Thoughts

It's interesting, a lot of homophobia in the church is tied to this idea that gay marriage is a "perversion" of God's will, and as such,
it's worse than being single and not getting married at all. But that's not an idea that's common in Christianity. I don't want to say that it has no theological basis, because it obviously depends on how you construct your theological moral system. Like I can kind of understand saying gay marriage is a lesser form of marriage but saying it's worse than not getting married at all is really difficult to support.
Link 1:48 p.m. Apr 25, 2026 UTC-4

More tests.
Link 12:17 p.m. Apr 25, 2026 UTC-4

Another test
Link 12:09 p.m. Apr 25, 2026 UTC-4

I'm so cooked.
Link 12:08 p.m. Apr 25, 2026 UTC-4

Hello? Config test
Link 12:06 p.m. Apr 25, 2026 UTC-4

The thing that I loved about Cohost is that they understood that "Posting" was an entertaining activity in and of itself.
Social media website where you get points for posting would be awesome. Posting streak. Combo bonus points for certain emojis. Quests in the form of writing prompts. A creative multimedia experience. Typing WPM leaderboard. Oh and I guess you like follow people to see their posts or something.
Link 3:29 p.m. Apr 23, 2026 UTC-4

I want to write a post on time as a budget. Because it doesn't work.
The empirical failure of time as a resource and the need to prematurely optimize. A common accusation leveled against practitioners of high level language is that they "don't care about performance." As a long time JavaScript and Ruby programmer, this is blatantly untrue. Myself and any advanced programmer in these languages will attempt to write performant code. However, the approach that we take is far less effective than the approach taken by systems programmers. In this post I want to attempt to describe how I approach writing performant code in high level languages, its strengths, and its (sometimes major) limitations. # The time-budget feedback loop I call this strategy "time as a resource." The idea is simple. 1. You define the time budget, the amount of time your program "should take" based on your desired user experience. 2. You make decisions about the architecture of your program based on this target runtime. 3. You write your code. 4. If the actual runtime exceeds the target acceptable runtime, then you benchmark the program to determine the bottleneck (the thing taking the most time), and rearchitect the program to remove it. 5. Repeat from 2 as many times as necessary. I most often use this strategy in a web development context, since that's where I work. In this context, the target time is anchored by the network connection, which normally takes on the order of 100ms (assuming you and the server are in different geographical regions). This means that to maintain the user experience, our server-side and client-side code just needs to take around 100ms. Since I'm spelling it out, the fallacy should be obvious—my website code runtime is added to the network latency, and so to minimize the sum, the code runtime needs to be minimized, but this isn't how most web developers think. They think, "no one has ever faulted a website that takes 200ms to load as slow, so if my code runs in 100ms, it won't be slow." This isn't unique to web development though. In a game-design context, you may talk about a 16ms time budget which is required to hit 60Hz. In a robotics context, your processing just needs to be fast enough to calculate the next instruction before the robot finishes executing the previous one. In a backend context, your database may be your bottleneck, and so you may say, as long as we're faster than the database, our program isn't slow. This strategy is very effective (at least at first) and many advanced programmers use it. # Don't Prematurely Optimize I believe it comes directly from the "don't prematurely optimize" school of thought. => https://wiki.c2.com/?MakeItWorkMakeItRightMakeItFast Make it Work; Make it Right; Make it Fast => https://wiki.c2.com/?RulesOfOptimization Rules of Optimizing => https://wiki.c2.com/?ProfileBeforeOptimizing Profile Before Optimizing This motto is promoted frequently and explicitly taught to beginner programming students. And I think this is a good thing. Beginner programmers don't have an intuition for how to write fast code—they don't understand how long a network request or a database read or a single if-statement takes. They don't understand big-O analysis. Even many "advanced" programmers, and systems programmers don't have a good understanding of those timings. (There are plenty of systems programmers who "optimize" by counting the number of instructions in the generated assembly, as if all instructions take the same time.) So we tell people, don't worry about performance, write good code, then if it's slow, we can optimize it later. # The limits of Don't Prematurely Optimize Over time, over years of writing end-to-end programs, you begin to develop an intuition for what's slow. You know that an if-statement takes orders of magnitude less time than a network request, but still significantly more time than an addition. You know about branch prediction and garbage collection (even working in a high-level language) and you know what things make your program slow. On the other hand, you learn to be comfortable writing slow code in an environment where it doesn't matter, like a scheduled job that runs at midnight. After a certain point, at a certain experience level, writing slow code becomes a mistake. An advanced Ruby programmer writes O(1) algorithms whenever possible, chooses the correct data structures, and doesn't store unnecessary data or do unnecessary work. In short, a good engineer using a "time as a resource" strategy writes good code, even if their programs are still slow. # The limits of profiling At this point Profiling doesn't help. Profile a beginner's program, in any language, and it has many hotspots that can be easily refactored out. But telling an advanced programmer to profile and fix the hotspots doesn't work, because they have an intuition for what introduces pointless hotspots, and they wrote the program without any. Either there are no hotspots in the profiler, or the hotspots are the important parts. C2 calls this Uniformly Slow Code. => https://wiki.c2.com/?UniformlySlowCode > UniformlySlowCode is not necessarily optimal code for either the machine or the language at that stage. Sub-optimal, but evenly distributed, performance can result from: bad compilers, a bad mismatch between program and architecture, or an inefficient coding style for a particular language. In such cases there is little you can do after the fact that doesn't involve a substantial amount of ReWriting. # Death by a thousand paper cuts What's so interesting to me is that there are lots of advanced programmers who still write slow code, it's just uniformly slow. > The secret is that Rails apps aren't slow or hard to scale by default - they die a slow death by a thousand papercuts. => https://www.railsspeed.com/ I worry that we've turned this one technique into the goal. The should be writing fast programs. Premature optimization doesn't get you fast programs, but avoiding premature optimization doesn't get you fast programs either. # What's the alternative The alternative is Tiger Style. > The best time to solve performance, to get the huge 1000x wins, is in the design phase, which is precisely when we can't measure or profile.
Link 5:38 p.m. Apr 22, 2026 UTC-4

