corytheboyd

Sorry, I often conflate flex and grid, they occupy the same spot in my mind as “1 dimensional” and “2 dimensional” versions of the same underlying thing. I am wrong.

This sounds crazy to me, the flex system is exactly this but also semantically correct. You can’t possibly be telling me that table hacks of yore are easier than this, a clearly designed system for the actual problem you want to solve https://css-tricks.com/wp-content/uploads/2022/02/css-flexbo...

I’m not at all a frontend person, I just think flex is one of the best systems I have ever used. It usually gets the job done in a way that I expected to work (and that bar is pretty low lol)

I have been wondering when I would start to feel aged out of the tech industry… gosh is it here already?

> The promise of "parallelism" is overstated.

100% my takeaway after trying to parallelize using worktrees. While Claude has no problem managing more than one context instance, I sure as hell do. It’s exhausting, to the point of slowing me down.

It’s so dumb. There is no way to verify the numbers, and yet, this stupidity weaseled its way into the LinkedIn cinematic universe of corporate bullshit. The same point but without the “X by Y%” hits the same for me— besides I know what questions to ask to judge if you are actually capable of moving the needle, which is all I care about as a conductor of interviews.

Mention of Perl will always reminds me of that time I put Perl in a PR at a Ruby shop. It was a beautiful, portable little piece of code that they made me take outside and shoot in the back of the head. Fair enough, but I will never get over it lol. I don’t even write Perl scripts for personal work anymore, I use python (only because uv makes it not awful), but it’s not about that!

Because my employer tells me it’s fine so long as we’re using it with AWS Bedrock. Do I believe that Amazon then wouldn’t be siphoning all this as model training data? It’s a fun conspiracy…

Yeah me too, keeping the code modular being a huge part of it. In the same vein is actually finishing feature flags and removing the old code, killing off truly dead code, tests, documentation… basically eliminate as much noise as you can, so that it reads as much exemplary code as possible.

> it shows you the dollar amount spent which is a bit disconcerting

I can assure you that I don’t at all care about the MAYBE $10 charge my monster Claude Code session billed the company. They also clearly said “don’t worry about cost, just go figure out how to work with it”

Been in a weird position where I had to learn Java real fast, then learn Kotlin real fast to replace it. I find Java to be so incredibly unproductive compared to Kotlin, for all the reasons a sibling comment listed. I mean come on, default argument values and named input parameters are table stakes— it kills off the billion overloads, billion positional arguments, annd builder pattern nonsense all at once. Kotlin immutability by default is fantastic, I am 1000% behind default immutable data structures. The stupid, awful nullability madness disappears (mostly, need to train people to never use ! and !!). I appreciate Java, it got us to Kotlin, but Kotlin is ACTUALLY FUN to use, and expresses all ideas I have cleanly. Less, cleaner code just makes things easier to maintain, full stop.

I’m very effective with JetBrains IDEs, VSCode is not in the same league. I’m very effective with a terminal. Claude Code is just enough of a bridge between models and my IDE to enhance my workflow without ruining my IDE. That’s it. I don’t pay $200.

Thank you stranger <3

Yeah I am a moron. Updated.

Well thats great, guess I have carried that baggage with me as misinformation for years now.

I’ll just kill the comment. It said Symbol isn’t garbage collected. It has been since 2.2 and I wasn’t aware. Sorry.

Good reminder that anyone can go on the internet, just say stuff, and be wrong.

Oh you’re assuming I just go through the motions. I do care about the craft of my work. Eh okay.

I ask myself a lot of things, but they all require money to do. This is just the society we made, you do some shit that doesn’t matter for a few decades, get some money, then go do whatever you want. I’m fine with it. Would I rather not do it? Absolutely!

> If you are in a toxic work environment, and you convince yourself it's not so bad, you might end up staying in a situation that is actively harming you.

I hate a toxic work environment too, but let’s be real, everything is fake at software companies, so there is no “harm” happening if you stay. So what if leadership is terrible, if you can play their little games, take their money, and not tailspin into a depression… congrats?

Very, very perfect, I love it

Ah yeah Í keep seeing it come up, I will take a look! This weekend was gonna toy with C++ to rebuild what I got done with MonoGame, but a next mini experiment can perhaps be Odin!

Agree, the recency sort sort ends up normalizing history over time anyway. Use fzf with your history to see more than one match a time anyway, and you can at least glance through a few of the iterative commands you went through if your history isn’t “normalized” yet. I like keeping shell history dead simple— it’s a stack of executed statements. Period.

I do pull very often run things into functions and source them into every shell session. Usually only parameterized things that aren’t trivial to use reverse search for.

Hah I just recently was doing some soul searching for a game dev environment to get started with. I did a wee bit of 2D game dev in college 15 years ago, and wanted to scratch the same itch.

At this point, I demand a good IDE experience: as much intellisense as possible, great debugger, hot reloading.

