Defaults matter way more than many think. More often than not, defaults are what inspire distro hopping.
Why? Because the path to the desired result from a big-name distro is frequently non-intuitive, often to the point that the user may not even realize it's possible. When something doesn't work as expected, the response isn't "I need to figure out which packages to install and what config files to change," it's "oh I guess this distro isn't what I'm looking for".
I think it would do an immense amount of good if the big distros did more to address this. If they made it such that a fresh install could be made to fit any remotely common use case and hardware combination with no more than 1-3 clicks that would make tiny distros much less appealing.
A handful of distros have the right idea by offering an install ISO with preconfigured proprietary Nvidia drivers for example, but even that could be improved upon by just rolling some heuristics into the stock install ISO to figure out if the user needs Nvidia drivers or not.
People generally want something that works, without tinkering - particularly on an entertainement device. I'll happily let Valve etc. pick the kernel and driver versions, set up the compositors, make the controllers work, etc.
The control is both a blessing and a curse. It’s really easy to accidentally screw things up when e.g. trying to polish some of the rough edges or otherwise make the system function as desired. It also may not be of any help if the issue you’re facing is too esoteric for anybody else to have posted about it online (or for LLMs to be of any assistance).
It would help a lot if there were a distro that was polished and complete enough that most people – even those of us who are more technical and are more demanding – rarely if ever have any need to dive under the hood. Then the control becomes purely an asset.
This is literally Linux Mint, Zorin, and several other distros. I haven't had to "go under the hood" on my daily driver machines that run either of these distros for over 7 years.
I think at this point people are just (reasonably) making excuses not to change.
Those and other big distros are better in that regard, but they're still not perfect. Depending on one's machine and needs, there can still be pain.
One recent example I experienced is jumping through hoops to get virtualization enabled in Fedora… it takes several steps that are not obvious at all. I understand not having it enabled by default since many won't need it, but there's no reason that can't just be a single CLI command that does it all.
What exactly did you need to do? All I've ever had to do to get QEMU working properly has been to make sure KVM is enabled in the BIOS (which you have to do on all OSs).
Just run a KVM based Windows VM (via GNOME Boxes, virt-manager, etc. On my Fedora install I had to install the @virtualization meta-package and enable dameons among other things, and the only reason I knew to do that is because I looked it up. Without that Boxes, etc just throws an unhelpful error that doesn’t suggest that more packages or config changes are needed.
I had to enable virtualization features in BIOS too, but that’s entirely separate and not the fault of any Linux distro.
Ah, I guess I might be a little unusual in that I use the QEMU CLI directly. I tried some QEMU GUIs in the past but they were annoying to get working so I just learned the CLI.
Things like that can be unbelievably annoying and confusing on Windows or Macs, too. Even worse, they can just turn out to be impossible: the company can actively be preventing you from doing the thing that you want to do, refuses to give you enough access to your own system to do the thing you want to do, and/or sells permission to do what you want to do as an upgrade that you have to renew yearly.
These are things that don't happen in Linux. Doing what you want to do might be difficult (depending on how unusual it is), but there's no one actively trying to stop you from doing it for their own purposes (except systemd.)
Also, as an aside, a reason that Windows and Macs might have easy virtualization (I have no idea if they do) is because of how often they're running Linux VMs.
One needs to go a fair ways off the beaten path before they'll start running into trouble like that under macOS and Windows.
For macOS in particular, most trouble that more tinker-y users might encounter disappears if guardrails (immutable system image, etc) are disabled. Virtualization generally "just works" by way of the stock Virtualization.framework and Hypervisor.framework, which virtualization apps like QEMU can then use, but bespoke virtualization like that QEMU also ships with or that built into VirtualBox and VMWare works fine too. No toggles or terminal commands necessary. Linux does get virtualized a lot, but people frequently virtualize Windows and macOS as well.
> It’s really easy to accidentally screw things up when e.g. trying to polish some of the rough edges or otherwise make the system function as desired.
'Similar to Windows' System Restore and macOS's Time Machine', the Linux 'Timeshift' tool can be used to do make periodic saves of your OS files & settings. (They can be saved elsewhere.) Restoration is a cinch.
Mint program 'Backup Tool' allows users to save and restore files within their home directory (incl. config folder and separately installed apps).
There's several distros that are fully usable without ever touching a terminal. The control is a gradient, some distros give you all the control and others (eg. SteamOS) lock down your root filesystem and sandbox everything from the internet.
You do have to know what you're doing. A complete OS has a bunch of components that work together. But an out of the box distro hides all that do you end up fiddling with incomplete knowledge.
Gentoo is great for learning what all the individual components are. You install it by booting a kernel from a USB stick then chrooting into your newly installed system to start installing and configuring everything. Just knowing the existence of individual components helps a lot. Plus Gentoo gives you more control than almost any other distro (much more than Arch, for example).
I wonder how many more failed tiny financial band-aids it will take before governments figure out that moving the needle on birth rate requires that deep systemic issues be addressed.