My fear of commitment and rejection makes me into a generalist, doing safe boring things, which conflicts with my value for in-depth
understanding and simplicity, which would have me pick a subject and become an expert.
Link 1:16 p.m. Apr 22, 2026 UTC-4

I need somebody to kill me slowly.
Link 12:57 p.m. Apr 22, 2026 UTC-4

Hello. I am alive. ZWJ my beloved. I am fundamentally unlovable. We talk about how programmers aren't exploited but when you compare us to
musicians we kind of are. I think the comparison is fair because software written by one person is used by millions of people and songs written by one person are written by millions of people. There are singers who are billionaires but practically no programmers—if you count CEOs who know how to program as CEOs and not programmers.
Link 12:29 p.m. Apr 22, 2026 UTC-4

I'm not a 100 gecs fan in particular but you have to admit Hollywood Baby is good.
Link 3:58 p.m. Apr 21, 2026 UTC-4

Note: `docker compose run --remove-orphans web sh -c 'git status > /dev/null && git stash create; echo $?'` fails (no std error, prints 1)
(IIRC with a dirty index, staged, and untracked files) when using gRPC FUSE in Docker Desktop on mac. (docker-compose.yml maps `.` to `/app`). Probably a correctness issue with Docker Desktop but would be more fun if it were a Git bug.
Link 3:40 p.m. Apr 21, 2026 UTC-4

uhghhhhg
Link 1:40 p.m. Apr 20, 2026 UTC-4

ADKSMDKLAM DLKAMS LKDMALSK MDKLAM LKAMS DLKASM DLKMAS LML DMA Ideas for how to kill myslef??????????
Link 9:55 a.m. Apr 20, 2026 UTC-4

My problem is that the perfectionistic side of me is also the side that wants to solve problems. Like. If I'm a high agency, rich, person
with vision and power and control over my own life, then I have to obviously fix problems. I have to make my bed and eat dinner and etc. But like my life is okay. IDK
Link 3:23 p.m. Apr 19, 2026 UTC-4

