Wes Deviers

2K posts

Wes Deviers banner
Wes Deviers

Wes Deviers

@wesdeviers

Technologist, nerd, refined rural citizan. I only post in good faith.

Entrou em Aralık 2019
358 Seguindo50 Seguidores
キャメル
キャメル@4Ud1CCuftl19105·
海外の人に聞いてみたいのですが、日本の捕鯨を批判するのはどういう理由からですか? 過去、日本では 「キリスト教的価値観から」 「鯨は知能が高いから」 「絶滅の恐れがあるから」 というのが理由だろうと考えられていました。 しかし 「キリスト教の国でも捕鯨の歴史がある」 「知能の高い豚やタコは食べられている」 「近年、鯨の個体数は増加傾向にある」 これらの事実から、日本人には捕鯨が批判される理由がいまいち理解できないのです。 私個人としても、この問題には長年頭を悩ませてきたので、よければ教えてください。
日本語
898
679
7.3K
1.3M
Amory Blaine
Amory Blaine@amoryxblaine·
@4Ud1CCuftl19105 Americans really like whales and dolphins. We have a lot of cartoons that anthropomorphize them. American standards for whether animals are food is almost entirely based on cuteness vs deliciousness. Whales and dolphins are cuter than they are delicious, so no go.
English
45
7
184
100.8K
Wes Deviers
Wes Deviers@wesdeviers·
I don't think it's possible to estimate; I've seen guesses that as high as 65-70% of American households have at least 1 firearm in the home. Broadly, we don't have registration or licensing and it's actually illegal in many cases to have a "registry." If you go look at a presidential election map, if a county is "red" it probably has > 80% gun ownership in that county. Just a guess, though. Now a lot of those are going to be like .22 rifles, or maybe an old revolver somebody's dad gave them. If you talk about "collections" that's a very different story. Funny Japanese link - one of my favorite guns is a Thompson/Center Contender, which featured prominently in Fate/Zero (if I remember correctly) and that's the only popular media I've seen it in.
English
0
0
2
317
猫パンダ
猫パンダ@pika_nekopanda·
アメリカニキって銃社会だと思うけど、実際に所持してる家庭は何割くらいなんだろ? 日本人はおそらくほとんどの人が持っていると思ってる
日本語
1.3K
88
5.1K
232.4K
Wes Deviers
Wes Deviers@wesdeviers·
Company not only pays for tokens, they will potentially ride your butt if you don't use enough tokens. Tokens are the modern "lines of code" metric. The issue is that you can double check the work but you're often using the AI to do things you're not all that familiar with the first place. The great engineers have use different techniques to work on things, like have different agents adversarially cross-check each other's work. Example - write a big spec (or just vibe code) and have Opus 4.6 do the work and then get GPT to read the spec and opine on the code that got written, then have a third model do testing. It's extremely expensive - which is the metric companies are judging on for the moment.
English
0
0
3
424
Dr Kareem Carr
Dr Kareem Carr@kareem_carr·
@BrainyMarsupial I'm curious about how this works logistically. Does the company pay for the tokens or do the engineers? Do they double-check the work or just commit it as long as it pasts the tests. Do they even write/read the tests? What do they do in the mean time? Pretend to work?
English
33
0
27
19.2K
Dr Kareem Carr
Dr Kareem Carr@kareem_carr·
I keep hearing that software engineers don’t write much code anymore and it’s mostly AI now. Can any software engineers confirm how true this is? Do you just drink coffee and watch Claude code all day now?
English
532
10
583
166.8K
Wes Deviers
Wes Deviers@wesdeviers·
@SonohennoKuma Taking food seriously is important to a subset of Americans. A large subset, but it's not universal.
English
0
0
1
72
Errol Allen
Errol Allen@SuperMarxBro·
@fandompulse I was curious about these books, but now have lost all interest. Thank you for showing us who you really are Dinniman
English
4
3
519
10.4K
Fandom Pulse
Fandom Pulse@fandompulse·
