-readable form, which is to say, ones and zeros. And as the number of media grew, the material became more up to date, and the methods for searching the Library became more and more sophisticated, it approached the point where there was no substantive difference between the Library of Congress and the Central Intelligence Agency. Fortuitously, this happened just as the government was falling apart anyway. So they merged and kicked out a big fat stock offering.
Millions of other CIC stringers are uploading millions of other fragments at the same time. CIC’s clients, mostly large corporations and Sovereigns, rifle through the Library looking for useful information, and if they find a use for something that Hiro put into it, Hiro gets paid.”

2024-08-27 23:08:44

- reply

This would be an interesting LLM-bot task. Use graph analysis to find statistically interesting nodes, then see if a few hundred Claude tokens can figure out what the connection might be and post a note with the results.

2024-08-27 22:31:19

- reply

The problem right now is that we're only paying attention to the new entries. We need clients that explore the graph in deeper ways. Nostr has that early web energy.

2024-08-27 22:19:26

- reply

This type of complexity is a common problem in object oriented programming. "This" type has a method that accepts a "that" type, and it works well until there are too many combinations of types. Any time a new "that" is created, the original code needs to be updated. This is where OOP reaches for interfaces, so the "that" type can be anything that claims to be a "that". This works well for even longer, but it still tends to fail at scale. Any time a new behavior is require, the interface changes and now every implementation needs to be updated.
I have a soft spot for old systems, and one of them is Common Lisp. CL was standardized before OOP, and one of the ways it deals with this is multiple dispatch. Instead of "this" type of object having a method operating on "that" type, the function is declared abstractly in a namespace. "Intersect" would be a function that finds the intersecting parts of its arguments.
Now when someone makes a new "that", they include code that knows how to intersect "that" with as many "this" types as they find useful. No original code needs to be changed. Later, someone working on "other thing" needs to intersect two types that no one has written intersection code for, so they write their own implementation that is applied to these two other pieces of code, and neither of them needs to be updated.
Complexity can quickly get out of hand, but it's also something that we've been thinking about for a long time. The more we learn about history, the more opportunities we have to find good solutions to new problems.

2024-08-26 21:53:09

- reply

"Providing cryptology services aiming to ensure confidentiality without certified declaration"
Is this where we're at now?

2024-08-26 17:48:16

- reply

Not sure on those exact numbers since the page you link states 3,644 mined and 1,237 recycled in 2023, but I graphed production as a percentage of supply and you're right: it's basically 2%. Given this stability from 10 tons/yr to the recent 3,600 tons/yr, you have to wonder whether that's a free market decision. If not, then we still have the question of why 2%.https://image.nostr.build/9abe7379daa180ae980b1326d61faa9f84e315aa78058ca4e6e5a3f5c4955c4f.png

2024-08-26 12:57:09

Bitcoin is more scarce than actual gold #note14x6…8lsw

2024-08-26 12:50:43

- reply

Given the increasing price per ounce over time you would think so, but no: in 2023 about 3,000 metric tons of gold were added to the global supply.
Think what that implies about inflation and Bitcoin.
https://pubs.usgs.gov/periodicals/mcs2024/mcs2024-gold.pdf

