> AI is great at allowing you not to write the dumb boiler plate we all could crank...
I've actually started having a different view on this. After getting over the "glancing instead of reading llm suggestions" phase I started noticing that even for simple or boilerplate tasks, LLMs all too often produce quite wasteful results regardless the setting or your subscription. They are OK to get you going but in the last weeks I haven't accepted one Claude, devstral or gpt suggestion verbatim. Nevertheless, I often throw them boilerplate tasks even though I now know that typically I'll end up coding 6 out of 10 myself and only use the other four as skeletons. But just seeing the "naive" or "generic" implementation and deciding I don't like it is a plus as it seems to compress the time of thinking about it by a good part.
I don't see how it qualifies as a legitimate download or ownership. You cannot save the file to a disk you control and you have no way to ensure you have continued access to it. Apple or the IP holder can cause this "download" to dissapear from your device/account without prior warning. Its actually written in the terms.
Afaic, people designing circuits still do care about that.
> Good Assembly
The thing with the current state of coding is that we are not replacing "Coding Java" with something else. We are replacing it with "Coding Java via discussion". And that can be fine at times but it still is a game of diminishing returns.
LLMs still make surprising mistakes, they too often forget specifics, make naive assumptions and happily follow along local minima.
All of the above lead to inflated codebases in the long run which leads to bogged down projects and detached devs.
In case you weren't aware, MacOS uses zsh as the default installed shell. The bash version that comes with MacOS is some ancient 3.x version, from 2005.
Just that it would be `.zshrc` since zsh is already the default shell in the context of this post unless you enjoy a mid-2000s existence and none of the nice bash features of this generation (in which whatever `.bashrc` you've crafted is likely broken by time passing).
Maybe I'm underinformed, but I don't personally know any Linux users who convert to zsh, instead opting for fish or something else (oilsh? nushell?).
Yes, I'll throw my hat into this group too. Bash is fine.
YMMV but I have found using zsh too frictitious to be helpful. Sure, theoretically zsh living in a bash world (lets face it, all scripts are bash) is completely fine but reality seems to differ. Copied a one liner from shell history into your script? Crash. Use arrays? Weird bugs. Use shell builtins? Whoa unexpected interactivity!!! Etc...
Bash is absolutely fine as a default shell. As an added benefit, you don't feel like an invalid once logged in to a container or server.
I also think that it sends the right signal in terms of "Hey, this really doesn't need to be an app". I don't need an app for my newspaper, I need a shortcut/bookmark to its web page.
And once you start thinking about it, the same thing goes for a surprisingly large amount of apps.
I feel like in the coming years the facade big A and big G put up in order to push everyone into their distinctive walled garden of apps will crumble in public opinion.
It never was "yeah, it needs to be an app because the web platform doesn't have an API standard for it", geez, apple even forced a single web engine. They could have easily allowed access to their APIs on the browser. It just never was in their corporate interest to do so.
Okay, this devolved into an anti corporate rant without it being my intention to... So, go web!
I don't really know how to articulate exactly how I'd classify into one bucket or the other but I think there are two types of "app" and I tend to have differing preferences on whether they should be native apps or web apps as a result.
One is where relatively-static content is the priority, deep-linking is important or essential and the web platform is pretty ideal for those. News articles or blogs or Wikipedia pages or those sorts of things. Things where I might want to be switching between tabs or forgetting about for a while and coming back to later.
The other is where the app is primarily interactive or where the content is a lot more likely to be real-time or ephemeral. Not least because if you're on a low-bandwidth or high-RTT connection, navigating between web pages or having interactivity blocked behind a backlog of XHRs (particularly where caching isn't permitted) is utterly miserable. My experience is that native apps usually continue feeling responsive to input even when the network itself is not responsive but that is often not true with many clickable elements in many web pages.
PWAs might be the middle-ground here but they feel a lot like Electron apps to me: still foreign to all platforms, not responsive in the way that native UI controls are, weird/missing "back" behaviours and still no better support for deep-linking than the average app would have.
I've actually started having a different view on this. After getting over the "glancing instead of reading llm suggestions" phase I started noticing that even for simple or boilerplate tasks, LLMs all too often produce quite wasteful results regardless the setting or your subscription. They are OK to get you going but in the last weeks I haven't accepted one Claude, devstral or gpt suggestion verbatim. Nevertheless, I often throw them boilerplate tasks even though I now know that typically I'll end up coding 6 out of 10 myself and only use the other four as skeletons. But just seeing the "naive" or "generic" implementation and deciding I don't like it is a plus as it seems to compress the time of thinking about it by a good part.
reply