I'm like half way through the folding ideas Mr. Beast video and he's like, 'why do they have bad cameras pointed at the contestants and
the good camera pointed at Mr. Beast, it doesn't make any sense.' Well, you see, all Mr. Beast videos are actually about Mr. Beast reacting to the contestants and not the contestants themselves. => https://thoughts.learnerpages.com/?show=7c0599fd-d8b6-4a31-a252-71c98a5154bb He's like "why did they bring me here." and then he's also like "Why do they have this big wall with the subscriber count on 3 TVs." Like. Put the pieces together bud. The guy who cares about subscriber count is flexing on you. => https://youtu.be/Mfu_iFS-UY8?t=62 The Big Short — "Could you hold on for a second. I don't get it. Why are they confessing. / They're not confessing. / They're bragging."
Link 10:20 a.m. Apr 19, 2026 UTC-4

I don't think I can fit "shit mediocrity, inflict misery, lie our asses off, screw our customers, and make a whole shitload of money" on the
back of the T-shirt.
Link 9:53 p.m. Apr 18, 2026 UTC-4

I'm just so fucking tired all the time.
Link 3:08 p.m. Apr 17, 2026 UTC-4

=> https://ezhik.jp/losing-the-plot-that-was-never-there/ This is really an excellent post. Every paycheck I find myself thinking "wait,
I get paid to do this?" And don't get me, wrong, I hate my job as much as the next person. I'd love to sleep in and stay up late and play Minecraft for 16 hours straight sometimes and I hate that I have to show up to a cubicle for the sake of making it look like I'm working. I hate taking direction from people with less technical experience than me and I hate working on a code base with this much tech debt. That's what I'm getting paid for. But the programming? I'd do that for free.
Link 3:43 p.m. Apr 15, 2026 UTC-4

There's just no fucking clean air it's crazy I can't handle it.
My eyes burn from allergens, they're cutting the grass outside, and CO2 is over 2000 in my apartment.
Link 10:25 a.m. Apr 15, 2026 UTC-4

I just feel so bad all the time.
Link 11:35 a.m. Apr 14, 2026 UTC-4

I'm going to do it
Link 11:34 a.m. Apr 14, 2026 UTC-4

One of the reasons it's hard to post-optimize is that the code that impacts your performance the most is, inherently, doing the most.
Link 4:41 p.m. Apr 13, 2026 UTC-4

I'm going to lose my fucking mind man.
Link 9:59 a.m. Apr 13, 2026 UTC-4

Of course what's actually frustrating is that I'm self aware enough to know I make my own life miserable. 99% of the population would be
totally fucking fine with my body and bank account. It's all me.
Link 9:45 p.m. Apr 12, 2026 UTC-4

Ingress is less complicated than Hypixel The Pitt I'm pretty sure. You can't quantify this stuff but that's my feeling.
There's like 10 items in the game man => https://ingress.fandom.com/wiki/Items GregTech: New Horizons -> very complicated Veney, Vanilla Minecraft, Cookie Clicker, Hypixel: Skyblock -> complicated Balatro, Hypixel: The Pit, a Pokemon game? -> fully fleshed out Portal, I don't play a lot of games I think like Rocket League, Clash Royale -> Normal Alto's Adventure, Data Wing, Ingress -> Simple Flappy Bird, Doodle Jump -> Dead simple Okay I'm really hating on Ingress here it's probably normal. But it's like definitely less complicated than Clash Royale. Ingress players have a crazy superiority complex.
Link 7:00 p.m. Apr 12, 2026 UTC-4

So the good news is that I'm going to kill myself I guess idk.
Link 9:43 a.m. Apr 12, 2026 UTC-4

Link 9:24 p.m. Apr 11, 2026 UTC-4

"I don't think I read a book in college" -Feinberg
Link 8:01 p.m. Apr 11, 2026 UTC-4

Good news for all my haters, my life still fucking sucks. My head hurts so much
Link 10:17 a.m. Apr 11, 2026 UTC-4

"Slowtown" might be one of my favorite songs, period.
Link 2:57 p.m. Apr 10, 2026 UTC-4

Do you have teeth?
Link 10:07 a.m. Apr 10, 2026 UTC-4

> When you are having problems with a piece of software, don't claim you have found a bug unless you are very, very sure of your ground.
> Hint: unless you can provide a source-code patch that fixes the problem, or a regression test against a previous version that demonstrates incorrect behavior, you are probably not sure enough. => http://www.catb.org/esr/faqs/smart-questions.html
Link 10:02 a.m. Apr 10, 2026 UTC-4

