Please, FOSS world, we need something like ChromeOS Dear Santa. For Windows-10-end-of-support-day in October, please may we have a dead simple bulletproof all-free OS that gets old PCs online without a Google account, and does nothing else? There are a lot of desktop Linux distros out there. It's the number one thing non-Linux users complain about: too much choice. One of them …
I've used Linux since SLS came on infinity floppies, and was a SunOS admin when a single pizzabox with a 68K cpu ran a lab of green screens.
But it does occur to me that my home non-development / laptops only need to run Brave+Firefox and an ssh shell.
They would be Chromebooks except for signing into Google and yes of course all my phones run Grapheneos.
I *am* using what's available, for my personal and professional computing needs.
What was made quite clear in the article but which you seem to have missed, is that the request is not for the tech-savvy but rather for the gazillions of poor sods that don't inhabit the tech bubble that you (and I, and other Reginauts) live in. Come October they'll be either hapless users of an unsupported OS or have shelled out for ridiculously up-specced machines shoving Windows 11 and Clippy.AI down their throats.
They could use our help, not your scorn.
Perhaps Canonical could step up.
I agree that scorn isn't helping, but I think what they're getting from this suggestion is unfounded condescension. From the features you've added to your suggested solution, you'd think that the problem with existing Linux is one or all of these:
1. The users are too stupid to understand multiple icons.
2. The users will break anything, which is why it needs to be locked down.
3. It should be expected to break every five minutes, which is why we need several layers of autorecovery.
In reality, none of those problems are significant problems to Linux adoption. You can put a lot of users in front of Ubuntu or Mint with no modifications and they can figure out, surprisingly enough, how to click on the Firefox icon. I have done this several times and identifying how to launch the browser was always the users' first steps, even on that one occasion where I forgot to tell them that they were getting a Linux machine and they just got a surprise when booting it. While anyone who tries can break a Linux installation, it's not likely to do so on its own, and you do tend to have to work at it.
This also ignores why Chrome OS is as common as it is, and it has nothing to do with its features. It is successful because there's hardware you can buy running it. It's successful because that hardware is cheap. People have often purchased Chromebooks unaware of the software they'd end up with, and most unaware of the security death dates built into the thing. People who are willing to install your Chrome OS imitator will be fine with many of the choices we have now*, but your problem is all those who won't install it, either because they don't know it's an option, don't want to risk a change from what they have now, or have a problem with the wipe and replace process.
* There are distros that are just unstable. Those must be removed from consideration. There are ones that don't have graphical tools for the basics. Those also must be removed. There are a lot of things remaining, desktop-oriented GUI-default operating systems with a browser preinstalled, which will be fine for this purpose. The reason why the task is more difficult for the rest of us is that we're also trying to find a distro where the average user can, equally easily, use all sorts of other applications locally. It is harder to find a distro that makes it simple to run a piece of Windows software through Wine without needing to know much than to find one where the user can get to a browser.
I know what Chrome OS is. This isn't about Chrome OS, but whether an open source thing that's very similar is a helpful idea. I don't think it is because it misunderstands almost everything about what people need and what the problems have been with adoption of existing Linux distributions. It fixes none of the real problems and it introduces several more, so I think the idea is a bad one and not worth investing in unless some people choose to do it on their own anyway.
>> Productivity apps? Use Google Apps.
Bite me. Don't want Google apps, don't want a Google account, and later this year I'll be blocking both gmail and outlook.com at the server (as well as the .IN country level TLD) for excessive spam.
It's amazing all the money Google and Microsoft are spending on AI, but they cannot clean up their mail services.. So they're getting blocked.
Sorry, but there are many people out there who should NOT run Linux.
Windows exists for vacant look mouth breathers don't have the ability to operate anything else. Someone who has zero interest in having privacy or independent thoughts. People who think making Microsoft shareholders richer is a good thing. The people who don't read the EULA and like signing away their rights through complex one-sided legal agreements.
I have spent far too much of my limited lifetime installing, then uninstalling, Linux because they can't find Notepad.
The Greater Unwashed deserve Windows. It thinks for them.
It's certainly a fire up and just get on with it OS, and requires a lot less work to keep it running well. However, macOS only runs on Mac hardware which, despite being well made and an economically sound choice over its full working life, the starting price is just too high for many people. Chromebooks and basic Windows laptops start at under £300 - the cheapest (albeit new) MacBook starts at a little over three times that. I switched from Windows to Mac over a dozen years ago and, whilst I still have a Win 11VM running, it's rarely needed (in fact, it's fired up more often for updates than for anything I actually need it for). I also have a Ubuntu VM that is more use.