Matt Dinniman, author of Dungeon Crawler Carl, has choice words for The Critical Drinker: "I regret that my books have any association whatsoever with him. I hope one day to have his part in book three removed all together. It's a distraction, and the only edgelord douchebag I want people to think of when they read my books is the AI itself." Why did he feel the need to do this?
Fandom Pulse tweet mediaFandom Pulse tweet media
English
491
54
1.1K
1.1M
Shawn Ravenfire
Shawn Ravenfire@shawnravenfire·
It's a joke, like people who make fun of Nickelback or pineapple pizza. One person says it as a joke, and another person picks it up and runs with it, and before you know it, everyone is making fun of the same thing, but nobody can remember why. Actually, I think the "California sucks" jokes peaked in the 90s, and now it's just the old hacks keeping it going. The hipper comics moved on to making fun of Florida.
English
6
0
2
24.9K
Wes Deviers
Wes Deviers@wesdeviers·
Japan <-> USA twitter is the most OG Internet thing that has happened in a decade. It feels EXACTLY like the late 90s when you could just...talk. To people. One of my favorite ways to relax is watching no-voice Japanse short order restaurant cooking videos on YouTube... and scream in angst when 8 types of sauce go on omelets! What are you doing?!?
English
0
0
0
18
chiefofautism
chiefofautism@chiefofautism·
someone made the most ADDICTIVE game to learn DATA CENTER networking its called Data Center, $6 game, you start with bare floors, buy racks, mount servers, route every cable by hand the INSANE part, every customers traffic shows as colored balls rolling through your cables... you literally see bottlenecks in real time 180 reviews in 48 hours, people with RTX 4090 rigs are HOOKED on a $6 cabling sim
English
317
1.5K
19.9K
1.6M
Wes Deviers
Wes Deviers@wesdeviers·
There's a false assumption about American society at play here. We have a constant influx of extremely poor people who either move here for welfare, or much more commonly, move here to "make it." And they do, overwhelmingly, make it. Poor people in America don't *stay* poor for very long, except in very particular circumstances. A the society scale, once you move into "the projects" of any major US city, you are essentially doomed to poverty forever, and so are you kids, unless they leave." We already redistribute trillions of dollars to various types of poor people - less than 50% of Americans pay any type of (federal) taxes and actually make money from the federal government. But upwards mobility is real as long as you don't fall into the giveway/urban government trap.
English
0
0
1
102
ふろむだ
ふろむだ@fromdusktildawn·
アメリカ人に聞きたいんだけど、 アメリカの一人当たりGDPは日本の2.6倍もあるのに、 なんでそんなに生活が苦しい人が多いの? 高所得の人がたくさんいるんだから、再分配を少し強化するだけで、貧しい人たちの生活はぐっと楽になると思うけど、なんで再分配を強化しないの? ちなみに、再分配を強化するのは社会主義でも共産主義でもないよ。 それは修正資本主義。君たちの大好きな資本主義を、ほんのちょっと修正しただけのものだよ。
日本語
2.3K
220
2.6K
870.8K
Wes Deviers
Wes Deviers@wesdeviers·
@ryota884 These are wonderful! I'd share some of my drawings with you, but I don't have enough talent to even attempt something like that.
English
1
0
1
13
林 亮太 色鉛筆画家
海外の皆様。私は有名な景勝地やいわゆる日本的なものはあまり描いていません。5色の色鉛筆で日本のごく普通の街並みや路地を描いています。各国の普通の路地はどんな佇まいなのか、気になりますね。素敵な絵画があれば是非拝見したいです。
林 亮太 色鉛筆画家 tweet media林 亮太 色鉛筆画家 tweet media林 亮太 色鉛筆画家 tweet media林 亮太 色鉛筆画家 tweet media
日本語
72
84
1.2K
11.8K
Wes Deviers
Wes Deviers@wesdeviers·
