mtlmtlmtlmtl

This is the first I'm hearing about anti-NMDA receptor encephalitis; what a strange and scary condition. If you don't mind the prying, I'm curious about some things.

I noticed you said in another comment that you were treated with antipsychotics in the psych ward. Would you say they had any effect in lessening the symptoms? My intuition says no, since I'm not aware of any conventional antipsychotics that interact with NMDA receptors directly(mostly they act on dopamine and serotonin receptors), but psychopharmacology is tricky that way...

And, another thing. Have you ever tried dissociative drugs(like ketamine, PCP, DXM), and if so, how similar would you say your experience was to those drug experiences? Of course, feel free to tell me to fuck off if you don't want to discuss that in a public forum.

Glad to hear that you found your way out of the psych ward to get properly diagnosed and treated. I've witnessed first hand people getting trapped in the psych system with neurological or endocrinological conditions. It can be almost impossible to get out, especially if the "diagnosis" is a psychotic disorder. Once you have such a diagnosis pinned on you, anything you say can be dismissed as a delusion, and most psychiatrists are woefully bad at considering somatic explanations for symptoms. You definitely got very lucky.

It's very interesting because my 13900K has worked like a dream from day one and still to this day. Never had any of the voltage issues, never had any abnormal crashes in Firefox or any other software. I was undervolting it for a long while, so I wonder if somehow that saved me from the voltage issues before they were fixed?

Don't be ridiculous.

And which features have been removed, as you claim? Removing code is not necessarily removing features. I use GOS and I honestly can't think of a missing feature compared to the stock OS, other than stuff not in AOSP in the first place, like gemini.

You said removing features. This link is talking about making certain feautures optional and disabled by default, not removing them.

Don't often see the great Dang commenting in a non-moderation capacity, and that's really a shame, because this is one of the most interesting, well-formulated comments I've seen on here for a while. I fully agree with your frustration with the AI zeitgeist, but I won't pretend to have any input to alleviate that frustration. I would love to much less eloquently rant about lisp(and emacs) and other vaguely related things for a bit though, so here goes.

The tendency for the Lisp community to self-segregate into these bubbles is something that's perplexed me for years. You noted that you don't agree with it; I'd in be interested in why. I do agree with it, in the sense that this phenomenon does occur to an extent, and certainly to a larger extent in Lisp compared to other languages. The part I don't agree with is that this is a universal deal-breaker for Lisp. And there's still plenty of cooperation and sharing in the Lisp community. Emacs once again is the prime example. Yes, every .emacs is as unique as a fingerprint. But the reason GNU Emacs refuses to die is precisely its wonderful ecosystem of extensions that have in fact been shared and maintained by a wider community. And this ecosystem is in part possible because GNU Emacs is devoid of any sort of boundary between application code and extension code, which is a philosophy that can be traced back to the LISP machine era.

It's also interesting to tie this in with RMS(the author of GNU emacs, what a coincidence!) and his stance on Free Software really being about every individual user being free to modify the software they run. Of course, the deep implications of this philosophy, and its implementation in Emacs and other Lisp software, are clear to me, because I'm a Lisp programmer, just like RMS. And the Free Software movement originated in a time when most people with access to a computer had at least some amount of familiarity with programming. But as computing has become mainstream, RMS' original vision has morphed, and the focus has been on the importance of the wider community being able to modify software. And that's certainly important. But it's not really free software in the original sense unless you happen to be a programmer. Most software users today are not programmers and likely never will be, so regardless of how the software they're using is licenced, they're not really free in that original sense, are they?

Then LLMs showed up. And suddenly, I can see on the horizon a revival of free software in the original sense. And yet it feels quite far away. For this to become reality, we need a lot more than what we have currently, which is a pretty damn good search engine/meme generator for the "normies", and a pretty damn good boilerplate generator for the coders. I think we probably need an entirely new concept of what software even is. I think we probably need new foundational research. I don't necessarily see a role for current LLM architectures in this. I feel more like current SOTA is scratching at the surface of the real deal, and it may take a few years to understand how to make foundational progress. Maybe we even need another AI winter to force the capital expenditure into other avenues. Certainly we need foundation models that are open in all possible senses of the word. Being tied to proprietary, heavily censored blob running in a giant datacenter is a non-starter.