The most astute observation I've seen on the topic is that in a capitalistic system in which monetary value is assigned to everything, the value of children is deeply negative and therefore they are not desirable. By having children, most couples are putting their stability, wellbeing, and long-term prospects on the line. The opportunity cost is staggering. If more children is the desired outcome, that tradeoff must cease to exist, and a lousy $2k isn't anything remotely close to that.
Very few people understand the depth of what you just said.
$2K or even $20K is meaningless for a parent making $100K or more.
Kids have a negative value to a professional class member.
If you engage in agriculture or some similar activity, a child as old as 10 can be a helping hand in some way or the other. No surprises that Amish farmers have a high birth rate.
It's not clear exactly what the number is, but if one observes individuals who manage to climb out of the low and middle classes and accrue a certain amount of wealth (somewhere in the ballpark of $600k-$1m net worth and up, maybe), pretty consistently not long after that achievement they've settled down and started a family.
I think for many the desire is there, but sufficient de-risking is required for them to be comfortable with acting upon it.
$600,000 net worth is nothing these days, I’m worth about that much after saving for 7 years and can’t even afford the mortgage for a $300,000 house, even if I put 20% down.
Investments are so much better at earning money than working for wages, in my case the amount of retirement savings I have after 7 years is larger than my cumulative earnings during the same period, and I’ve been saving about 40% of my gross income. Part of my net worth is ESOP equity that I can’t monetize in any way so that’s part of the reason why my net worth is higher than my earnings over the same period.
I think there's just not enough money in the county to induce more babies. The cost would be a shock. Anyone wealthy enough to shoulder the cost would fight so hard against it, it would never stand a chance. IMO the number is probably something like $10k per year per kid. Foster Care pays somewhere between 8k-12k.
This is such an important point. As a father of two, children are turning out to be a very large investment...larger than anything else I ever will pour money into, probably by an order of magnitude (though not quite, since I have a house).
I talk to lots of people in SV, heads of design, engineers, as well as folks from around the world that I work with, from San Diego to Argentina and Chile. So many 20-30 year-olds have told me they are never having kids. Life is too fun, and they want to see the world. But training the next generation is hard work, and it's easy to do a terrible job. We want to incentivize people to have kids and be great parents. But that requires voluntary sacrifice, which is a hard sell.
If I hadn't had kids, I could retire now. As it is, I'll be lucky to be able to work and get a job so I can earn for the next couple of decades so I have enough to retire.
That would introduce a new problem of all of those seniors suddenly becoming more dependent on their younger family members, which is exacerbated by kids moving all over the country in search of greener pastures.
There's not really a solution that doesn't involve heavy restructuring in one place or another.
>That would introduce a new problem of all of those seniors suddenly becoming more dependent on their younger family members, which is exacerbated by kids moving all over the country in search of greener pastures.
That's how it always was. It used to be your kids were your retirement safety net.
What's different now than 100yr ago is that those working generations also have the state taking a 20-50% cut which used to be available to be sent home to help out mom and dad.
In the UK, for example, the retirement age was set to 60 for women and 65 for men, when life expectancy was substantially lower than today. On the current trajectory, a large number of "boomers" at death will have only "paid into" the system for half their lives, while extracting most of the economic reward of the last 70 years.
Even if React is technically the most popular, that’s somewhat moot because most folks who might be interested in writing desktop apps with Rust probably aren’t coming from a web background, but instead from old school desktop UI frameworks (AppKit, win32, GTK, Qt, etc).
For this crowd, the barebones “bring your own everything” style of React and its analogues and contortions required to write apps with complex UIs in a declarative framework are not very appealing. OO imperative UI frameworks with a wide and deep chest of batteries-included widgets still come out on top for this particular use case. Minimal declarative frameworks are fine for light utilities but become progressively more cumbersome every step beyond that.
> most folks who might be interested in writing desktop apps with Rust probably aren’t coming from a web background, but instead from old school desktop UI frameworks
You'd be surprised. A lot of people migrating to desktop apps are people coming from a web background (this is after all by far the largest pool of UI developers) who want something more efficient than electron.
“More efficient” is relative. While tauri and dioxus can lower the install size, they’re not going to be meaningfully more efficient than electron for memory usage or speed since Chrome probably does better than the other frameworks, and consistently across OSes which dioxus and tauri struggle with since they are dependent on the host OS’s web implementation. Slint is interesting here but then it’s more rust and less web coding.
Oh, "more efficient" here isn't a refence to Tauri. It's a reference to the parent comment's mention of "(AppKit, win32, GTK, Qt, etc)".
I work on a framework (Dioxus Native) which I think sits at an interesting intersection of "rust" and "web": it renders using HTML/CSS (custom implementation, not using a webview) but it uses Rust scripting instead of JavaScript.
What's "rust scripting"? I don't see any mention of it in the docs nor dioxus native roadmap and AFAIK https://github.com/DioxusLabs/dioxus/tree/main/packages/inte... is used as the interpreter although I can't for the life of me figure out where the JS engine is actually instantiated.
"interpreter.js" is for Dioxus's "web" backend which runs in browsers (who instantiate the JS engine). All it does is interpret "patch commands" to apply DOM mutations.
iMac G5s too! They even had a spot on their right side which was magnetic where the remote from that era could be stuck when not in use. My 20” iMac G5 doubled nicely as a TV in my high school years (and the 27” model with its dramatically nicer IPS panel even better in my college years).
