fellowniusmonk

Plain text is way harder than people think. Partly because the reflow rules change depending if you are making edits in front of or behind the cursor/point of action.

On anything structured that doesn't require re-flow this approach makes some sense.

Lists, Grids, etc, there are pro/cons but it makes sense.

In this case you ARE causing reflow, so content and format change at the same time.

With normal cut and paste the reflow happens at the point of attention, and the final state paste keeps your eye right at the point of action.

This jumping around stuff means you are doubley losing tracking if you paste below the cut point. This is effectively an out of bound focus shift. This has higher cognitive penalty because layout is no longer occurring in serial but in parallel in multiple places.

idempotence, types, rejection of global state, all of these "engineering" rules go out the window when you enter the realm of mereological nihilism that is text editing.

The best you can do it think of things as causal chains like some CRDTs do.

You see this very often, these kinds of things keep escalating and with the destruction of the rule of law and the need to suck up to an executive a country basically becomes a black hole for investors.

Every time the chain of evil is allowed to persist it corrupts a widening tranche of people and the inertia for compromise just increases until sudden system collapse.

It's just normal error propagation in the network. Only taking ethics and morality seriously at the grass roots level can stop this, but people have gone absolutely wacky for ingroup and apocalyptic thinking and the lie "everyones doing this" has been really pushed hard as a thought stopper.

The U.S. "executive" class is so obsessed with the "exploit" part of the explore/exploit cycle that it's very clear they are prematurely closing advancement. Better a little money and power for them now than a lot of money and power for their country/humanity.

This has an element of stochastic improvement so it's hard to predict but the chance of the U.S. "winning" this "race" is pretty bleak.

You see this all the time in communities that have internalized hierarchy as a "good", little kings of shit mountain vying for less and less at a higher and higher cost.

I think thats deeply uncharitable and fundementally superficial.

I think people (many/most) don't want slaves.

Our imaginations just outstrip our abilities and we desire them to match.

I have a UI I developed from first principles about 10 years ago, I use it to replace my os, its like my personal do everything emacs os but not at all emacs, and its funny to see other Uis iteratively convergence on it as context finally gets its due.

Chiltepin started growing in our yard thanks to some bird.

They are amazing, high heat small dose, I'll just crunch one or two in the morning, the only downside is my spice tolerance is now accidently extremely high.

The real hack is that the events aren't the community, they serve the community, the process of working together to put these types of events on is what builds deep connection.

People don't actually get where the deep value lies, the event income or social credibility for those involved in putting it on just helps ensure there is enough fuel for the fire of the real community.

This is entirely about social friction.

Blaming the people trained by the smartest people on earth (with population level ad sales and a/b testing) to reject friction until they start to feel it as a poison isn't their fault.

We built a low friction co-working space that was mostly a social club after work hours, and by reducing that friction even the most intense introverts ended up integrated.

It's not difficult it's just hard.

I basically did this for a couple years around 33. I was the highest paid IC for a multi billion dollar "digital transformation". I was writing code in 5+ languages, doing hardware retrofits, had r/w access to every system, climbing ladders to deploy hardware for my own skunkworks projects.

The whole thing was crazy. I mean actually insane. I look at my productivity from the time period and it blows me away.

Really enjoyed it. Being the one guy who was damned adamant I wouldn't manage anyone kept me out of the political fray as well. I drank and I knew things. I probably had more freedom and access than anyone ever has had at a company that size, I mean they literally ended up paying for me to have some degree of DHS clearance to operate unrestricted in passport zones, it was wild.

Is it good for a company to have those roles or not? I don't know, after starting several companies and also working in corporate, that was the most fun I think I've had. Moving the needle, unfucking things, reducing complexity.

Providing accountability where there was none before.

I left because they wanted to me to move. I was flown out every Monday and flew back every Friday but I didn't want to move cities and I had a startup I wanted to launch. I took 3 months traveling and then 2 months writing code for my 4th startup.

There are _almost_ no universal rules, my work saved them an insane amount of money and generated even more, but the methods only made sense to dig them out of the completely unaccountable traffic jam they had built for themselves.

If you ever read the James Bond books as a kid, there is inner monologue explaining why he is so overt, his whole modus operandi is to be conspicuous and to shake things loose in opaque situations. When your works starts holding peoples feet to the fire you have to come extremely correct and it paints a target on your back until you have sufficient track record. It helped me learn a lot about myself and ultimately taking the time to inhabit the role fully and lose myself in it helped me understand what I really value.

I mean we have all kinds of under synonym'ed words. Just look at how few we have for "smell" (as in the act of smelling), and then how overloaded the word smell even is.

I want a clock partially tied to gps that purely measures my dead reckoning movement along with a universal lock as near as we can tell.

A causal wristwatch that purely lets me label my self relative increments. I'd allow cesium oscillations on my person as well I guess.

We can all then crdt ourselves together for group interactions or I can just live on my own branch.

I mean it's a fucking joke, I kept getting refusals on a code base I wasn't familiar with and it was literally just because there are some vars named DNA. Just absolutely stupid.

I am absolutely shocked and freaked out by the number of long lived accounts on this thread that can't detect the fact that this is Ai composed.