I've installed Ubuntu on a number of older PCs (mostly laptops nowadays) and it works fine for the owners - not particularly PC literate but, once I've spent a bit of time showing them LibreOffice, Firefox and Thunderbird (plus set up email accounts and copy over files and browser bookmarks) they get on fine. A bit of ongoing handholding is needed, especially for those who want to do a bit more, but nothing too taxing on my time - more than my wife needs nowadays with her MacBook, but less than relatives/friends sticking with Windows.
Having said that, none of those people would have been able to switch without some initial handholding and support. It's a pity there aren't (m)any laptop suppliers prepared to offer them pre-installed with Linux (I think either Ubuntu or Mint would work - probably other flavours, but I've not tried them).
There is hardware with Linux pre installed: Tuxedo from Germany and System 76 from the US. They both have websites to order from and both roll their own distributions that they maintain and update.
All the knashing of teeth in this article. Jeez, just open your browser and do a search.
I looked for approximately the same type of OS for the same reason - something that could be installed onto laptops not supported by Win 11. In my car as a volunteer or a repair/reuse charity, looking towards the end of 10.
The best I could find was Fedora Silverblue, as I was looking for an environment where a minimal environment was provided which could be extended with Flatpak or similar via some GUI marketplace. Silverblue ticked almost all the boxes I had, though the one thing that was missing when I looked was some form of "factory reset" which could be performed from the UI.
I suspect this is isn't really a feature Fedora's team see as a high priority - the OS isn't typically pre-installed and the presumption is probably that the person installing the OS will be around to reinstall should it be required at a later date. This however was exactly the thing I was trying to avoid needing...
I kinda feel that it's this point - the expectation of a fairly tech literate person being on hand post install - that's leading to this use case being under served.
Nonsense. LMDE 6 is every bit as easy to use as Windows 11 save for the fact you HAVE to know and enter your password to install applications or updates. It even has a "store" for finding and installing them.
The ONLY "special knowledge " needed is how to remove the old windows partitions during the installation process.
You don't do things exactly the same as Windows with Chrome OS either, so there is a learning curve no matter what you migrate to.
I'd agree that Ubuntu, Mint etc are all pretty easy to use, and could fit the bill. But (having upgraded my Tosh Chromebook 2 to Mint) I can assure you the primary problem is the process of making that change. There's usually a need to flash the BIOS with a non-OEM one, possibly a need to dabble in the gubbins and remove hardware write protections, you'll probably need to nuke the built in drive and its contents, the command line interface is most unfriendly for the overwhelming majority of consumers. There's quite a lot of online help, but often that help is not being updated, so users may need to make some educated guesses about how to get the outcome they want. Once done, there's few challenges in running any sensibly chosen distro on a Chromebook, but getting there, that's interesting, with various moments where you're pretty sure that your next key press is going to only have a binary outcome, either (a) it worked and you're on to the next step, or (b) you just bricked your Chromebook.
I also have a Toshiba Chromebook 2, which I originally selected because it was one of the few Chromebooks available at the time that had an upgradable SSD and the bonus of having a backlit keyboard. I did put a bigger SSD in it and replaced the battery, which sadly needs replacement again.
I have yet to undertake the process of updating the BIOS to allow me to install a Linux distribution for some of the reasons you mentioned. I have gone through this process on another Chromebook (I forget which model) and I agree, it's not something the average non technical person could normally manage without encountering a fairly steep learning curve.
All that said, I believe the main thrust of the article was regarding an easy to use replacement OS for Windows machines, not Chromebooks.
Liam wasn't asking for a new OS for the Chromebook. He was asking for an OS that would turn laptops previously running Windows into something like a Chromebook, with every limitation that Chromebooks usually have, minus the limitation of having to sign into a Google account, plus the limitation of not being able to run Android apps, and in practice probably with a couple more limitations than even Chromebooks have when trying to run Linux software locally.
>> minus the limitation of having to sign into a Google account
Not really. He wrote to use Google apps. Needs a Google account.
>> Anything else that you want to do, you do in a browser window. Productivity apps? Use Google Apps
If you're going to do that, might as well just buy a Chromebook anyway. Because you're still going to need a Google account to use Google apps.
But most people would be better off with a cheap tablet. It's not like they do anything beyond surfing the web, doing some email, and maybe a video call or two - and that can be done with a $150 tablet (less when on sale)..
True, but in Liam's defense, that's one of multiple options. You could use any web-based word processor. So if you don't like Google, you can use Microsoft's or Apple's, or if you can convince people, Collabora Online. A Google account becomes optional. My problem with Liam's suggestion is that I don't see why that's any better than a normal Linux where, in addition to any or all of those, the user also has multiple offline word processors where they can save their files to their own hard drive without having to pay for a subscription service, either with cash or with their data. So far, the only reasons I can understand are either mistaken (the users will break everything so need something locked so tight that they can't) or come with tradeoffs (automatic backup of user files would be useful, but the privacy or financial consequences are ignored as if they don't matter).
