Posted by ranebo 17 hours ago
What would be most helpful for my workflow is something slightly different. I need to be able to launch specific browser profiles/windows in these workspaces. One space with all of the tabs for project X, another space with all of the tabs for project Y, and then another with all of the tabs for project Z. These might be in different browser profiles.
I don't see how I can achieve this under the common per-app paradigm of macOS space organizers unless macOS has some notion of Windows/Linux style shortcuts whereby command line arguments can specify the exact things that need to be in the browser window.
I remember the 2x2 grid in Ubuntu 12 being the best desktop UI I had ever used.
The current Gnome workspaces with a single row are a huge step backwards in terms of productivity. It must be easier for beginners, but it frustrates me every single day.
I've used Lemon Squeezy a couple years back, but after the acquisition I feel they've gone downhill. It's been a month since I submitted my product for review and I'm still waiting.
Stripe also has a MoR service now, I was able to set it up and ready to sell in a few hours
Plus the wharfs were cool, imo.
Hooo damn TextMate snippets, that brings back memories. Hard to convey how hyped I was to use these. That is also what drove me to Mac at that time. I remember writing hundreds of those snippets for every possible C++ construct, and <tab> to fill in variable name, type, loop counters and so on.
The real issue is how the ORDER of the desktops changes all the time which messes with that spatial memory and kills a lot of the productivity improvements. A consistent straight line would still be worse than a grid, but still MUCH better than the current situation.
This is depressing. I've been out of the field since Covid (after decade_s of work) and basically have to get back to work since kitty is gone, but this is definitely what I signed up for when I started on this career in software engineering.
If I'm gonna be reviewing all day, I'd rather manage humans rather than LLMs. How is it affecting managing engineering teams?
I think the point is:
Pre-LLM: 1. think, 2. write code & check, 3. review.
LLM: 1. think, 2. write prompt, <LLM writes code quickly>, 3. review.
If the thing that you enjoy about programming is writing code, you can have the LLM write code in the style you like. If you enjoyed getting to explore and understand a system, an LLM can help you do that quicker, too.
"Use LLM without thinking" won't get you substantially useful results.