bloomca

> The application had to be literally installed on the system to even debug it

I think that's because you chose "packaged" application, these apps need to be installed so that capabilities are handled correctly.

To be fair, macOS has the same issue, although they won't show in Launchpad, they still can be indexed by Spotlight.

Rendering engine + set of native components + APIs to make your own components.

Windows definitely shot themselves in a foot with building multiple renderers while building them on top of each other.

Basically I want an API available to build my own SwiftUI. Definitely not on the Core Graphics level :)

But good point, I actually think AppKit might be a good abstraction level. I'll play with it a bit and see if I can abstract it behind a good component model.

Nothing wrong with it, just want something a bit higher level and ideally with at least some native components/styles.

I really hope they do and the rendering engine is decently decoupled, I'll give a try building a framework on top of it.

I wish all platforms gave access to their rendering engine similar to DOM on the web, imo SwiftUI/WinUI (or WPF, but they are very similar) are not that good.

Haven't built anything native on Linux, though, no idea how good those are.

I use a similar approach. I tried to experiment going into a topic with no knowledge and it kinda fumbles, I highly recommend to have an overview.

But once you know basics, LLMs are really good to deepen the knowledge, but using only them is quite challenging. But as a complementary tool I find them excellent.

I think it is not really possible to fill all your time with "productive" activities, so to speak. I mean we are on HN, which is in general mostly a fun social network.

It is not a bad thing, just all about balance. If you are focused and consistently dedicate even little time to some activities you can achieve incredible results.

I am not really sure how bright the future is.

Sure, for some people it will be insanely good: you can go for as stupid questions as you need without feeling judgement, you can go deeper in specific topics, discuss certain things, skip some easy parts, etc.

But we are talking about averages. In the past we thought that the collective human knowledge available via the Internet will allow everyone to learn. I think it is fair to say that it didn't change much in the grand scheme of things.

I've been learning systems programming for a while and recently I discovered kilo[0], which inspired me to start building my own editor[1]. Not sure how far I'll go, but hopefully it will be usable -- how fun would it be to build another project of mine using my own editor?

[O] https://github.com/antirez/kilo

[1] https://github.com/bloomca/love

Yep, I think a lot of people genuinely want to disrupt the status quo. Sure there are some very questionable things attached, but in general people are extremely good at ignoring things if they get a better deal out of it.

I fully agree with you, I have similar insights (deep technical texbooks are also great), but I have to say that the value you extract is not equal depending on how you approach it.

If what you are learning is purely theoretical and you see zero application, you likely won't be super invested, might skip a few things and will likely forget a lot of them soon.

However, if the problem is relevant for you, or even better if you are working on something related, all of the sudden it quickly moves from pure theory to a very applicable thing.

I think that's why people who tinkered for years before university if they keep the passion can build insane things during/shortly after the university.

> I find so cool about learning with AI, you get to try, mess up, and improve without judgment

I agree. There is no judgement in going very deep in as simple concepts as you need, whereas with real people you'd feel like you are wasting their time and that you should know that already, or at least find out on your own.

Depends on the business logic, sometimes summaries (or a short demo explanation) help a lot to understand the made tradeoffs, so the reviewer can contribute more without spending too much time. It is especially helpful if the part is somewhat isolated.

But most of the time it is not very necessary.

i think it is the opposite. Users don't care about the backend, so as long as security is fine and DB queries are not too bad, it will be okay. You still need somebody oversee it, but the endpoints can be generated easily.

Any UI client, though, needs to look authentic or people will hate it. Maybe generic stuff like dashboards or internal tools is fine, but any premium product needs to have good looking front, and that is really tricky.

You can get a VPS in any cloud provider and treat it as your server, just copy files to it. It depends on the scope of your projects, though, if you need a lot of resources it can get pretty pricey for just toy projects.

I pay $12 for a server on Digital Ocean, and I host multiple servers behind an Nginx. But I am primarily a frontend/Desktop developer, so they are simple.

There is also an issue that ranking is easily available now.

Of course, recommendations existed in the past as well, and stuff like classic literature was ranked for centuries at this point, but still, I think we relied on word of mouth more.

Nowadays you can easily get a list of "top ..." in any area and chances are high that it is all old stuff.

I broke some of my devices, and some have battery become useless, and the price of changing is just not worth it, but overall? They last really long time. I even have some shitty 7 years old Chromebook still working okay passed to a family member, and Macbooks in general last very long.

And upgrading laptop components after 5 years just doesn't sound like a good value proposition.

Personally, I think that debugger is very helpful in understanding what is going on, but once you are familiar with the code and data structures, I am very often pretty close in my assessment, so scanning code and inserting multiple print lines is both faster and more productive.

I only used debugger recently in C# and C, when I was learning and practicing them.

Isn't MSIX packaged apps on Windows basically the same?

They will even list all the capabilities before installing, and I believe it can handle auto updates outside the MS Store as well.

But I think the main issue is that you can't give granular permissions. I would like to make my own sandbox rules, like only enable a single domain for networking for the app, only allow specific folders, etc.

I don't think you can get this granularity on macOS/Windows right now.

System in itself is not super productive. It can work with implied goals (usually if you have a lot of experience).

The most resilient structure is a web of interconnected goals: even if you fail at some, the web will not break. The more interconnected your goals are, the higher your chances of success.

Money, it's just business. I think every big corp is morally bankrupt (otherwise they wouldn't be big). There are some exceptions, of course, if a company found a sustainable way to monetize their output.

But the baseline is really bad.

Hijacking scroll, icons with no explanation, auto hiding content, not enough contrast between content and background, etc.

Frankly this ship has sailed long time ago, as sad as it sounds.

The thing about MS I recently realized is that whatever they do (and most of the technologies they output), they target it from the enterprise angle. So they check boxes with features, they just need to make sure they are available/usable, but they don't particularly care how nice they are to use.

So it is indeed an intentional choice just to make a good enough product and move on to something else. They never want to polish whatever they have.

Yep, I was discussing it with my partner a few months ago. It is just too good right now, in a lot of cases you just slip past all the fluff which are impossible to avoid with traditional search.

The opportunities for AI providers to capitalize on that are too prominent. No idea how long it will take, but imo it's inevitable.

Tauri uses native WebView (can be very outdated in old OS versions) and compiles to native machine code. Electron bundles full Chromium (rendering engine for HTML/CSS+V8 for JS) AND Node.js for your app code.

Honestly if there was an Electron without Node.js which would use literally any compiled language (although Rust is probably too low level), it would've been more tolerable.

Yes, it is the biggest selling point. You almost never need to debug user/platform specific UI bugs -- if the app starts, it will render the same. There might be some GPU issues, but that's not due to the framework itself.

Electron apps often fully share codebase with the Web apps, so on the app backend you implement native functionality and communicate with your app via IPC.

Just read other replies, if you use the web platform in any capacity, you end up with tons of hardly reproducible issues, and it seems on Linux performance is bad no matter what.

Chromium is superior to the native web view unless you have latest version of Windows or Mac.

I use Windows alongside my Mac Mini, and I would say they perform pretty similarly (but M-chip is definitely more power efficient).

I don't use Slack, but I don't think anything takes 20 seconds for me. Maybe XCode, but I don't use it often enough to be annoyed.

There are a lot of things, but one warning I can give is try to avoid perfectionism.

Cook some novel dish, or maybe just significantly modify your regular one with some spices. Write a blog post. Sit and think about some library you wanted to write, but didn't have time. Even writing a draft with API is a pretty good example of producing.

You can fix things by yourself. E.g. service your bike, maybe reorganize your workspace, stuff like that.

Just start small and think about things which bother you and what you can do about it.