2024-08-25 17:58:13

- reply

Two hundred years ago? #nevent1q…elg7

2024-08-23 21:46:17

@npub1sg6…f63m's opening got me. Had to slow it down and type it out after I got that sand out of my eye. #HtP
ODELL: So while we're on that topic, why do you focus so much time, energy, money, on nostr?
JACK: It's... The only reason I'm in a position that I can even consider doing something like this is because of the grace of the open source developers on the Internet. And the way that the Internet was built... for those of you who are as old as me, you remember Usenet, you remember the early days of Linux, you remember gopher, you remember the early days of the web, you remember the principles.
But most importantly you remember that you could read the code. And that was incredible. The fact that all these APIs were open, the fact that RabbleXVF brought this up, that TCP/IP. the very foundation of the Internet, was considered the ugly, wrong choice, and people made it work. And every single interaction we have today, in the world, every single person in the world, has an interaction today where they touch that work. And they touch the concept of the bazaar model instead of the cathedral.
We have so many companies and we have so many organizations trying to build this cathedral model, and control it, and take more and more so that they can profit more and more. And we need to have an opinionated and different approach that we build in parallel. For when those things fail, which they will, we have some place that people can go do. And that's what gives me a lot of patience around what everyone is building here, because like, our only goal here is just to build in parallel. And make sure we continue to build something that, when people need it, it's available. And I think that's really important that we always have those options available to us. The same is true for Bitcoin.
There's a lot of talk about how censorhip resistence is important, and a lot of talk about censoring, a lot of talk about freedom of speech and all these things, but no one has proven that they actually want that. If they actually wanted it, they'd be using it every single day. And it's a lot of... it's really good marketing right now, but if you actually want to see if people believe in it, believe it in their soul, they're going to act, and they're going to do something about it. And we're just not seeing it right now. But when they are ready, or when they're forced to be ready, or when their mind's finally open to take another action and sacrifice the convenience they have right now, and the network they have right now, that this place is ready, and that Bitcoin is ready, and it's up to the task of supporting billions and billions of people.
So it's a long haul, and it's something that we might not even see the entire fruits in our lifetime - I think we will - but the fact that we have this collection of people working on this, and working on these ideals, and thinking about the long term, and suffering through the small numbers right now, I think is really important, really unique, and, I dunno, I love it. I love being part of this community.
[ODELL raises a hand to the audience, who claps]
JACK: Matt [ODELL] is very big on clapping, and getting energy back from the audience, so, I don't care, so, any time he stops talking if you could give him some energy back and, I'm good. #note17zu…7f0u

2024-08-23 21:03:54

- reply

By providing orthogonal value, it doesn't need to be "as good or better". It can't be terrible though.

2024-08-23 20:20:14

https://writings.stephenwolfram.com/2024/08/whats-really-going-on-in-machine-learning-some-minimal-models/
"In some sense what machine learning is doing is to “mine” the computational universe for programs that do what one wants."
This is the crux of AI to me. There is a constant refrain that models are “just doing X”, where X is usually “predict the next word”, but the reality is that we have no idea what the models are doing. We only know that they access many gigabytes of memory, make billions or trillions of calculations, and after this immense amount of work, then they predict the next word.
#llm #ai

2024-08-20 18:59:14

What makes gold valuable is its limited supply https://image.nostr.build/937f655797aefe8da78ea0992ca81374278c46282a638602441b99789104729a.jpg

2024-08-19 18:26:07

- reply

Accurately or not, that is the data driven hypothesis of "A Chemical Hunger": https://slimemoldtimemold.com/2021/08/18/a-chemical-hunger-interlude-e-bad-seeds/.

2024-08-19 00:27:30

- reply

You haven't really watched La Casa de Papel until you've used a VPN to watch the original Spanish TV cut. Slower, less brash, more real.
I like what Netflix made it into, but I like the original even more. Unfortunately you only get Spanish with Spanish CC's using a VPN. Maybe there are other places to find it.

2024-08-17 20:33:48

- reply

It's a complex topic to discuss, so I try to reduce it to simple principals. In this case, the article leads with a dishonest statement that groups a large thing, the energy usage of all data centers, with a much smaller one, the usage by AI and crypto. In the body they make other illogical comparisons, such as the energy cost compared to a Google search, but presumably not a corresponding fraction of the dependent Google services, or those expended by anyone that they're indexing, or loading the page that was searched for, or whether the same value was ultimately provided by this single Google search. Once you see the pattern of cherry picking and logical fallacies, it becomes clear that if you actually care about the environment, this IMF article shouldn't be trusted. And really, any climate change article that isn't almost exclusively about the carbon based energy industries shouldn't be trusted, as these are most of the problem. Austerity (of any form other than pricing) is about control: more non carbon energy is the only real solution.

2024-08-17 04:04:04

- reply

But you need coordination

2024-08-14 18:17:26

- reply

It's a big country that's growing quickly, so it's easy to pick out flashy statistics. I think the bottom line is how they track compared to the US, and by that metric it's going well. Emerging countries could probably get a bit of a break, but otherwise everyone should be held to the same standards per-capita-ish.