Good morning.
Link 9:51 a.m. Apr 10, 2026 UTC-4

This is me writing in 2018
> I believe there is an absolute standard of truth, defined by God, but that we can find it by thinking critically. The Puritan standard of truth, that it comes only from scripture, is limiting, in that it doesn’t really provide for human writings or observations to be “true.” If truth is defined by the Bible, then it becomes very difficult for the general sum of human knowledge to increase. However, the Bible is necessary for providing some information, for example, knowledge of the moral standards humans are held to. However, what might be referred to as scientific observation and knowledge, is not explained by the Bible; however, it is no less true. However, unlike Pragmatism, which states that truth is only defined by how useful it is to science, I believe that in most things there is an absolute standard of truth. Or at least a set of absolute truths which can be logically and empirically reasoned to from a given set of assumptions. For example, if both a colorblind person and a non-colorblind person see the same object, and perceive it to be different colors, a Pragmatist would argue that making a distinction between the two colors was irrelevant; if one person said “blue” and meant the color they perceived one object as being, and someone else perceived a different color on the same object, and also called it blue, then there’s no pragmatic difference. However, I would say that there is a difference, and that at most only one of the people, and likely neither of them, is perceiving the true color of the object. In this way, I am not pragmatic about my views of truth, for I am unwilling to believe truth changes, only our perception of truth. Similarly, I don’t have a Modern view of truth. I value absolute truth highly, and believe that I can obtain it by reasoning, something that goes against the ideas of most Modern writers. My Epistemology is most similar to that of the Romantic period. Like Emerson, I believe that there is an absolute standard of truth that is up to each man to find himself. The notable difference between my standard that of Emerson or Thoreau is that I think it’s important to be around other people, to learn from them, and to collect knowledge from society, whereas a purely Romantic view of truth is something that can be obtained by one man, alone, in nature. A couple of thoughts. Obviously, cross reference Frosted Pane of Glass. I think the big problem with frosted pane of glass is how strongly it implies that there is some barrier which prevents us from obtaining truth. I don’t think that was the point. I think the point was always “we will have an imperfect view of truth” but only in the way that your 2 dimensional view of a 3d object is imperfect. With that in mind it’s very similar to the above. Sorry, to address the actual quoted section. I imply (I think, if I know my own writing style and remember correctly) that truth itself is elegant and good. That’s the view that is kind of challenged by GEB and which has changed over time. I would no longer confidently describe truth as elegant—I would describe it as fractal and messy. Maybe it isn’t different but it feels different to be trying to meet a simple glorious standard like “always be kind” and fail sometimes, versus trying to reach a standard which is the equivalent of “be kind to certain people on specific days and be mean to this group of people and the rules are infinitely complicated” and fail sometimes. I think it’s important to remember, this depression started back in the fall, not with epistemological confusion, but with a task that I was certain was morally and logically correct, but which I was unable to act on. I created the confusion in my brain as a defense—if being kind isn’t right then I don’t have to worry about this either.
Link 9:09 p.m. Apr 09, 2026 UTC-4

It’s like I’m logically determined that it’s correct to be a moral-less asshole. And I don’t want to do that or be wrong.
Link 8:40 p.m. Apr 09, 2026 UTC-4

I’m not an incel but I believe their beliefs, ya know.
It’s just fucked up. We defeated misogyny and replaced it with nothing. No interaction between the genders, everyone’s single forever. We have this fucked up version of feminism where the man still has to ask out the woman 100% of the time but now if she wants she can report him for sexual harassment and ruin his life. Our society is about seeing the worst in people all the time. It doesn’t matter why you actually believe if you break from the societal consensus and expectation and say something “maybe a woman should ask me out” that makes you a mysogonist because you’re not normal and fucking 14 year old on Reddit don’t know what mysogonist means. But I’m not an idiot. I’m cursed with intellect self awareness. I know I sound exactly like an incel so I want to kill myself.
Link 8:38 p.m. Apr 09, 2026 UTC-4

I don't even think this is a strength of mine, I think it's a weakness, but I would much rather spend 5 hours writing a patching I
can send upstream than spend 5 minutes changing a setting locally. "turn on this setting to work around the bug" you may as well be speaking French.
Link 4:24 p.m. Apr 09, 2026 UTC-4

