atraac

While I also dislike LinkedIn a lot, I unsubscribed once in the past and I haven't received a single email like that in years.

Youtrack was made long before Kotlin so I suspect the rewrite just wasn't worth it

Absolutely everyone who doesn't work with hardware, or on specific problems, because it's far more popular, easy to hire for, easy to deliver an MVP with and easy to have people jump between frontend and backend with. I'm saying this as someone who specialized in .NET before and now writes Node/TS. I simply grew out of the opinion that language/architecture is everything. When you make a product delivering quicker matters more than choosing a marginally more performant language, unless you're solving a very specific problem that requires that performance. Adding one more server costs less than hiring engineer for a more niche language and if you have the success that requires you to scale up that much, money is probably not the problem anymore. Granted we're in LLM era where everyone can write anything but it's still better to have people vibe in language they actually know, so they understand what's happening. You can always extract services later on and rewrite them in a more performant way, noone's stopping you.

I personally like Codex much more as vscode integration. It's actually nice to use, looks great, handles images much, much better, can even send me images in the chat to see what it's doing(my side project is based heavily on image processing), and what's most important to me is that /steer actually works. When I type something to Claude mid-task, it'll maybe acknowledge that within next few minutes, although it seems to be actually quicker last few days but still takes a minute or two, whereas Codex will almost instantly read it, switch what it's doing or answer me. It feels much more polished in some ways(though usage page in settings almost never loads for me).

I disagree. I needed a small text to json model that would parse basic info into a json, nothing else. I instantly knew to start with nano and if not good enough, use mini. This was obvious to me, just looking at the name. Now you have to actually know what those names mean. And I can guarantee you they'll add 5 more to create more hype, or make new names in whatever the next-gen-world-breaking-dangerous-model will be.

Nothing. It had marginal gains. People just romanticize it cause it's gone.

I have the same feeling. I'm even trying to argument to myself upgrading my phone to something newer that would support AI better but I genuinely can't. The only use case I could ever need is maybe(!) using it to fill up shopping list out loud but at this point it, using shared Google Keep it's roughly the same amount of work to just type it. In professional matter though, having an AI agent on Slack, that can query readonly dbs for us, check things, debug problems reported by our users, it's pretty great.

Website catalogues exist for more than a decade. They were one of the first attempts at gaming SEO.

We forgot already that getting paid for your work is your right and LLMs were trained on stolen property?

I work in crypto and this is happening practically every other day. I refuse anyone on LinkedIn that I don't know personally and has web3 or crypto anywhere in the description. It's all fake accounts with fake job offers. It's a pretty known scam.

Unless your job is purely producing code pointlessly, this is not a really good comparison. Most of the time really is spent on understanding the problem and figuring out solutions, not waiting on CPU.

How would getting rid of postinstall break patch-package? If people use a package, and that package needs some kind of step to get working, user of that package should decide when that step happens. He can very well just call patch before building on his own. There's zero issues with that approach and the upside is he actually has control.

I work in a monorepo where running install calls dozens of deeply nested postinstalls of some elaborate NextJs or React Native dependencies other projects use. It's borderline insane. Unless you regularly screen everything, it's impossible to know whether one of those is compromised, especially in the world of Node where is-even is being used and the sheer amount of crypto scams around.

postinstall scripts should've been removed long time ago, it's the cancer of NPM packages. There's so many deeply nested, uncontrolled postinstalls that run randomly when you pull something it's insane, I don't know how someone at some point ever though that was a good idea.

These laws only apply to megacorps. It's not an existential risk to them, as Apple is clearly proving now.

Who is saying that enforcing companies to open their systems to competition is making them mediocre? Maybe if that's the end result, they should put more time into designing systems that wouldn't become mediocre just by allowing third parties to do things with those said systems? We need to stop defending corporates for abusing their monopolies.

We use almost exclusively Valkey now, mostly because we host on AWS and Render, which both use Valkey. It's faster, cheaper and compatible. I'd consider Garnet too but I believe it doesn't support LUA(or didn't at the time we needed it).

I've used one of the flagship Surface Books two jobs ago, really beefy, could run Cyberpunk(don't ask how I know). Insanely bad machine, the overall finish was terrible, hinges broke often, charging port broke often(like every 3rd-4th person in the office), it overheated with basic office work, fans went rocket level loud when compiling basic things, after a while it was stuttering even with casual use. Had a really good Dell XPS at home too that was dying when compiling big Typescript codebase. I bought an MBP M3Pro 1.5y ago and I could never go back. It's insane how far ahead Apple is when it comes to performance. Noone seems to grasp that until they actually try to work on M-series. Three of my ex colleagues also switched to Macbooks after I told them, they also are never coming back.

My bet is they won't. It's hype driven development. I work in the space(we build one of the bigger exchanges based on Hyperliquid) and our design/product people are spasming at the thought of releasing MCP/Openclaw skill for trading. I'm 99% sure it will all be a flop, month from now noone will ever know these exist but this is what everyone in that space is doing right now, quite literally, everyone. Not a single sane person will give meaningful amount of money to LLM for actual trading.