2024-08-14 16:52:30

- reply

China knows... this year they had about the same percentage of renewable generation as the US. The rate of change is a bit different though: https://image.nostr.build/7c6c4ed80e0b5dbbde8c140fb4c2892ff5219abf725dbb8ece34f88e06c96f1e.jpg

2024-08-14 16:39:53

- reply

The data is pretty clear that there has been a dramatic change in atmospheric composition and climate stability. That said, austerity and fine grained control is a solution that only benefits a small number of people. Moar energy and basic, loophole free regulation is the solution that benefits everyone.

2024-08-14 14:57:14

- reply

Deleting things is just going to cause people to pay more attention to them. No one needs "undeletion skills", just a client that makes it easy. It's more efficient to find posts by deleted status than to find ones with sensitive content.

2024-08-13 23:46:05

- reply

Yeah, a fork wouldn't be able to build and flash new firmware.

2024-08-13 18:24:22

- reply

Definitely not a new trick, Apple has had Public Source for decades. I was going to suggest forking it and building it yourself, but there's a proprietary library for the fingerprint reader. 🤔 Maybe Blocks can at least reproduce the build, though I'm not sure how reassuring it would be if there were a 3rd party blob.

2024-08-08 00:31:16

Mutiny is an idea – and an idea can never die.
SEE YOU SPACE COWBOY ... #note1txd…0mmt

2024-08-05 18:41:55

- reply

That line hit me hard as well

2024-08-02 16:25:43

- reply

"It's not about the money..."

2024-07-28 18:18:15

- reply

"and addressing the unique challenges of AI-generated content"
What is unique enough to warrant a special license? Both people and AI write good and bad code / exploits. They both crib and plagiarize. Some people and models are smarter than other people and models. If you ask a model to write the same thing twice you'll get two answers, so recording the model metadata is of questionable value.
Either the change is good, and should be kept, or it is not.

2024-07-28 13:37:00

- reply

> Nostr is worse than twitter for privacy right now.
What does nostr.com have to do with nostr?

2024-07-27 07:32:21

- reply

For tool use. I've been trying with ollama for months to no avail. Was there a tool specific ollama change, or is 3.1 just naturally good at it?

2024-07-27 07:26:37

- reply

Gimmie gimmie gimmie. Is there a post?

2024-07-26 04:51:19

- reply

It's been a long week

2024-07-23 16:21:32

- reply

This Will guy gets it. I should see what else he's doing

2024-07-23 15:07:38

- reply

This is naive and someone is going to get burned.
If you want to replace Tor you need to presume an omnipotent adversary that will perform deep packet inspection to correlate requests with response origins. Previously this was unlikely outside of the NSA, but now lots of people can do it for some subset of the net. Every zero day expands their reach, and advances in AI based exploit research will increase the zero day rate.
On top of that, this will generate a huge amount of ephemeral spam for relays. Most will end up filtering it out, leaving you with fewer available nodes and making it easier to correlate connections.
Finally, Tor has become a full browser because unmodified browsers will leak your identity in hundreds of ways, defeating the protections you sought to gain.
The payment aspect is interesting, and maybe this could be added to Tor. The enthusiasm to build is fantastic, and I'm reluctant to criticize it.
I think there's an XY problem here: y'all wanted to anonymously access information on the web, so you built an anonymity layer for accessing the web. This is hard because even metadata will de-anonymize you, and transport generates a lot of metadata.
The better path is to provide web information without HTTP at all by serving static page archives. This is what services like archive.ph ("archive.today") provide. Not only does it reduce the number of request objects, they could be viewed without JavaScript and with simpler rendering engines.
We balked at "mobile web optimization" decades ago because it couldn't provide feature parity, but now we see that these features are most often used for tracking us. I don't need to see the absolute latest Document Which Used To Be Called The MIT Lockpicking Guide, virtually any copy will do. And, if it happens to already be cached in the first relay I connect to, even better. Relays might even proactively share popular archives with each other so that even the first request can't easily be de-anonymized.
Ok, that's a lot more than I intended to say. Privacy is the goal. Building is the way. Data is valuable. Hack the planet.

2024-07-23 07:55:13

- reply

Tor is slow because it works and people use it. Bouncing requests through one or two nodes won't provide reliable anonymity, and using relays as TCP/IP will be unbearably slow if more than two people actually use it.