Good morning.
Link 3:35 p.m. Apr 09, 2026 UTC-4

How the actual fuck are you supposed to develop software targeting unix if you can't fucking read the fucking standard.
Link 12:05 p.m. Apr 09, 2026 UTC-4

Explicit error handling makes the code better, the UX better, and the debugging experience better.
Link 10:53 a.m. Apr 09, 2026 UTC-4

Good morning.
Link 9:48 a.m. Apr 09, 2026 UTC-4

I said “Truth is fractal” a long time ago but I’m only just realizing that increasing levels of accuracy require increasing amounts to data
at least to share, which isn’t true for lies. I feel like I should have dealt with this as like a 12 year old because it’s really obvious. I had this naive idea that truth was something you could obtain through easy learning or thinking or logic when obviously if you put in more work you get more truth. It means that “maximum truth” can’t be a goal because that implies maximum work. I feel like I should read more of my old writing because I definitely had ideas around this. If I didn’t have this thought it’s only because I would have disagreed with it, there’s no way I missed it.
Link 6:26 p.m. Apr 08, 2026 UTC-4

T-shirt ideas for if I had balls: - Uriel was right - Fuck the Digital Millennium Copyright Act
- what makes life worth living is the fact that stereotypes aren't true, It's the complexity of life / (back) except for this one. There has been no entity in human history with less complexity or nuance to it than Oracle
Link 6:04 p.m. Apr 07, 2026 UTC-4

It's so weird to look at apple.com and think that they should charge more money. I bought a refurbished MacBook Pro in 2020 for $2,000.
The MacBook Pro today starts at $1,699 and completely mogs that 2019 laptop in every metric. Same RAM, M5 instead of a i5, new, and 300 dollars cheaper. For $1,699 you can get a Lemur Pro (System76) with half the RAM. Which means that a Mac laptop is cheaper than a Linux laptop with the same specs. What's going on? I'm on Lenovo's website and I can't even focus on the laptops because they are using fucking ^ < > for their arrows. I used a greater than sign for a right arrow when I was 14. Why don't they have a design team? If you click on "X9 Series" and "View X9 Series Laptops" you get "Sorry, there are no results." Do people not buy laptops online???? Microcenter has some laptops with the same specs for $699 so maybe it's just System76 being weird. But $699 is really cheap.
Link 3:28 p.m. Apr 06, 2026 UTC-5

Soooooo many beees in my head
Link 2:53 p.m. Apr 06, 2026 UTC-5

“Oh it evolved to escape” I fucking hate Darwinists what the fuck is this horseshit.
Link 1:20 a.m. Apr 05, 2026 UTC-5

Project Hail Mary is about how single autistic people have to be ritually murdered by the government in order to keep society functioning.
Link 1:18 a.m. Apr 05, 2026 UTC-5

Project Hail Mary (movie) was fucking awful. None of the science was explained at all. Polar opposite of The Martian. I’ve read fantasy
books with better science. The bridge is multi-use but the bridge retractor is single use so you have to use the magical bridge cutting gun. We’re 12 light years from Earth but it’s going to take 4 years to get there. We’re traveling near light speed but no one is going to do a relativity calculation for the audience because no one cares about engineering or science or numbers. “I don’t have enough to fuel to do both” just go slower there’s no friction in space what are you talking about. I’m sure the math in the book was right and they just cut out all of the everything for the movie but it’s so insulting and I thought we proved that audiences can handle hearing a real engineering thought process but no, let’s show the scientist’s hangover montage and show absolutely none of the actual scientific method at any point.
Link 9:27 p.m. Apr 04, 2026 UTC-5

A Count of Monte Cristo adaptation where the Count is a black man would fix society I think.
It just hits so many threads, so many different angles.
Link 5:53 p.m. Apr 04, 2026 UTC-5

People hate hate hate hate hate hate hate hate being corrected and so if you notice them making a mistake, you should say nothing ever.
Link 11:46 a.m. Apr 03, 2026 UTC-5

Oh my poor brain.
Link 1:39 p.m. Apr 02, 2026 UTC-4

I'm going to lose it. I'm going to fucking lose it. The amount of bad code I have to put up with on a day-to-day basis has driven me insane.
I can't be normal about it. I can't fucking do it.
Link 10:24 p.m. Apr 01, 2026 UTC-4