Your users don't have to use those extensions, so I don't understand how that's relevant? People who do, should be made aware of risks and that's it. This is not a good argument against taking away their option to have that customization.

But 100$ Claude subscription also gets me easily entire week of coding 6-8 hours a day? What on earth do you do to run out of limits on Max? Do you vibe multiple new codebases every day for a living? The benefit of Claude is also not gaslighting me every time I tell it it's wrong.

12% for software development, 8.5% for design/management. The caveat being, you can't deduct anything from tax, only VAT(under some assumptions). If you have actual expenses it's 12/32% progressive or 19% linear tax. Of course all of those are assuming you own a one man company and work B2B. Most devs here do. Otherwise regular contract of employment is progressive 12/32% tax, plus Healthcare and employer payments. Much less beneficial to both sides hence why it's not preferred by most.

I'm Polish, working for globally remote companies. I second the communication issue. Most Polish devs are so ashamed of their english(even if it's perfectly communicative) that it makes it hard to discuss technical ideas with them. As for technical knowledge, I guess that's cognitive bias, most Polish devs I met were far better at tech stuff than most f.e. Germans I worked with.

Anecdotal evidence but last week or two Claude changed something related to their quotas. I'm a Pro user(now Team Standard) and while I did quite a lot before with that subscription, past week the 5h quota quite literally lasts maybe 5 semi sized prompts. I don't "vibe" anything, I give it clearly defined tasks or things to debug/fix, nothing hardcore. I ran out of the quota every single day past week, often twice a day, this never happened before. It's rather unusable for actual professional usage now. I'm tempted to test Codex over next week to compare hence why we're waiting with going to Claude Max sub.

Was in the same boat. I ended up not using Zed because it had a bunch of minor quirks that annoyed me but I moved to vscode. I primarily write Typescript and C# these days. I was a JetBrains fanboy for years and it feels way too bloated now, stuff notoriously hangs or takes too long on my M3 Pro. I also love Claude Code integration with vscode just a bit too much to give it up for CLI.

I use Pro professionally and didn't hit limits most of the time. I believe I used up 5hr quota once or twice. We switched to Team sub and I'm on Standard(which is Pro x1.25 I believe). I don't vibecode entire applications, I ask it to make boilerplate, smaller, well scoped features or fix some errors. I don't let it go off with a prompt "make another netflix clone" cause I just don't see any real value in that

What does this even have to do with the thread? They're hosted there cause it's cheap and extremely easy to do so. Not because it's "specially crafted" for scams.

Default double shift in zed goes to Execute a command window, sadly it doesn't support file searching within the same window and that's what I suppose he meant. I know cause I'm an avid JetBrains fanboy too and I can't use any other IDE because of that and few other features. Also, they're working on Search Anywhere Jet-Brains style feature right now if you believe Github Issues.

With design just being more sensitive and more overlooked area

My take is that design stopped being for the user. It's a rat race on who will make it look the most fancy or closest to whatever new trend AirBnB starts as quickly as possible. More animations, more colors, darker, lighter, let's make this transparent. Noone really thinks about UX anymore, it's all just portfolio-driven-design.

In my case, I use JetBrains Junie(it uses various models underneath) and it mostly works fine, but I don't vibe code entire products with it, I just give it easy, neatly defined tasks. Improve readme, re-implement something using the same approach as X, create a script that does Y etc. It's fairly good at making one-off tools I need, we messed up something and need a tool to f.e. fill up db with missing records? I'll just make a console app that does this. We need a simple app that will just run somewhere and do one thing(like listen to new files and insert something to a db). It's perfectly fine for that. I wouldn't trust it with day to day job, features, anything more advanced(I do mostly backend work). I also have to verify thoroughly what it ends up doing, it tends to mess up sometimes but since most of what it does is non-critical, I don't mind. I too don't believe the claims of people who just straight up Claude they way through a codebase to do 'grand' things but I have a small sample.

I've learned that being truthful in your resume as a SWE doesn't work anymore. I've had nearly 0 response rate to my applications, even if I was nearly a perfect match. Dude that we fired a while ago for being abysmally bad and non-productive gets interviews, raises and just got into ycombinator backed company by making up 75% of his resume(which I saw). We need a reset, this is getting ridiculous.

I write mostly backend stuff for a living, big chunk of it in Node/TS but also C# with modern .NET. I also have to dabble with Unity and Unreal both for work and a hobby project. I technically learned C++ in uni but really, I hate every single second I have to spend doing Unreal Engine work. I genuinely despise how obsolete and hard to write C++ is compared to modern languages. It didn't bother me in university because that was all I knew at the time but it's hard to justify existence of header files today. Add macros everywhere, really bad compilation errors, really hard to read syntax with a lot of unnecessary bloat and you get something that is just not likable. I'm sure you can get used to it given enough time spent in that ecosystem, but I can tell you as someone writing 4 different languages on day to day basis, C++ is difficult, and it's not because of pointers.