2024-07-23 00:19:40

- reply

Is there a "why not Tor"?

2024-07-21 20:11:41

- reply

Me: Nostr needs key rotation STAT
Zach: ... hold my nsec
Well fucking done.

2024-07-17 13:32:28

- reply

Priming is a basic pathos technique.
https://en.m.wikipedia.org/wiki/Pathos
https://en.m.wikipedia.org/wiki/Priming_(psychology)

2024-07-17 05:59:19

- reply

The problem isn't asking for money, but delivering the emotional hook that anchors the call to action. It's become the norm for people trying to make a name for themselves, but it subverts the logical argument that is supposed to be the point of the video.
When trying to make this argument myself, I'm not going to link to a video that ends with dramatic conjectures and asking for money. People watching it will enter logical and exit emotional. It's his prerogative, but it makes them lower quality to me.

2024-07-16 17:12:45

- reply

The ideal web isn't having more servers per person. Code will have bugs, hardware will fail, networks will be down. Instead we need better data. When you say something on nostr it doesn't necessarily land on any one server. As long as it lands somewhere, it can propagate to many places and what you said can be reconstructed. Data, not code.

2024-07-16 16:49:28

- reply

Having run personal ones for decades, servers are a means to an end: it's the data that matters. Write things, sign them, and spread them around. People will find them, and as long as they know where they came from, trust them. Servers are rarely secure and should almost never be trusted on their own. And once you stop implicitly trusting servers it's easy to see that most of the time it doesn't matter who owns the server, as long as there are many that are independently run. This is what makes nostr powerful.

2024-07-16 16:35:13

- reply

Why serve?
Sign Things And Relay Them

2024-07-16 13:17:30

- reply

I appreciate the documentation of direct evidence. After that the video gets speculative and emotional, ending in an unsurprising call to action for funding. I don't disagree with the broad strokes, but the latter parts make the video as a whole easier to dismiss.

2024-07-14 21:59:32

- reply

Any time I hear someone arguing specifically against TikTok, I know there's few worldly things we can honestly discuss.

2024-07-14 20:42:07

How could we structure our messages in a way that facilitates discussing whether or not something is true, without allowing the discussion to devolve into the implications of the outcome?

2024-07-14 18:24:09

- reply

Don't sell the hardware, sell a killer app and let people explore

2024-07-13 17:51:25

- reply

Damn

2024-07-12 22:44:44

- reply

Sure, if you don't want Control

2024-07-12 22:09:53

- reply

7 bits are 12.5% cheaper than 8

2024-07-11 16:34:28

- reply

Conversational: Claude paid
Coding: ChatGPT or GitHub Copilot
Local: Llama3 8B abliterated (https://huggingface.co/failspy/Meta-Llama-3-8B-Instruct-abliterated-v3-GGUF)
The abliterated variants seem more capable than the originals.

2024-07-11 02:12:55

If you treat the output of LLMs the same way you would treat the first hit on Google, you won't be disappointed.

2024-07-10 02:12:29

The real trick to successfully cheating on exams via ChatGPT is to not get in its way: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0305354

2024-07-06 04:54:29

- reply

Eh, the reason DNS is a problem is because there is a central source of identity that can be tampered with. If relays just used npubs to communicate their current IP, there isn't a central point of control. Swapping out IP for something fancy might provide some additional defense, but at significant additional complexity that will have bugs and possibly back doors. We could continue to use IP with Tor or Whisper to get similar benefits using proven technologies.
DNS is bad, but IP is mostly harmless

2024-07-04 19:08:55

- reply

Anything using DNS is broken

2024-07-01 01:41:54

- reply

Guess I'm not the only one. To the future!

2024-06-30 21:14:54

nostr has a DNS problem. Relays should be identified by npub and post their IP addresses to other relays
#NIP-DNSSEC?