Yeah, most of my computer use is with a properly adjusted desk setup with external monitors and while it doesn’t bother me to use a laptop to jot down some notes or for a short study session, if I try to do “real” work at all I quickly become uncomfortable. A cheap folding laptop stand (which elevates the laptop enough that the middle of its screen is eye level) and wireless KB+mouse dramatically improves comfort (and productivity) but the tradeoff is that you need a table or other sizable, stable flat surface.
The exception is if there happens to be a reclined-position chair (IKEA POÄNG or similar) around; this gives back support and reduces neck craning enough to make longer sessions more viable, but it’s far from a given that this kind of seating will be available.
In theory. In practice, a lot of EVs (and hybrids, which could do the same thing to a more limited extent) ship with the same cheap flooded lead acid 12v batteries that ship with ICE cars and don't handle constant charging/discharging well.
This puts a cap on how much the "smart" systems can do because it dramatically increases cycle count and thus the risk of the 12v battery losing the ability to produce enough voltage to start the car, leaving the driver marooned somewhere.
It could also result in a noticeable "vampire" drain on the high voltage battery which looks bad and could put you at a disadvantage vs. competitors.
I don't think many people really understand how awful automobile-scale internal combustion engines are at efficiency. The only reason they work at all is thanks to the absurd energy density of the fuels they burn.
Only speaking for myself, but the problem with Android is that it and the hardware needed to make it run acceptably are absurd overkill for the use case, which drives up cost, cuts down on battery life, and adds a layer of unnecessary complexity (suddenly you need to think about what player app to use, for example).
Basically part of the charm of a single-purpose device is that it can be built to serve it purpose ridiculously well and do nothing else, and the second general purpose software enters the picture much of that is lost.
The endless amount of Chinese Android-based single purpose mp3 player devices that are obviously iPod Nano/Classic clones basically cost ~$30 and have 50hr+ of battery life. You don't have to think about what player app to use, they ship with the only one that runs. The rest of the Androidness is stripped out.
Then yes, there's obviously the other end of the extreme where the mp3 player is very obviously a phone without a radio with a price tag to match. And everything in-between.
I'd say there's actually too many choices cause the silicon and battery cost required to simply play music has gotten so cheap that it doesn't make sense to optimize the OS further than Android. I'm sure the economics of scale means the actual hardware wouldn't be cheaper by any noticeable amount either.
> Only speaking for myself, but the problem with Android is that it and the hardware needed to make it run acceptably are absurd overkill for the use case, which drives up cost, cuts down on battery life, and adds a layer of unnecessary complexity (suddenly you need to think about what player app to use, for example).
The battery life is fine on modern DAPs. Excellent, even.
I understand why an engineer would want a completely application specific, built-from-scratch OS that does one thing perfectly, but that's a pipe dream for a niche market.
A powerful and efficient SoC that runs Android is ultra-cheap these days. Less than $1. Hiring an engineering team to write and maintain a custom OS for a niche product would incur so much R&D cost that it would wipe out any money you'd save by using a smaller microcontroller and drive the final cost up.
Just think: How much salary would you have to pay a team of engineers to write the custom OS and maintain it? If you could optimistically sell 500,000 of these devices (good luck) then how much would you have to save in order to pay for the R&D?
You don't need "OS" to play some music, drive display and talk via USB/BLE. It's trivial task and could be done with a few event loops. A lot of firmwares is being written without OS. May be FreeRTOS/Zephyr to somewhat simplify the programming, but that's definitely not "OS" in a commonly accepted sense. You don't need team of engineers, one hobbyist could easily do that. I wrote firmware for a device of similar complexity (work with ADC, implements USB, BLE, some UI with buttons and leds) and I'm not even a professional.
I can't vouch for it personally since I don't own one, but I saw a video on YouTube mentioning the Innioasis Y1[0], which supposedly does a decent job of replicating the iPod experience with some modern features like USB-C and Bluetooth at a decent price. Can be flashed with RockBox. No external SD slot, but it can be opened to swap out the SD card it comes with. Reportedly doesn't feel nearly as nice in hand as a real iPod does but that's pretty standard at this price point.
Damn this looks great. I'd guess the main difference in feel is the weight. iPod classic was about 2-3x heavier, which seems to be the main factor in premium feel.
Why? Because the path to the desired result from a big-name distro is frequently non-intuitive, often to the point that the user may not even realize it's possible. When something doesn't work as expected, the response isn't "I need to figure out which packages to install and what config files to change," it's "oh I guess this distro isn't what I'm looking for".
I think it would do an immense amount of good if the big distros did more to address this. If they made it such that a fresh install could be made to fit any remotely common use case and hardware combination with no more than 1-3 clicks that would make tiny distros much less appealing.
A handful of distros have the right idea by offering an install ISO with preconfigured proprietary Nvidia drivers for example, but even that could be improved upon by just rolling some heuristics into the stock install ISO to figure out if the user needs Nvidia drivers or not.
reply