pico303

As a software engineer who has developed on Macs (and Linux) for most of my career and has recently started a job that requires me to use Windows again, I can tell you from experience that Office on the Mac is far, far more stable, easy to use, and considerably faster than on Win11. Microsoft’s macOS team are really good at their jobs.

But then I don’t find macOS to be slow or a buggy mess, so mileage may vary.

I remember reading that even the Start menu is now a React app.

I absolutely hate this article. So you want me to learn, understand, and write code in a format an LLM can ingest so you can more easily copy and paste my knowledge? Seems like the worst of all possible worlds: I work for you for nothing, you and the LLM make money on my work, and you get to produce terrible software someone else has to maintain when you move on to the next "problem." How mercenary of you.

When I started working on one of my side projects a year or so ago, I realized I didn't have time to figure out how to style each and every component, so I paid for Tailwind Plus. It was pricey, and I definitely had to think about it for a few days, but I'm so glad I did. It saved me way more time than the dollar value of the product, and it has continued to get better.

If you are using Tailwind, I highly recommend Tailwind Plus. You'll learn so much about what Tailwind can do using that library, and it is so easy to adapt into your own offerings. It is 100% worth it.

Hearing that they're struggling, I may have to also bite the bullet and pick up Refactoring UI.

Note: I am in no way connected to the Tailwind folks other than through my credit card.

Maybe I’m not smart enough to grasp all these flowery words, but is this suggesting if I spend a few years writing some code, you should get to copy it for your own interests and without compensating me as long as your sales and marketing is better than mine?

I don’t think Rockchip learned from the ffmoeg code. They simply copied it outright without attribution.

I feel terrible for the author, having to work in this kind of structure. Someone should tell mytaxi that even Spotify never got “tribes and squads” to work, it sounds like for the same reasons/problems mytaxi ran into.

Edit: accidentally hit update instead of scrolling…

One thing missing from the article that I don’t think has been mentioned is confidence and setting expectations. I’ve found if I expect certain results and convey confidence, people are more likely to follow your lead, or at least listen to you. Don’t act like a know it all, and be sure to encourage and question others so the environment is collaborative (solve problems as a group; don’t be a hero). But set expectations.

Also, I don’t think you mean “intimacy.” Do you mean “empathy?”

Right there with you. I’m working on fixing an app deployment this weekend myself and dreading picking my way through GitHub actions, ansible scripts, container configs, and deployment APIs to figure how why the thing stopped deploying. Thank goodness it’s just deploying to VMs and not Kube, or I’d probably lose a week.

One of my coworkers (not a developer, but familiar enough with technology) asked about a new “low code” app and wondered if our engineering team used that. My response, I think, provides something of an answer to this paradox: it’s not writing the code that’s time consuming; it’s the requirements gathering. It’s not enough to generate code or even have the LLM build or deploy the software. It’s all the knowledge and experience around crafting an app that’s likely preventing the shovelware. Turns out code is only a part of our job.

Also, why is this the “Gorman” Paradox? He literally links to the article that I remember as first proposing this paradox (though I don’t think the original author called it a paradox). This author just summarizes that article. It should be the Judge Paradox, after the original author.

The example in the article of letting Claude deploy the app worries me. It has me thinking of that line, “AI is really good until you know what you’re talking about.” If the author was clueless of how to deploy the app, how do they know the app was deployed safely or securely?

Just this past week I asked Claude for some help with C++ and a library I was somewhat unfamiliar with. What it produced looked great—-if you didn’t know C++ very well. It turned out Claude knew even less about this library than I did, generating tons of code that was completely incorrect. I eventually solve my problem through research and trial and error, and it was nothing like what Claude recommended. It certainly didn’t leave me feeling confident enough to let the LLM have the level of control over my computer or project that the author is allowing it in the article.

I’m not looking forward to a future spending all my time cleaning up the messes LLM’s create.

This somewhat reflects my sentiment to this article. It felt very condescending. This "self-limiting beliefs" and the implication that Seattle engineers are less than San Francisco engineers because they haven't bought into AI...well, neither have all the SF engineers.

One interesting take away from the article and the discussion is that there seem to be two kinds of engineers: those that buy into the hype and call it "AI," and those that see it for the fancy search engine it is and call it an "LLM." I'm pretty sure these days when someone mentions "AI" to me I roll my eyes. But if they say, "LLM," ok, let's have a discussion.

I agree. This "cooldown" approach seems antithetical to some basic tenants of security in the open source world, namely that more eyeballs makes for better code. If we all stop looking at or using the thing, are these security professionals really going to find the supply-chain problems for us in the thing, for free?

Instead of a period where you don't use the new version, shouldn't we instead be promoting a best practice of not just blindly using a package or library in production? This "cooldown" should be a period of use in dev or QA environments while we take the time to investigate the libraries we use and their dependencies. I know this can be difficult in many languages and package managers, given the plethora of libraries and dependencies (I'm looking at you in particular JavaScript). But "it's hard" shouldn't really be a good excuse for our best efforts to maintain secure and stable applications.

On the Steam Deck you boot into desktop mode and it’s a standard Linux. Install what you want. I have Heroic Launcher on mine, running games from GOG and Epic alongside Steam games.

I slapped a nice sticker of a boat anchor on my work laptop. Seemed fitting. Slow (thanks to 4 separate malware/virus scanners), weighs a ton, lasts for an hour on battery.