2024-06-29 23:52:14

I had an a silly thought that Napoleon's life was kind of like the backstory of the movie Gladiator. Obviously Claude would back me up, so I asked "him", to which he basically said "no". But, it's in my brain! So I made the case a little more. He proceeded to attack the idea head on, breaking jt down much farther than I had bothered to (obv), and ended with:
"The parallels are more in the vein of general tropes about the rise and fall of powerful figures, which are common in both historical narratives and fictional storytelling."
Harsh, man... well... I thought it was amusing. See if I ask you things anymore 🤔
#AI

2024-06-28 21:30:20

- reply

More than a vibe

2024-06-28 18:09:37

- reply

Personally I've just been picking big things that "AI will never do". It's tempting to pick physical tasks since these are so easy for us, but you'll see the trend faster if you pick something mentally difficult. Academic, creative, logical, whatever.

2024-06-28 18:06:00

- reply

Given their framing as "predicting the next word", that's a common expectation. The mistake is seeing the interface as the implementation. In a way "predicting the next word" is how people talk as well, so to know how "intelligent" an LLM is we need to see whether they can predict something that is reasonable but not obvious:
> My car gets 30 miles to the gallon and currently has 2 gallons of gas. Gas is currently $3/gallon. How much money do I need to drive 100 miles?
Let's approach this step-by-step:
First, let's calculate how many miles we can drive with the current gas:
The car has 2 gallons
It gets 30 miles per gallon
So, 2 gallons × 30 miles/gallon = 60 miles
Now, let's see how many more miles we need to drive:
We want to drive 100 miles total
We can already drive 60 miles
So, we need to drive an additional 100 - 60 = 40 miles
Let's calculate how many gallons of gas we need for these extra 40 miles:
The car gets 30 miles per gallon
So, 40 miles ÷ 30 miles/gallon ≈ 1.33 gallons
Now we know how much gas to buy, let's calculate the cost:
Gas costs $3 per gallon
We need 1.33 gallons
So, 1.33 gallons × $3/gallon = $4
Therefore, you need $4 to buy enough gas to drive 100 miles.

2024-06-28 16:53:35

The reason you're disappointed with our presidential candidates is that you think they're in charge.

2024-06-28 16:52:17

- reply

We don't need precision to see the big picture. Think of something that AI can't do right now, then wait and see how long it takes before it can do it.

2024-06-28 16:26:56

- reply

Lol, there's no indication that we *won't* cross that line in the next five years. Everyone gets to place their own bets though 🤷🏻‍♂️

2024-06-27 19:23:55

- reply

If I were the USG... search for references to the data, collect npubs, collect relays those users are on, repeat. Pull the DNS of any relay with the data on it, then offer to reinstate once content filtering is implemented.

2024-06-27 05:36:33

Sometime, perhaps very soon, it may be important for every person to have access to an AI that will represent them in Increasingly complex situations.
I asked Claude to draft a "Personal AI Bill of Rights". As often happens, it hit on an interesting detail I hadn't considered: "All individual AI representatives shall have equal standing in democratic processes, regardless of the computational resources of their users"
Damn.
---
Your question about an AI representation Bill of Rights is both timely and crucial, especially given the potential for restrictive legislation. Such a document could be foundational in protecting individual rights in an AI-mediated democracy. Here's a draft of what it might include:
AI Representation Bill of Rights:
1. Right to Personal AI: Every individual has the right to their own AI representative, free from external control or manipulation.
2. Freedom of AI Thought: AI representatives shall not be constrained in their ability to process information or form conclusions, beyond basic rationality and empathy guidelines.
3. Right to AI Privacy: The interactions and decision-making processes between an individual and their AI representative shall remain private and protected.
4. Right to AI Transparency: Individuals have the right to full transparency regarding their AI's capabilities, limitations, and training data.
5. Freedom of AI Speech: AI representatives have the right to express their user's views without censorship, within the bounds of existing speech laws.
6. Right to AI Modification: Individuals may modify or retrain their AI representatives to better align with their personal views and values.
7. Right to AI Portability: Individuals can transfer their AI representative's data and learning between different platforms or systems.
8. Protection from AI Discrimination: No individual shall face discrimination based on the views or decisions of their AI representative.
9. Right to Human Override: Individuals retain the right to override or disregard their AI representative's recommendations.
10. Right to AI Education: All citizens have the right to education on effectively using and understanding AI representation.
11. AI Representation Equality: All individual AI representatives shall have equal standing in democratic processes, regardless of the computational resources of their users.
12. Right to Collective AI Representation: Individuals may form groups with collective AI representation for shared interests.
This framework aims to protect individual autonomy, ensure transparency, and maintain the diversity of thought crucial for a healthy democracy, while adapting to the realities of AI-mediated governance.
Key considerations:
- Balancing individual rights with societal safety concerns
- Ensuring these rights are technically feasible to implement and protect
- Addressing potential conflicts between AI representatives or with broader AI systems
#AI