It seems like the majority of people in the field have some intense tunnel vision about LLMs and transformer architectures. I'd love to see more variation there. I know Lecunn is doing some unique stuff. I can't pretend to understand whether it has legs, but I applaud the effort. Certainly we need to address the issue of energy expenditure, the absurd amount of data and training iterations required, and the lack of online learning. Human brains are vastly superior at all 3 of those metrics. So like, how about addressing that instead of just building ever larger sandcastles filled with nvidia chips? Anyone?

Those are some of the thoughts percolating through my head. I can't really figure out what to so with them. Maybe someone else can. Please? Thank you.

It doesn't have to be, but that's not really an argument for claiming it isn't. Considering how deeply embedded privacy violation is in Meta's corporate DNA, is there any reason other than hilariously naïve and inexplicably charitable, hypothetical speculation to believe this is not motivated by more privacy violation for profit, just like literally every single thing Meta has done in the entire history of the company? No? Didn't think so.

As an avid Spelunky player(still trying to complete the Cosmic Ocean...), I recently decided to explore some of Spelunky's roots, and set out to learn Nethack, and fell in love with the game. After a few weeks of dying repeatedly, perusing the wiki, and watching the Ascending in Nethack Overexplained series on youtube(highly recommended), I managed to ascend a valkyrie. Planning on trying a harder role soon. It's amazing how tense it can be despite the turn based nature of the game.

I do like the nerfs in this release. Making excalibur harder to get for Valkyries is a good one, as well as nerfing the unicorn horn. The run where I ascended felt a bit too easy at times. But of course valkyrie will still be by far the easiest role, I think. I bet I'll be stuck for quite a while trying to ascend anything else.

Although I agree INTC is overbought currently(as is the entire sector, and good news are pretty much universally overbought anyway, that's just how the stock market works), I'm having a hard time taking this analysis seriously.

Yes, it's true that some(specifically 18A) of the successes Intel have had since the CEO switch were essentially part of the plan under Gelsinger. But to say that nothing has changed seems simply uninformed, and that's me being charitable.

LBT has done lots of things, including securing significant investment, cutting costs, made some good deals in the DC segment, and most importantly, started making real progress(still early days though) in finding actual foundry customers willing to commit to 14A. Intel under Gelsinger was approaching its foundry business in a "build it and they will come" sort of way, hemhorrhaging huge piles of cash and merely crossing their fingers that customers would show up at this point. One of the first things LBT did as CEO was make future development of 14A contingent on finding customers for it, which is precisely ehat he's been trying to do.

This quarterly report was the first during his tenure that showed some real financial progress: the quarter was profitable when analysts expected breakeven, sequential foundry revenue grew, 18A yields are up, guidance looked good. This, to you, is practically nothing? It seems to me that going from bleeding cash at an accelerating rate, to turning a profit, is about very much something.

And again, I agree that hype is a bit overzealous right now, but arguably the fact that there's market optimism about INTC again is a huge victory on LBT's part. Because how exactly is IFS supposed to attract customers willing to make multiyear commitments to their services if the company is universally seen as a sinking ship? To state the obvious, market sentiment can make or break a public company, no matter how disconnected that sentiment is from the fundamentals.

And to say Intel has no AI strategy seems disingenuous. LBT has addressed AI strategy quite explicitly in public, and you can look it up if you want.

I'm cautiously bullish on INTC in the long term. I do expect a correction at some point, just like I would for any stock that goes up more than 20% in one day, but to me that's just market swings. The daily machinations of the stock market are like a tale told by an idiot, full of sound and fury, signifying nothing.

At least here, there are randomly triggered checks by shop staff where they have to manually rescan anything before they let you leave. And possibly, those checks are more easily triggered if you do certain very strange things like buying nothing but many separate instances of "bananas' with widely varying weights. Wouldn't be too hard to program a set of rules for the most obvious red flags.

And of course, the area is wide open and well covered by cameras, and usually self-checkout means paying by card or google pay or something, which will tie your identity to the purchase.