Anyway, C# and MonoGame was what won at the end of the day, at least to just get the ball rolling. Rider is a great IDE, and MonoGame… at least has great getting started documentation to get me going again. Once I knock out a few cutesy games I’ll probably go build my own thing on top of SDL3, so that I’m not held back by MonoGame. Also the MonoGame content builder is just completely broken on Mac, so on the side I am building my own Rider plugin to bring actually good intellisense to the mgcb file, so that I don’t need the broken editor gui (all it does is manage the mgcb file anyway, arguably should have just been an IDE plugin the whole time)

I tried Dragonruby, really wanted to like it, but the development experience is just… not there. It live reloads, but seems to offer no IDE tooling whatsoever. They talk a lot about emacs, maybe you’re supposed to use it with emacs, which I just have zero interest in doing. You’re also not using MRI, or the full Ruby language, so a lot of tooling for “normal Ruby” may not work. C# is just C# with MonoGame, I can use all the tools, import all the nuget packages (if you really want to), etc.

I felt like going with a C based project would lock me out of having hot reloading, but that might not be true, please let me know if I am wrong!. I know of a C++ hot reloading thing specifically made for game dev, it’s in my notes somewhere, just not sure if it’s suitable for use (you sure do have to sift through lots of abandoned experimental stuff in this world)

> Just last week i asked for a script to do image segmentation with a basic UI and claude just generated that for me in under 1 Minute.

I agree that this is useful! It will even take natural language and augment the script, and maybe get it right! Nice!

The AI is combing through scraped data with an LLM, and conjuring forth some imagemagick snippets into a shell script. This is very useful, and if you’re like most people, who don’t know imagemagick intimately, it’s going to save you tons of time.

Where it gets incredibly frustrating is tech leadership seeing these trivial examples, and assuming it extrapolates to general software engineering at their companies. “Oh it writes code, or makes our engineers faster, or whatever. Get the managers mandating this, now! Also, we need to get started on the layoffs. Have them stack rank their reports by who uses AI the best, so that we are ready to pull the trigger.”

But every real engineer who uses these tools on real (as in huge, poorly written) codebases, if they are being honest (they may not be, given the stack ranking), will tell you “on a good day it multiplies my productivity by, let’s say, 1.1-2x? On a bad day, I end up scrapping 10k lines of LLM code, reading some documentation on my own, and solving the problem with 5 lines of intentional code.”

Please, PLEASE pay attention to this details that I added: Huge, poorly written codebases. This is just the reality at most software companies that have graduated from series A startup. What my colleagues and I are trying to tell you, leadership, is that these “it made a script” and “it made a html form with a backend” examples ARE NOT cleanly extrapolating to the flaming dumpster fire codebases we actually work with. Sometimes the tools help! Sometimes, they don’t.

It’s as if LLM is just another tool we use sometimes.

This is why I am annoyed. It’s incredibly frustrating to be told by your boss “use tool or get fired” when that tool doesn’t always fit the task at hand. It DOES NOT mean I see zero value in LLMs.

I genuinely cannot tell if this is satire.

I’d love to skip to the end of the AI capitalism any% speedrun.

We will emerge from the tunnel, still with LLMs, but they will be tastefully integrated as sub-components of greater systems.

The displaced software engineers will be hired back to do the janitorial work of scrubbing chatbots and all otherwise broken token-streaming text boxes off of the internet. Hopefully they get hired back on full time.

We will know who all the idiots are during that brief period of bag holding where the LinkedIn profiles are still plastered with nonsense like “Senior Principal AI Adopter & Ethics Officer.” It won’t matter, they will still fail up, but it will be funny.

Nah it’s all good, I appreciate the decluttering of the link!

Sorry, I just used the copy link button on Amazon

The tiniest amount of light will keep me from sleeping. Seriously. A neighbor can turn their lights on across the valley and if I have line of sight through a window, it’s over. I tried lots of things, and at long last, I found this mask that I absolutely swear by https://a.co/d/cDbUv9J

I lost it once, and bought two more, so that I would never be without one again. It’s the only sleep mask that works. It stays on all night. It does not leak light around the edges. Everything “shaped like your eyes” is a complete waste of time. This is the correct implementation of sleep mask, and I will never go back.

It’s super comfortable, and even slightly muffles noise since it wraps around your entire head. I do like the effect of good sealing ear plugs (Mack’s 31 Db only), but find them extremely uncomfortable to wear multiple days in a row. I’ll bring the Mack’s on vacation, just in case I am put next to an obnoxious snorer, there is very loud city noise all night, etc.

I don’t have issues sleeping in too late, because I actually get sleep with this thing, and after some time, I was just back on a normal rhythm, very consistently waking up at the same time.

I care deeply about sleep mask. Maybe it won’t work for everyone, but just in case it can help a single person, I am unloading my mask manifesto.

But that’s exactly what I’m saying, a big selling point WAS reinventing things. There would be a stdlib “http” package for all HTTP client needs, an “orm” package, and so on. This is how I remember Deno being sold many years ago, and I thought it was a noble cause.

One of the original selling points of Deno seemed to be that it didn’t use NPM. It would build its own broadly useful and good standard library.

The NPM compatibility madness really does kill it for me. Probably something to do with taking VC money and “expanding addressable market” or whatever.