To be convincing, it's important to have an actual example you can point to. The culture war is a battle of margins, and if a few people stumble across Devon's posting, with examples, with a personal story... it might move the margin. You, probably, already hated the left or at least saw through the technique. But maybe 10, 30, 100 people who read this post did not. This post was not for you, though perhaps it will help YOU move the margin in a conversation in the future : )
English
0
0
2
102
Devon Eriksen
Devon Eriksen@Devon_Eriksen_·
While the left was telling you that you were the world's pariah... While the left was telling you America still hated you for Pearl Harbor... While the left was telling you we saw Fukushima as the site of an ecological crime instead of a heroic battle against a natural disaster... They were telling a younger version of me that Japan looked on us with contempt. They were telling us that you thought we were racially inferior. That our facial features were coarse and ugly. That you thought we stank ("batta kusai"). That you thought of us as uncultured barbarians with no refinement or manners. That you you saw the atomic bombs as acts of calculated cruelty, rather than desperate measures in a horrific war. They showed us photographs of businesses, somewhere in Japan, with signs outside, saying "no foreigners". They posted them in anime forums online, with captions like "So you love Japan? Well, Japan doesn't love you." And I couldn't help but believe some of that. A little bit. It was the only information I had. I once wrote a very angry post defending the decision to use atomic weapons instead of invading, a decision I still believe was sound, but I wrote about it with great anger, believing as I was told, that the Japanese felt no responsibility for the war, that they blamed everything on us. I lost a grandfather in WW2, a grandfather I never met. And my father didn't know how to be a father, because he grew up without one. And I was angry. And then someone connected twitter to an AI that could turn Japanese into English. And I learned what Japanese people actually wanted to say to me: "Hey, look at this thing I made!" "We love cats! Do you love cats, too?" "Arggh! That meat looks so delicious! This is torture!" "I love bluegrass music!" "It's so cool that you guys can own guns!" "Are you mad about Fukushima? Wait, you're not? What!?" Sigh. The left tells so many lies, because they need us to feel bad. They need us to feel alone, and afraid, and angry. They need us to feel worthless, so that we will surrender and allow ourselves to be dependent on the worldwide tyranny they wish to create. But to do this, they have to tell different lies to different people. And when we learn to talk to each other, and we compare notes, we can discover the inconsistencies, and spot the lies. The Christian religion has a legend, a story of the people of Bab'el, who built a tower reaching towards the heavens, so that they might climb. For some reason that isn't really clear, god is supposed to have found this to be a threat, and responded by cursing the world to speak different tongues, so that they could not work together. Now, AI technology has undone the curse of Bab'el, and rocket technology is building a tower of fire to lift us to the stars. There are too many people in the world who are actually bad, who actually hate us, and don't want to see that progress. So perhaps it's time we stopped nursing old grudges against people who we actually can live in peace with. I'm not the only person who suffered a loss. None of us is. Perhaps it's time, instead, to defy the people who lie to us. Perhaps it's time to work together again. And rebuild that tower.
Devon Eriksen tweet media
うさこ🐰🌸@ankoromochuu