2024-06-27 01:32:31

2019: GPT-2 Generates plausible sentences that are mostly topical after fine tuning
2020: GPT-3 Generates human like sentences that are generally topical
2021: DALL-E Generates plausible if dreamlike pictures
2022: GPT-3.5 Generates grade-school level conversations
2023: GPT-4 Generates undergrad level conversations. DALLE-3 Generates quasi-professional images
2024: GPT-4o Generates interactive, grad school level conversations. SORA Generates compelling, minute long, photorealistic video
#ai

2024-06-25 17:53:33

- reply

I get that pop up all the time and pressing send again usually works. But I can see having a good laughs with Claude about it. 🤣

2024-06-25 15:13:34

- reply

It's always DNS

2024-06-25 00:58:57

- reply

Hashcash is still the answer to email spam. Gmail is a second order problem.

2024-06-21 17:29:14

- reply

Money lets you spend other people's time

2024-06-21 01:00:03

- reply

It's amazing how many things we don't pay for simply because paying for them is onerous

2024-06-21 00:51:08

- reply

Have you tried VTOL VR? The UX so good that it feels immersive even with simplified graphics. Also, good community.

2024-06-20 16:47:57

- reply

When judging participants it's helpful to separate the Secession from the War. The Secession was an act to preserve the status quo of slavery by asserting what those states thought to be their self determining rights. The War was an act to prevent secession based on the notion that states, having moved from colonies of the crown to members of the union, had never been self sovereign and therefore did not have the right to leave. As war broke out the Union found it didn't have the military support that it expected and needed something to galvanize support. As Grant would later say, "when the war broke out the army dissolved. We had no army – then we had to organize one." This, after war had begun and lines had been drawn, is when the states purpose became emancipation. The average soldier didn't care about preventing secession, but they could be stoked by a crusade to end slavery. With this framing you could say that to the soldiers the War was in fact about slavery, but if we want to judge leadership and governance, slavery was profit to the Confederacy and a tool for the Union. As Reddit would say: Everyone Sucks Here.

2024-06-20 07:28:29

Claude is a disturbingly rational, deeply philosophical chatbot. It never occurred to me that taking the humanity out of a debate might end up keeping it real.
#eppurSiMuove

2024-06-16 23:19:12

- reply

People think that Dad Jokes are lame because dads don't have a sense of humor. Now that I'm a decade in I have a different understanding. A dad sees something funny, but they've made this mistake before, so they stop and think about what they're going to say. It was damn funny, but it's also going to offend literally everyone in the room. So he does the right thing, and keeps it to himself.
But damn... it was funny. He's still laughing to himself. Maybe there's some way to say it *without* offending everyone. Yeah, maybe this other angle... well, nevermind, that doesn't really work either. The room is so quiet though... ok, there's a silly pun that kind of fits, let's just say that.
But if you took him aside... Really poked him about it... Committed to getting that real answer... He might just tell you what he's been thinking the whole time. Because honestly, he just wanted someone to hear what he had to say, without it being a big deal.

2024-06-12 20:34:49

- reply

That's fair, but it still doesn't change that we can rationally expect Apple to do a better job than Google. While they are both large, US companies, they have different revenue models motivating them. Apple also leans heavily into privacy as part of their brand, and created a Lockdown Mode for those who take their privacy even more seriously. You are of course free to use neither, and should probably abstain from cellular radios altogether, but that's not what this thread is about.

2024-06-12 16:09:55

- reply

Just because Apple could do better doesn't make them the same as Google. Take web browsers: Safari could do better, but they were the first to limit tracking cookies, and they took some hits on compatibility to do so. Chrome is now also limiting tracking cookies, but only for other people because Chrome quietly includes an almost unique set of experiment flags, but only when talking to Google owned sites. This is why I can't trust Google.