Thank you for posting this. I had no idea this was out there!

I don’t mean this to sound like I’m being a jerk, but why would I want my MacBook to be awake with the lid closed? If I want it to be awake doing something, I leave the lid open and let the screen sleep. Maybe I’ve been using a Mac too long, but lid closed means sleep to me. I’ll even do that when I have it plugged into a monitor: close the lid to make it sleep.

Same boat, and 100% agree. I couldn’t find a single example of Windows or Windows software where I think the experience is in any way better. Windows only saving grace, as a developer, is WSL.

For a simple example, no app remembers the last directory you were working in. The keys each app uses are completely inconsistent from app to app. And it was only in Windows 11 that Windows started remembering my window configuration when I plugged and unplugged a monitor. Then there’s the Windows 95-style dialog boxes mixed in with the Windows 11-style dialog boxes; what a UI mess. I spoke with one vendor the other day who was actually proud they’d adopted a ribbon interface in their UI “just like Office” and I verbally laughed.

From a hardware perspective, I still don’t understand why Windows and laptop manufacturers can’t get sleep working right. My Intel MacBook Pro with an old battery still sleeps and wakes and lasts for several hours, while my new Windows laptop lasts about an hour and won’t wake from hibernate half the time without a hard reboot.

I think Windows is the “good enough” for most people.

What a lame excuse. “The G-Drive’s structure did not allow for backups” is a blatant lie. It’s code for, “I don’t value other employees’ time and efforts enough to figure out a reliable backup system; I have better things to do.”

Whoever made this excuse should be demoted to a journeyman ops engineer. Firing would be too good for them.

My company literally has four different apps “protecting” me now, including two different malware scanners. Neovim runs like it’s a 286. That said, before they’d installed everything it still wasn’t any faster than my Mac.

My “fancy” Windows work laptop has 45 minutes of battery life, while my M3 MacBook Pro will go 14 hours compiling C++ or running JavaScript and Docker images, and do so twice as fast as my work laptop could. I’d say you get what you pay for, but my work laptop was around the same price as my M3.

I wouldn’t be opposed to going back to Linux. But once you stop looking for power sockets all the time and start treating your laptop like a device you can just use all day at any moment, it’s hard to go back.

It’s not about swapping implementations, like the original poster suggested, so that you can later come in and swap out different implementations in a local sense. Interfaces are used to decouple parts of an application; for example mocking test interfaces. Another way to look at it is that interfaces are a contract that needs to be fulfilled.

Declaring a local variable as an interface to hide functionality so you can swap out functionality later is misunderstanding the fundamental theories around interfaces. Your variable should be whatever the function or method returns. If you want to abstract the type so it can be swapped out, create a function that returns an abstract type. Don’t tell the HashMap constructor it did the wrong thing. Now that I think about it, I’d recommend you use “var” in all cases, and not try to redefine the return values from a function.

No, you shouldn't write that. Your variable should represent the complete implementation. Where you use it should be as generic as possible, for example, the parameter in a function call.

I love how the initial commit is "48 years ago."

What a great piece. I’m so glad not only that his daughter will get to know her dad, but that her dad is going to appreciate every moment he has with her.

At least one CEO seems to get it. Anyone touting this idea of skipping junior talent in favor of AI is dooming their company in the long run. When your senior talent leaves to start their own companies, where will that leave you?

I’m not even sure AI is good for any engineer, let alone junior engineers. Software engineering at any level is a journey of discovery and learning. Any time I use it I can hear my algebra teacher telling me not to use a calculator or I won’t learn anything.

But overall I’m starting to feel like AI is simply the natural culmination of US economic policy for the last 45 years: short term gains for the top 1% at the expense of a healthy business and the economy in the long term for the rest of us. Jack Welch would be so proud.

I was thinking that too. I wouldn’t even trust that the “offline” tests didn’t have the questions and answers posted online somewhere. This might really be an analysis of how extensive the dataset is for each LLM, not how much smarter one LLM is from another.

I thought the headline was implying something else. When I read the article, I thought, “Is it a monopoly to worry about if you can just do it yourself?” Feels like complaining there is a monopoly on services to pour you a glass of water from the tap.

I use neovim for that.

I’m a dev working with AI to build tools for others, but I don’t use them personally. Why? Because they make your writing sound like everyone else, they produce shoddy and broken code (unless you’re doing something really commonplace), and they dull your own creativity. If you’re relying on someone else to do your work, you’re going to lose the ability to think for yourself.

AI is built essentially on averages. It’s the summary of the most common approach to everything. All the art, writing, podcasts, and code look the same. Is that the bland, unimaginative world we’re looking for?

I love the bit in the study about the “fear” of AI. I’m not “afraid” it’ll produce bad code. I know it will; I’ve seen it do it 100 times. AI is fine as one tool to help you learn and think about things, but don’t use it as a replacement for thinking and learning in the first place.

The target market for this editor seems infinitesimal. Who on Windows these days groks the terminal on Windows, wants a terminal editor instead of Notepad++ or VSCode, and finds Neovim overwhelming? And certainly no one on Ubuntu is looking for this.

That said, give me QuickBasic back in the terminal on Windows, Mac, and Linux, and now you’ve got my attention…

And after all that, your company gets hacked through a misconfigured router.