Riyadh, a city of 7 million people gets basically all of its drinking water from desal plants in the Persian Gulf. If those plants get knocked out, they're just gonna "fly in drinking water"? So with some napkin math, assuming 1 litre of water per person per day(which is extremely conservative considering they're in the desert heading into the hottest part of the year), that's 7 million litres of water every day. Can they "fly in" 7000 tons of water every single day? And where is all this water coming from? I have serious doubts about Saudi Arabia having "no problem" doing this.

I often think about the shuttle program in relation to all these crazy complicated, wildly expensive, and incredibly fragile space telescopes we're sending to LEO or the Earth-Sun L2. Would be damn useful to be able to repair/upgrade these things like with Hubble.

Obviously I realise the shuttle program was pretty far away from being able to head out to the Earth-Sun L2(AB, and wasn't even working towards it. But man, it would be nice to have that ability.

Dude, Lichess is entirely funded by donations. There's only so much money to go around.

And Thibault iirc is the kind of person that's not terribly interested in earning lots of money. Of he wanted to, I'm sure he could make bajillions elsewhere in tech, because he's that good. But he apparently prefers to only work for a "measly" upper middle class salary and doing something he's really passionate about. And I thank him for it, because lichess is awesome.

Why not? FreeBSD has never been intended as a batteries included, everything "just works" out of the box OS. It's meant to have a bare minimum install and let the user choose how things are set up. You can disagree with that philosophy, but that's not an indictment of FreeBSD. Just go use something that aligns better with your preferences.

Why? Nothing wrong with running your network interface in a VM. There are reasons for doing so even if drivers aren't an issue. Qubes OS does this, for instance, for security reasons.

Why do people(you, in this case, but this is a very common fallacy) assume that advocating for one thing(idleness) is implicitly advocating against its opposite(work)? We can do both, just not simultaneously.

Leaving aside the wealth gap factor, which I do agree is important:

When I think back to using computers as a kid, both at school(starting in 1999) and at home I don't think it's all that black and white wrt just playing at home vs learning useful skills at school.

At some point in the early 00s my underfunded elementary school acquired a bunch of old windows 95 computers. We would have classes where we mostly did basic touch typing, MS Office etc. At home, my middle class parents had also managed to find me some old outdated clunker. And yes, most of my time at home was spent playing games, chatting with friends on msn, pirating mp3s etc.

But I'd say I learned orders of magnitude more from my frivolous activities than from whatever we did at school. At home I was learning things like: online research(into warcraft cheat codes or quest guides for Runescape etc); software troubleshooting(having to reinstall windows because I downloaded malware on limewire or otherwise borked my install somehow); fast typing(from chatting with friends about whatever 12 year olds like to discuss. Probably 90% of my typing practice back then came at home, not at school, and there was no touch typing. I could type 100+ wpm on just 4 fingers by the time I was in middle school. Never actually learned to properly touch type until I had force myself into it 5 years ago due to RSI); English as a second language(from various forums, irc, etc, hard to avoid back then); And I believe one of my first forays into programming was trying to get a cracked game with a broken .bat installer off TPB to work. I had a friend who got into it via Morrowind modding.

Actually, come to think of it, most of computer class was also in reality spent sneakily playing flash games and/or messing around with the computer settings just to screw with the next student/teacher to use it.

Even generalising beyond computers, I think a remarkable portion of the skills and interests that end up defining us as people, can be traced back to stuff we did trying to avoid boredom as children.

To summarise though, I do think computers have a place in school. But especially at an elementary school level, I think play should be a significant portion of their use, because play is how kids explore the world and themselves.

MLID has been saying Arc was cancelled since before the first Alchemist cards were released.

The next(I think? It's in -CURRENT now anyway.) version of OpenBSD will be adding VLAN awareness to veb(4). Should make my OpenBSD home router experience much easier.

Tbf I haven't actually played much tennis, but I've watched plenty of it, and been playing padel for a couple years now. I would agree that tennis has a steeper learning curve, and is definitely more physically demanding. Besides the learning curve though, I'm not sure padel can be said to be any less technical than tennis, at higher levels. Padel seems to me to have a much bigger variety of shot types all requiring specific technique. And positioning is also much more fluid and complex than in tennis. Same thing for leg work.