The worst thing about being good at something (programming) is that I can't turn it on at will. I can't pick the problem to magically have
insight into. I still write plenty of bad code and I still misunderstand plenty of code. All "being good at programming" means is that sometimes I look at a piece of code and I just see that there's something wrong with it without really trying.
Link 4:06 p.m. Apr 01, 2026 UTC-4

underscores is already sold out. Gosh heck. Should've been on top of it. Uhghhgh
I looked like 6 months ago, pretty much right after the last tour and now the new album is popping off and I wasn't waiting for the minute tickets opened so I'm cooked. Fricking skill issue. I need to check all the venues and see if I can slide in somewhere.
Link 3:30 p.m. Apr 01, 2026 UTC-4

Good morning.
Link 12:33 p.m. Apr 01, 2026 UTC-4

On one hand, it's probably my depressed mind warping my perception when I say that no one appreciates me or wants me to exist, but on the
other hand, I shouldn't need strangers to design the world to accommodate me in order to stay alive. I was treated kind of rudely by a Starbucks worker earlier today; I didn't order anything but I tipped a couple of dollars and asked to use the bathroom. Obviously we can play the AITAH game, and I did at first, but the thing that transcends that is the systematic issue. You are expected, in society, to drink coffee. And you are absolutely expected to drink coffee if you walk into a Starbucks. I don't drink coffee and I don't plan to start, so Starbuck is not designed for me. I need to be okay with that. But it just tears me up. I'm trying to live my normal life and not drink coffee and it sucks that my normal life is so abnormal and weird and freaky for many people. Because it leads to friction. It leads to "was I rude to this customer service employee" because I'm asking them things that are off-script for them. I need to be comfortable making other people uncomfortable. It's hard letting go of little conversations. I want the customer service employee to think I'm cool and friendly and nice. Maybe I need to channel my Dale Carnie more and ask them about their hobbies or something. Like maybe I need to do a better job at it. As someone who aligns with society's standards a lot (well dressed white man) I get what I want 70% of the time just because I look normal and express myself normally and I don't need intentional charm because people assume I'm acting in good faith. (I wouldn't even call this privilege because everyone should be treated like this.) But sometimes I'm outside of the norm and I need to either explain myself with more eloquence, or, be okay with being misunderstood. There are times when I'm doing something a little unusual (climbing a tree, sitting on the floor, programming for fun on vacation, walking into the Starbucks and not ordering anything), and someone asks a reasonable question, and I take offense because it reminds me that it's not normal for everyone. (Of course, there's no "societal normal." Normal is different for everyone which makes preempting this impossible.) Ref: tomato party Ref: vic bit I didn't post about: on AHA where she's like "I make people uncomfortable? me?"
Link 10:50 p.m. Mar 31, 2026 UTC-4

I think I need to subscribe to a magazine. I need a non-internet source of novelty. Pretty pictures shipped direct to my mailbox monthly.
Link 10:17 p.m. Mar 31, 2026 UTC-4

It would be awesome to build a new database. There's an elegance to relational databases, but they're very difficult to build on top of
correctly. Data races need to be fixed with locking and locking kills your parallel throughput. ACID (as implemented by major SQL-based databases) somehow makes both performance and correctness difficult. SQL is actually nice; relational tables and normalized data are nice.
Link 4:52 p.m. Mar 30, 2026 UTC-4

Druidic. Tgvbug his by. T CB he y g bf c
Link 2:05 p.m. Mar 30, 2026 UTC-4

Ideas on how to kill myself, ?
Link 12:47 p.m. Mar 30, 2026 UTC-4

I definitely have autism
Link 12:20 p.m. Mar 30, 2026 UTC-4

You don't need to be angry or stressed to get stuff done. You can work consistently, steadily, and quickly without stress or anger.
Link 4:31 p.m. Mar 27, 2026 UTC-4

People post stories of them guessing quantities exactly to the nearest <unit> as if we, humans, didn't invent <unit> based on how precisely
we can perceive it.
Link 4:50 p.m. Mar 26, 2026 UTC-4

It's so annoying how the Claude closed-source cli tool is priced completely differently (cheaper), than using Anthropic models with an API
key and a different tool like OpenCode
Link 4:08 p.m. Mar 24, 2026 UTC-4