I had a file that had a couple places where vars were named DNA and got just total refusals during the first launch. Came away thinking the model was total trash. The guardrail classifiers are for sure total trash.

Some random thoughts:

It is very odd, I do more deep reading then ever before but its curated through llms.

I do the exploration then the dive on papers.

Many novels we love were released as serials.

I haven't gotten stuck trying to understand an idea because its poorly explained in a book in a while.

I think humans have been very inefficient at finding gaps in logical progression in explanations because anyone who already knows it subconsciously skips steps in the explanation.

That said, llms, woof. Still so much misunderstood.

I think that's a generally fair statement. I understand its historically pretty common to see a theology to law pipeline.

How else do you propose a system that is fucked like that ever gets unfucked?

I have one area of my education that I highly value but its very hard to explain without people importing a lot of assumptions.

I like to call it critical listening but also its textual evaluation.

In addition to some didactic instruction my Father gave me a short book on the principles of hermeneutics around 13. We went to different churches over the years growing up but I would bring my bible, take notes, and on the drive home from service he would ask me if anything unsubstantiated by the text was snuck in, anything against the text, etc.

In the hundreds of sermons I took notes on over the years there were only 3 without obvious butchering of the text, statements directly contradicting the very text being examined, nightmarish hermenutical implications, outright fabrications, etc.

The shear volume of evaluation I did against a static text was interesting.

It helped me understand how to parse language, how to do evaluation, just a lot of stuff in a way that was more dynamic than something like debate club.

It also helped me understand how self servingly imprecise people can be and the ways in which deceptive and misleading language is used.

The revenge of the _nearly a dozen_ philosophers.

Awesome! Besides size, how does this compare to gte-small?

Discovery, which was hard for so long, is now hard again.

No longer because communication and transmission is expensive but because it is so cheap.

Interesting how much of this writing was dictated.

Combine the two generals game with the implications of value based pricing. Catastrophic unaffordability is a guarantee.

Whoever wrote this, I find them incredibly unsympathetic.

Some people have the desire to gamble their lives away, to take on extreme amounts of debt and then die in despair.

Same goes for alcoholism.

Same goes for ponzi schemes.

We don't ban alcohol because bans of it drive organized crime.

On media consumption apps the west blocks children from seeing porn, adults from seeing snuff films, adults from seeing how to make a bomb.

I don't think I believe anyone who claims to stand on these things on principle unless they are also pro right to repair and pro open source. I won't believe you are advocating for freedom if you only ever argue for exploitative extraction of resources via preying on human frailty.

And personally I think we should ban skinners box algorithms from being the default on any app.

We explicitly make apps to be as lazy as possible to use by people and low IQ people or obsessive people with poor impulse control and to extract as much of their money and attention as possible, and behind closed door and in full view tech is very explicit about this.

Separately, some people are destructive idiot or do idiot things, they do actual damage to our society and make everyone have less resources, breaking windows, burning money, living in "maladaptive fantasy", etc.

These types of very simple arguments about human nature and freedom as a virtue coming from someone openly bragging about running exploits against human nature seems like a wolf bemoaning the sheep dog.

Yeah, I think this is a bad take on the part of the EFF.

The full dataset will be in private hands and available everywhere to everyone to do anything the moment this goes live.

I protect data for a living, cost asymmetry and proof of work are really the only tools we have.

If this goes live, the next time I get subpoenaed to testify for something I'm going to throw in so many random but couched accusations at people just to screw them over when the data goes live at scale via data brokers and torrent dumps.

Bill Jones (accused in court testimony of killing puppies) is requesting the HOA get off of his back.

A collection of thoughts on this.

Pierces Firstness is exactly what drives this.

The move from thinking to semantic conversion is important for investigation/introspection.

Arguing with yourself also seems to engage your brains "theory of mind" centers, so different pathways get activated to examine the problem space.

The problem with Ai is the fact that it hallucinates and if you're doing anything truly novel in an integration or framing sense it bottoms out very quickly and can't engage. A human operator can decompose the problem and get accuracy checks for known areas in the training data of course.

Now to be I'm not saying Ai can't produce novel work on the edge but in my experience it is antagonistic towards those goals.

Case in point, CRDTs, many don't use tombstones but they are the minority, and if you try iterate a new CRDT off of one that doesn't use tombstones, let's say diamond-types, it will keep pulling you back to tombstones.

The problem is that the number of humans who understand dynamic investigation and the push pull of exploring an idea you don't hold with someone has always been very small, and now with reflexive internet argument culture driving how we view "debate" and "discussion".

I don't know if we've reduced the leisure to think or what but things are not great for finding speculative thinking partners.

Consciousness is a fuzzy word and I would call it a joke if it wasn't actually dangerous currently.

And currently its widespread usage in how people relate to and talk about Ai is actively harmful.

And dont get me started on how terrible the "hard problem" is, yeesh.

This seems like a clear false dichotomy.

I have a philosophy pre-print about "empirical ontologies" I use for testing new models reasoning abilities, and it also degrades, there is no way around it and it always refuses.

It's not that the model is complete trash, it's that anthropics new approach to forcing epistemic crisis will make any model behind it complete trash.

At least they reverted the shitty mobile Keep integration that was not only an insanely distracting UI but made the whole interface laggy as hell.