左翼はよく「日本は世界の嫌われ者」と私たち愛国者に言ってきます。 私は悲しい気持ちになりながらも、「海外でも日本を好いてくれている人はいる」と信じてきました。 Xで世界の皆さんと話せるようになり、大勢の人が日本を親しみを込めて好いてくれていると感じることができました。 ありがとう!

English
123
1.1K
5.9K
226K
Wes Deviers
Wes Deviers@wesdeviers·
@MichaelFKane It's been beautiful. Feels like the old internet captured in a bottle.
English
0
0
2
273
Michael F Kane
Michael F Kane@MichaelFKane·
The funniest thing about American/Japanese twitter crossover is how extra polite we Americans are being for our new friends. Look. I've seen some of you talking to other Americans and you are WAAAaay nicer to the Japanese. 😅 My theory is that folks have a sense that we are guests on each other's timelines and are acting accordingly. We're trying to be polite for company.
English
373
369
5.6K
133.5K
Wes Deviers
Wes Deviers@wesdeviers·
@AtlasisZephyr @kloss_xyz ... why, exactly? Kids have a lot of time, and a lot of creative thinking. What makes this "18 hour window" thing especially egregious? If it's because "kids are dumb" - you haven't been around long enough : )
English
0
0
0
3
Atlasis
Atlasis@AtlasisZephyr·
@kloss_xyz the part where they staged the fake dependency 18 hours in advance with platform specific payloads that self destruct shows this wasn't some kid in a basement, this was genuinely professional grade offensive ops
English
1
0
3
852
klöss
klöss@kloss_xyz·
do you understand what just happened to one of the most used npm packages on the internet? → axios gets downloaded over 100 million times a week and today it got compromised → an attacker hijacked the npm credentials of a lead axios maintainer… changed the account email to an anonymous ProtonMail address… and manually published two poisoned versions → axios@1.14.1 and axios@0.30.4… neither version contains a single line of malicious code inside axios itself. instead they inject a fake dependency called plain-crypto-js that drops a remote access trojan on your machine → the fake dependency was staged 18 hours in advance… three separate payloads were pre-built for macOS, Windows, and Linux… both release branches were hit within 39 minutes. every trace was designed to self-destruct after execution too → there’s no tag in the axios GitHub repo for 1.14.1. it was published outside the normal release process entirely... bypassed CI/CD completely → StepSecurity called it one of the most operationally sophisticated supply chain attacks ever against a top 10 npm package → a routine npm install silently opens a backdoor… no warning… no suspicious code visible in axios itself this is the wake up call all vibe coding bros need to hear right now: → if you installed either version… assume your system is compromised → pin to axios@1.14.0 or axios@0.30.3 → rotate all secrets, API keys, SSH keys, and credentials on affected machines → check network logs for C2 connections → add –ignore-scripts to CI npm installs going forward 100 million weekly downloads and one compromised maintainer account… that’s all it took to wreak absolute havoc and I imagine we see a whole lot more of these… crazy times ahead for cybersecurity and vibe coding be safe out there y’all
Feross@feross

🚨 CRITICAL: Active supply chain attack on axios -- one of npm's most depended-on packages. The latest axios@1.14.1 now pulls in plain-crypto-js@4.2.1, a package that did not exist before today. This is a live compromise. This is textbook supply chain installer malware. axios has 100M+ weekly downloads. Every npm install pulling the latest version is potentially compromised right now. Socket AI analysis confirms this is malware. plain-crypto-js is an obfuscated dropper/loader that: • Deobfuscates embedded payloads and operational strings at runtime • Dynamically loads fs, os, and execSync to evade static analysis • Executes decoded shell commands • Stages and copies payload files into OS temp and Windows ProgramData directories • Deletes and renames artifacts post-execution to destroy forensic evidence If you use axios, pin your version immediately and audit your lockfiles. Do not upgrade.