TiNAD is so relatable because I woke up at 7:00am today and I have no idea what I did for the last 3 hours. Like where did my time go.
Link 9:50 a.m. Mar 24, 2026 UTC-4

I just hate arguing because there's never any happy ending. Like even if I agree with you that it's a problem that Gemini is fundamentally
racist you're not going to do anything about it and I guess I just need to shut up and not talk about Gemini ever again and take down my Gemini capsule. Like what's the point. What's the goal. Sorry.
Link 4:29 p.m. Mar 23, 2026 UTC-4

I'm trying to balance being chill with being productive. High executive function not high strung, ya know
Link 1:48 p.m. Mar 23, 2026 UTC-4

Good morning.
Link 11:14 a.m. Mar 23, 2026 UTC-4

Finished TiNAD yesterday, making it a one-day read. I had previously read part 1 online (but reread it because it’s good). 5 star book.
A couple of weird “potholes” where the printed book refers to the SCP Foundation but can’t refer to it by name. Also, a couple of pretty hand-wavy explanations, but I don’t mind suspending disbelief. It’s a story about fighting monsters that can tinker with the concepts in your mind. It’s hard to explain because there’s not a single system. Sometimes they’re basically invisible, by virtue of being impossible to conceive of, but you can remember that you saw *something*. Other times you can see them fine, but forget that they exist after. Sometimes you have to apply amnesia to yourself, because if you remembered it would kill you. Sometimes forgetting means having false memories, sometimes it means forgetting just the monster, sometimes you have no memory of the time period and no idea that anything happened at all. It’s, in a way, it’s similar to my time-sickness. There’s an existential horror in looking back in your memories and realizing you’ve forgot about a time in your life, an aspect of who you used to be, which doesn’t require supernatural memory loss to experience.
Link 9:59 a.m. Mar 23, 2026 UTC-4

Crying a little bit reading *There is no Antimemetics Division*
It just makes me emotional to see such valiant shows of emotional fortitude. And it also mirrors my struggles with depression. You relapse. You forget how to be happy, how to have hope. It’s too late for my life to be perfect. I’ve already lost hope in the past and there’s no way for me to go back in the past and get it back. I need to rebuild my life. If strength is staying standing up then I’ve failed because I’ve been knocked down, but if strength is standing back up then I can still be strong. The characters in the book keep losing their memories and they have to rebuild and keep going.
Link 8:04 p.m. Mar 22, 2026 UTC-4

People who use Ad Blockers are like contagious but asymptomatic patients.
Without knowing it, they upvote, share, and click onto ad-infected sites, allowing these sites to survive and spread. My ideal ad blocker would: * replace ads with a banner [AD BLOCKED] instead of operating invisibly * show a warning before clicking through to sites with a lot of ads (which would include options to visit the site with or without ads) * flag links to ad-infected sites in search results and link aggregators Critically, reducing traffic to sites with ads would encourage sites to remove or reduce the ads. With traditional ad-block, there’s no moderation—it doesn’t matter how unobtrusive the ads are, they’re all blocked. Site owners obviously aren’t happy about this but there’s nothing they can do about it.
Link 5:08 p.m. Mar 22, 2026 UTC-4

I'm just so reserved with my opinions. I have opinions on everything but I try to only share them 1) after hearing out the other person's
views, and 2) in areas where I've put in considerable thought. And the problem is that this attitude invites people to preach to me. People start assuming I don't have opinions. I'm interested in understanding what they're saying and they're like, okay let me convince this person. Which just feels so invalidating. Because I actually have my own opinion. I'm letting you share first but that doesn't mean that I'm permanently waiving my right to have a belief. But I don't like talking so I intentionally talk after other people have talked and only say things that haven't been said. I don't feel the desire to be the main person expressing the consensus opinion, I want to bring something new to the table. But that means most of the time I'm bringing something to the table that doesn't align with the consensus opinion. People assume that since I didn't rush to jump in and start talking first that I must care. I don't feel the need to convince *you* but I still have a strong belief. But I don't know, maybe I don't have a strong belief and I have no opinion and the real men can talk without me and it's fine.
Link 4:34 p.m. Mar 22, 2026 UTC-4