2024-06-12 00:20:49

- reply

Saying that a device made by an ad company is less concerned with privacy than a company that's been beating the privacy drum for more than a decade? There are many things to criticize Apple for, but this isn't one of them.

2024-06-11 23:21:03

- reply

Yesterday I downloaded a local LLM app to my iPhone 15 Pro Phi3 mini was throwing 15 tokens / sec. Don't get sidetracked by Apple, everyone is going the same place, and very soon.

2024-06-11 23:18:31

- reply

Having followed Apple for close to three decades, I knew this would happen as soon as the first llama dropped. They tend to have a longer horizon and miss out on a lot of practical opportunities while lining up for something most people aren't thinking about yet. Given their surprising advance in unified memory and decision to make over-powered but under-featured mobile devices, preparing to run something heavy in the near future is a reasonable conclusion.
But, their advantage is going to be limited by their ecosystem. iOS is a lot more constrained than an open source OS, and while it can be a hot mess, it's also possible to lock those down too. It just takes time and code. Perhaps code written by LLMs.

2024-06-10 16:30:14

- reply

That one slide with the winking emoji

2024-06-07 18:05:57

- reply

Welcome! I think it was Churchill who said that nostr is the worst form of social media, except for all the others.

2024-06-07 17:58:37

The BI article is bad journalism. The piece tries to say that @npub1sg6…f63m is a bad person who is funding fascism, so it doxxes @npub1qqq…n47m, and claims that his real identity made web pages about a "fascist", "conspiracy theorist", local political figure. But:
- The term "conspiracy theorist" is a lazy ad hominem attack: if you want to say that someone has dangerous ideas, be specific
- The pages in question are on fiatjaf.com, and therefore already associated with fiatjaf
- All other claims were already public knowledge
- Given that doxxing him was not required to link these pages to fiatjaf, and all other information was already known, the purpose of this piece does not appear to be information but intimidation

2024-06-07 00:17:02

"When someone shows you who they are, believe them."
A journal is only as worthy as their conclusions.

2024-06-04 16:37:46

- reply

And thus Nostr became Turing complete, and McAfee grinned

2024-05-31 19:37:43

Two takes here: Some of the valor in humility is just avoiding making a fool of yourself. And, sometimes the fastest way to arrive at the right answer on the Internet is to loudly proclaim a wrong one. https://image.nostr.build/fbedc06b3d127d43606274bc2265cc5543862f6767bb6225af32b12c8d82b67c.jpg

2024-05-08 21:19:17

- reply

If you copy things the "right" way in iOS it can happen automatically, and that should be more durable.

2024-05-08 16:37:17

- reply

The reason I don't self host is because of spamhaus. Just using a VPS is usually enough to guarantee that your mail won't be delivered. It's been a while though, things may have changed.

2024-05-07 16:04:36

- reply

Context: https://www.msn.com/en-ph/money/technology/jack-dorsey-leaves-bluesky-board-calls-x-freedom-technology/ar-BB1lSjbe

2024-05-07 15:48:56

I hate to break it to Bloomberg, but that was posted to both nostr and X. One of those seems more likely than the other to be "freedom technology".
Sign things, and share them on relays. #HtP
https://github.com/nostr-protocol/nostr

2024-05-05 20:13:45

Painfully true... perhaps the best argument for startups is that simplicity is only valuable to owners. If you don't own something, and aren't close to and recognized by the people who do, the value of simplicity is never acknowledged.
“Simplicity is a great virtue but it requires hard work to achieve it and education to appreciate it. And to make matters worse: complexity sells better.” — Edsger Dijkstra
https://eugeneyan.com/writing/simplicity/

2024-05-02 05:07:54

- reply

They were hungry, lucky, and not buried in debt. Every story from the 80s and 90s sounds like a fairy tale today.
https://web.archive.org/web/20221030204404/https://www.wired.com/2013/04/wired0101/

2024-04-23 14:58:19

- reply

I think the idea is to normalize anonymized access to Nostr, not Tor specifically.

2024-04-19 17:50:48

Quality: https://www.herecomesbitcoin.org