To be clear I realise you were mostly talking about learning curve specifically; this is not an attempt at refuting what you said, just adding to the discussion.

My sennheiser earbuds are now down to 15 minutes of battery life. Less if it's a cold day. Sure, they're not completely dead yet, but they're effectively useless. And it's not like I can easily replace the batteries. Most wired earbuds or headphones at a similar price point have replaceable cables.

About BankID: There was a regression in the app back in june that broke the app entirely. Back then I emailed the developers complaining about it, and their response indicated that there was no deliberate attempt at breaking BankID on GrapheneOS, and the specific developer who replied to me said he was a fan of the OS.

Biometric login was also confirmed to work around the same time. I can however confirm that it doesn't work on the latest app version. It complains that the webview isn't Google Chrome.

This is probably just an oversight. I will email them again; good chance they'll push a fix to recognise Vanadium webview.

Well, demonstrably you have at least some measure of interest in interaction with other humans based on the undeniable fact that you are posting on this site, seemingly several times a day based on a cursory glance at your history.

I'm so fed up with Proton. I will be taking my business elsewhere. Instead of making a great product for X, they've decided to make a series of extremely mediocre products for P, Q, X, Y, Z and W, all of which are left missing the most basic features for years. Features which even the free alternatives already have. Things like supporting unicode in email headers without having to use punycode, creating mailboxes from sieve filters and a bunch of other sieve expansions, and decent, portable, non-bugridden integration with email clients. Protondrive has such dogshit speeds it's basically completely useless. The nat-pmp support on their vpn servers is very strange, and it took me a couple weeks to craft a script that could handle all of its idiosyncrasies, none of which are documented. I haven't even bothered trying their calendar, password manager, or the Yet Another AI Service they keep sending me upselling emails for. I don't need any of those things, but I'm sure they have similarly lacklustre feature parity.

Doesn't help that when i notify them about these things, their support people just gaslight me. "I've notified our development team about this". Then nothing happens. I told them about the speed issue with protondrive when it was new, that was years ago now. Still not fixed, no updates, nada.

I will be moving to something like fastmail, plus some other vpn service, since those are the only two products of theirs I'm actually using. It seems like I'll get a far better product in both cases for almost half the overall cost.

I read this merely as not likely corresponding to any psychedelic compound known to occur in nature, not that it isn't psychedelic. Of course, even that isn't a given, since there could be some interaction between a known compound and a second compound affecting the metabolism of the first one. Although the description of its effects doesn't really sound like any psychedelic I've tried either, and I've tried quite a few

I'm still open to it being psychedelic(primarily acting on the 5-HT-2 receptor family) though. It could just be that there's enough folklore surrounding these mushrooms in the local culture to explain the very specific effects. After all, cultural beliefs are a part of "set and setting".

4 hours later, the 24h change is down just shy of 90%. To me this looks like a classic shitcoin pump and dump.

I write them down. I've been following the GTD system(using org-mode with the excellent org-gtd package). The basic concept is just that you write down things when they pop into your head. Once a day, you go through the list of stuff you wrote down and sort it into categories like "do this asap", "this is a multistep project with these individual steps that I can work on over time", "This is a habitual task that repeats on some interval. "maybe do this some day", or just discard it. I definitely get the flurry of ideas you describe. But not just ideas, mundane shit I gotta do. I'm at the store and a thought will pop into my head like I gotta clean the filter in my washing machine. Or I'm at home coding and I remember I need to buy milk. But when I'm at the store, I don't remember the milk. And when I'm at home, I don't remember the pump filter. So not only do these things not get done, they keep popping up in contexts where I can't do anything about them, and distracting me from what I'm actually trying to do.

GTD helps address that disconnect. The thought pops up, I dump it in the inbox. It is now "dealt with", asynchronously, and the thought can stop bouncing around my head. If it pops up again, I remember that I wrote it down, and let it go. Sometimes I'm not sure, so I just write it down again, no biggie if it gets duplicated.

And it also serves as a filter. So many of my ideas are just silly flights of fancy that, when I'm confronted with them even the next day, I already hate them, and just discard them.