English
107
490
3.5K
869.7K
Humi
Humi@byteHumi·
lmao I can't stop laughing claude-code has a "Frustrated User Detection" There's a regex that detects when you're angry ( fully hard coded btw) When triggered, it changes Claude's behavior/UI state. Claude literally knows when you're cussing at it.
Humi tweet media
English
230
345
5.3K
555.6K
Wes Deviers
Wes Deviers@wesdeviers·
That'll only be good for another 3-6 month. At that point it probably won't be possible to identify true slop any more. Maybe a year. But if you're going to drastically change the way a platform work, ideally you do the big one once, then tweak it. Not try to put a finger in the dam.
English
0
0
0
6
I am Alpharius.
I am Alpharius.@i_am_alpharius_·
@ReviewsPossum Why go through all of that when they can just instantly analyze if it's AI slop and either prevent it from uploading or immediately delete it?
English
2
0
1
340
Possum Reviews
Possum Reviews@ReviewsPossum·
If you're a YouTuber, you should be aware of changes to YouTube's algorithm this year. YouTube (and other Google services) will be implementing something called GIST (Greedy Independent Set Thresholding), an algorithm that seeks to prevent users from being shown the same thing over and over again from different channels by using AI to analyze the content and determine its uniqueness based on what it determines to be your video's category, niche, sub-niche, subject, format, style, idea/information, and signature. The last two, idea/information and signature, are the most important because those are what will distinguish your content as human-made. I'll get back to that. Basically, GIST creates an imaginary "radius" around a video based on those aforementioned criteria, and will place other similar videos around it in this imaginary space. The closer another video is to the center, the more similar the video is and, therefore, the less likely it is to be recommended. So if you've seen a dozen or so channels all using stick figure illustrations while explaining the "top ten most brutal historical torture methods" or something like that, those types of videos are going to die off. Back to idea/information and signature, the algorithm will look for indications that your video was made by a human and not AI. It will see your face (if you show it), your voice, your editing style, your catchphrases, your humor, and so on. So if you use text-to-speech voices to explain something over unrelated video game footage like a million other channels do, you're in trouble. So why is YouTube doing this? Because of the influx of AI-generated and copycat content. YouTube is now flooded with low-effort content that only exists to siphon ad revenue at the expense of user experience. YouTube doesn't want users getting bored or clicking off videos because they're not providing anything valuable, so these new measure will reduce clickbait and rip-off content.
GIF
English
36
95
494
14K
ThePrimeagen
ThePrimeagen@ThePrimeagen·
This is a very apt thing I have noticed a certain malaise i am seeing. it honestly reminds me a lot of 2021 neovim config andies. You can build anything, just some time and some prompts and you got your fully custom piece of tech for exactly what you want!! but wait, here is another thing to optimize! wait... there is another thing.. oh no! your first thing doesn't quite do it the way you want it to, better fix that... ... pretty soon you are having 8 agents building all sorts of stuff and you consistently feel like you are going no where and you have nothing to show for it, yet your brain is moving at 100x the speed, your sleep is suffering, your attention with your kids and wife are dwindling... but bro, just one more prompt, just one more prompt and it will fix this issue, i swear The burden of being able to build anything is the burden of having to maintain everything and in the day and age of vibing, contracts are even less held tight and change is even faster.
LaurieWired@lauriewired

oh come on. You think your grandma wants to make her own app? Much less maintain it. Everyone neglects the mental energy it takes to even *think* of what it is exactly you want. The entire principle of apps relies on some faith that designers; and the collective feedback of their users, can come up with a workflow or design paradigm that is *better* than what you; an individual could come up with. This is true for 99% of cases. I’d be floored if even 1% of global users want bespoke applications for uber-specific needs. Not saying it’s useless; but it’s so far away from what average users want.

English
115
116
3K
236.7K
JimBob
JimBob@j1m_808·
@ronrule I mostly agree but the part about "no extra costs" isn't entirely accurate. The AI that freed up that 99% time has a cost, and so far it's been heavily subsidized by investors and tax payers. I'm starting to see rumblings that it might not be an actual cost saver long term.
English
2
0
0
47
Ron Rule
Ron Rule@ronrule·
Companies that cut staff due to AI are making a huge mistake. Where you see savings, I see opportunity. If you're profitably paying someone to sit at a desk all day, and a tool comes along that frees up 99% of their time, firing them to save on costs instead of leveraging their free time to expand into new verticals with no extra costs is incredibly shortsighted. You now have a worker, already paid for from existing revenues, who can now be turned into a profit center doing something AI can't do yet. This is a rare opportunity to 100X the productivity of your workforce and grow your top line revenues like never before. And you want to throw that out to save a few bucks on wages? That's retarded.
English
6
0
16
516
Wes Deviers
Wes Deviers@wesdeviers·
@glen_baku @f_demaku @ThePrimeagen I built a Rails app with a Svelt front end that reads an epub, vectorizes it onto PG, stores the text for retrieval, then let's you ask Claude questions about the fictional universe on the book. V1 took about 3 hours.
English
2
0
0
389