I think updating Zed may also cause you to lose unsaved files without any warning prompts, but I can't update on demand to test
Link 3:50 p.m. Mar 18, 2026 UTC-4

This is the most BS I've ever freaking seen man => https://github.com/zed-industries/zed/pull/49262
Yeah yeah let's add this 20-star LSP that no one in the history of software has ever asked for or used which does exactly the same thing as prism, the official parser of ruby, except worse, because it uses an older syntax reference, and it's enabled by default for me because I configured Zed to use ruby-lsp using the provided example in the Ruby extension for Zed documentation. The maintainer clarified of course that "The expectation is to look" at the general "Working with Language Servers" page, which of course has the list of supported Ruby language servers. Of course. What IDIOT would configure their Ruby LSP using an example from Ruby LSP docs page??? It's just so stupid top to bottom. There's just no reason for kanayago to even exist.
Link 3:13 p.m. Mar 18, 2026 UTC-4

I feel, sometimes, like the guy in 1984 who is enthusiastic about developing Newspeak. Like I've missed the point in some fundamental way
Link 2:02 p.m. Mar 18, 2026 UTC-4

I’m having an identity crisis, to a large degree
Link 1:36 p.m. Mar 18, 2026 UTC-4

Good morning.
Link 12:40 p.m. Mar 18, 2026 UTC-4

Good morning.
Link 12:22 p.m. Mar 18, 2026 UTC-4

Ideas on how to kill myself? ?? ? ?? ? ?? ? ?? ?/ ? ? awlkdmn ajk dnkan djkan djakn skjdnjkan skjdnk jankjs nd
Link 11:16 a.m. Mar 17, 2026 UTC-4

I’m going to fuckijg kill myself
Link 9:12 p.m. Mar 16, 2026 UTC-4

I want to be able to like people but everyone I see is an idiot and loser.
Link 5:27 p.m. Mar 16, 2026 UTC-4

One of the reasons it's so hard to rationally determine truth in real life is that in math and logic, it's possible to reason about
"all possible options" in varying senses. As Sherlock Homes says, "when you have eliminated the impossible whatever remains, however improbable, must be the truth." And of course, that's just not true, because it's also possible that you've missed some possibility or that your framing of the question is wrong. Just because every argument you've heard in favor of a principle is bad doesn't mean the principle is bad. Just because there are aren't any remaining options that you can see doesn't mean there aren't more options.
Link 4:28 p.m. Mar 16, 2026 UTC-4

I'm so fricking tired.
Link 3:14 p.m. Mar 16, 2026 UTC-4

I'm going to lose it.
Link 12:53 p.m. Mar 16, 2026 UTC-4

My kingdom for an IDE that detects that when over 90% of the lines on screen have errors and disables lints.
Link 12:49 p.m. Mar 16, 2026 UTC-4

"Headshot" is good but it's not "Picture Perfect" (Lydia Night songs)
Link 12:28 p.m. Mar 16, 2026 UTC-4

Integrated Dynamics has to be the craziest minecraft mod. It's just Haskell. It's just a point-free functional curried programming language
A screenshot of an item in Minecraft, the tooltip describes it as, "Variable Card, Type: Operator, Signature: Any -> Any -> Boolean, Value: Relational Equals".
Link 11:59 a.m. Mar 15, 2026 UTC-4

Of course libgit2 revwalk is O(n^2) on the number of commits being walked because that makes sense. How do you mess up traversing a tree?
Link 11:29 a.m. Mar 13, 2026 UTC-4

People don’t want the world to be better.
I have this terrible curse: I’m always capable of imagining things as better than they are. But people don’t want change or they don’t trust me or they don’t want to do the work or they can’t imagine what I do. So it’s a gift that’s completely useless except to make me miserable.
Link 12:42 p.m. Mar 12, 2026 UTC-4

Good morning.
Link 3:11 p.m. Mar 11, 2026 UTC-4

Good morning.
Link 12:48 p.m. Mar 11, 2026 UTC-4

It's a good morning for a browser bug. Holy moly.
Link 1:56 p.m. Mar 07, 2026 UTC-5

"calc is short for calculator" is such a good bit because anyone can do it. Everyone has a pretentious teaching voice. "Im just using slang"
Link 11:18 p.m. Mar 04, 2026 UTC-5