It took me months to get used to it, and get consistent with even remembering to write stuff down. But you get better at it, and eventually you'll have less random thoughts bouncing around your head all the time. It's actually helped me be less stressed and calmed me down a bunch.

It's important to note that the system is ever evolving. It wraps in on itself. As I use the system, ideas and frustrations about the system pops up. I write them down as well, and if actionable, do something about them. So for instance, my list of "asap" tasks was getting cluttered with: groceries I need to buy, various changes to my emacs/wm/OS configuration, things that needed to get done, but were not really urgent, like "clean the shower". Obviously cleaning the shower is a good idea, but it's not like the world will end if I don't do it right this minute. This resulted in me always having more of these single tasks that I could typically finish in a day, which was distracting, because there were all these items on my agenda pulling my mind in different directions. So I herded these things into places where they make more sense. Made a shopping list and a wishlist. The shopping list gets synced to my phone and I have a widget for it via the orgzly app(which I also use as my mobile inbox). These things are not on my agenda, because whenever I'm looking at my agenda I'm not in a position to buy milk. "Config stuff" is now its own list, and I have a weekly habit item where I spend a couple hours on stuff in that list. I don't touch it otherwise and it doesn't clog up my agenda. More tedious, less urgent chores have their own list as well, and I similarly try to spend at least an hour a week on those.

Now, the block of my agenda with these "single tasks" will rarely be more than 5 tasks, most of them taking less than 10 minutes. I can easily clear it almost every day. And if I have a bad day and do nothing, the list is still manageable the next day. And ofc I made a daily habit item to that effect. That contributes to a sense of mastery, and I also get to work on the "fun stuff" without being plagued by guilt about all the other tasks I'm neglecting. Fuck yes.

Now I'm noticing another issue, that I'm getting an increasing list of these multistep projects. And I'm currently working on making a weekly review process where i prioritise projects. Ideally, I want a list of active projects with an explicit goal that I should check off at least one subtask in every active project in a week. And a procedure to manage which projects are active, not active, etc. that way I can keep my agenda free of projects that I've fallen out of love with that just sit in my agenda making me guilty.

Of course, the GTD book goes into all of this in detail. But I don't have the book. I prefer to develop this system myself, incrementally. One major issue I've had in the past with these systems is a sort of "system overwhelm", where the ADHD tendency to take things too far leads me to "commit" to adopting some ambitious, complicated system virtually overnight, and it's good for a couple days. Then I have a bad day, fail to follow it, and the ADHD propensity to see myself as a failure takes over, leading me to discard the system because I think I'm just not capable of following it. No more. Start with something dead simple, that you can do even on your shittiest days. Write some stuff down, go through it the next day. Just do that one thing. If after processing the inbox, all you can manage is to fuck off and play video games all day, fine. It's ok to have bad days. I certainly still have them. But even people with ADHD can develop habits. We're just bad at the initial phase ehere the habit isn't automatic yet. So make it at simole as possible at first, and then build on it slowly over time is my main advice.

There are transposition tables for that though. They don't store the board state actually. For Stockfish, transposition table entries are 10 bytes each, 16 bits of which are the low bits(or high? Can't remember) of a zobrist hash of the board state. The other 48 bits of the hash are used for addressing into the hash table, but aren't stored in it. The rest of the entry will be stuff like the best move found during the previous search(16 bits), the depth of that search(8 bits), evaluation(2 different ones at 16 bits each), and various bits of data like node type and age of the entry(for deciding which entry to replace, because this table is always full). Collisions can occasionally happen, but saving a full board state to eliminate them would cost far too much, since no matter how big you make the table, it'll never be big enough to cache all the board states a search visits.

In Stockfish, there will only be one full-fledged board state in memory per search thread. So the size of the board state is pretty much irrelevant to performance. What's important is reducing the overhead of generating possible moves, applying those moves to the board state, and hashing the board state, which is what magic bitboards are for.

This is why I run claude inside a thin jail. If I need it to work on some code, I make a nullfs mount to it in there.

Because indeed, one of the first times i played around with claude, I asked it to make a change to my emacs config, which is in a non-standard location. It then wanted to search my entire home directory for it(it did ask permission though).