And this is why Linux is going to struggle. You are only seeing this from the point of view of someone who already understands Linux. Like them or hate the Microsoft, Apple and Google all have an OS that comes pre installed with a (mostly) usable and consitent interface. Applications can be downloaded and installed from the corresponding App Store.
Updates happen with no interaction and within reason for the majority of users there are no issues.
Now take Linux, as a complete Newby where do you start?
Forums are full of people all proposing there favourite distribution and pet applications.
What is being suggested simply brings the convenience, eae of use and some form of privacy to a FOSS operating system in attempt to give an alternative that is not tied to a big tech company.
Neither of which is a big deal, because most of the users we're trying to convert didn't have those before and dealt with it just fine. Windows can sync everything through the cloud, if you set it up. Most people didn't. Storing some documents in OneDrive because various Microsoft applications default there is not the same thing. Nor did they have single sign-on. They had account-specific passwords and maybe stored them in a browser's password manager. These are not the most important features. Incidentally, they're also not what you spent most of your article on. You mentioned those features along with plenty of other ones that either don't matter to those users (update stability is great and would be welcome, but people don't find themselves needing it very often) or harmful (fixing the "how do we convince users that Linux can do as much as Windows" problem by making sure that it can't but telling them that's a good thing).
" Like them or hate the Microsoft, Apple and Google all have an OS that comes pre installed with a (mostly) usable and consitent interface. Applications can be downloaded and installed from the corresponding App Store. Updates happen with no interaction and within reason for the majority of users there are no issues."
Ubuntu Desktop which has all the above you just listed.
Sure, there may be a plethora of choice, but if you can't search and find a "beginner" linux distro with one search, then literally everything else is going to be beyond you anyway.
Hint: You can install Ubuntu Desktop from Hyper-V Quick Create if your Windows edition is high enough.
Yes, the clue is in the name: Personal Computer
It's why they prospered in the first place, rather than an array of minis and mainframes serving up sessions. The user was in control. And yes, the user did (does) need a little bit of grey matter. But that's the price of both inspiration and freedom.
For most users they need an automatic backup of everything as they will not know/care about data protection until it is gone. Seen that with folks PCs and phones in the past, now with Apple & Google & MS automatically forcing a web account and some means to sync data they have that.
What the lose is privacy of course, but how does one fund a universal backup system without whoring your users to every slimy advertising agency around? Paid is the obvious answer but rarely a settled aspect in the FOSS world, then how much data / what rates, etc?
> but how does one fund a universal backup system without whoring your users to every slimy advertising agency around
You... ask money for it? You seem to think that advertising..., I mean, theft is the only funding model, and I'm not sure why...
Their comment did talk about that suggestion. Paid backup is the logical answer, but they were considering people who don't think about the problem until after they needed it. In that case, a paid system often doesn't work because people don't pay for it until after they think about it. A system that does not require payment because it destroys the user's privacy results with them still having some of their data, whereas a paid solution that they don't use means they still lose it.
I see this as the user's problem. I'll make sure backup software is available, I'll give them options to back up to local hardware or to an external location, neither of which will be free, and if they choose not to and lose some data, that's an expected result of their actions. The problem the previous poster was talking about arises when you want to have a good result even when the user hasn't done what they need to, and my version does not solve that problem. If you want to solve it, that's where the straightforward "ask them for money" plan develops holes.
Give up trying to make it foolproof.
Accept that there are some people lacking the will -- and many-fewer, the ability -- to read instructions, comprehend them, choose to follow them, to think logically, and to exercise good judgement.
These are the people who drive their cars onto bike/walk paths (I've seen that, and once had to slam on my bicycle brakes to avoid being struck), on to goat paths, into creeks and rivers, down embankments, onto commuter-train trackways (saw that once, too; they got their car stuck) because they either couldn't figure things out for themselves, or because their sat-navi system said, in a commanding voice, "Turn right, now."
Ignorance generally can be fixed with education.
Arrogance, indifference, and willful ignorance ("I don't have time to read. Just make it work!") generally can be fixed (well ... eliminated, in specific cases) with a bullet through the relevant skull. That's not something I'm willing to do; I hope that most other people are also unwilling to do this.
That leaves adjusting our target audience to the non-goombahs in the world.
For most users the most important thing to back up would ve their usernames and passwords to various services, and especially important would be the backup of the username and password to access their usernames and passwords, because that one, and the password to the recovery email is lost first, and at the same time as their phone number changes.
Solve that.
No problem. Keypass. Multiple laptops? Still no problem. The central backup to this is Nextcloud under my control, not Google's, not Microsoft's. There;s a separate Keypass account on the NC server and all the NC desktop clients all use it to sync with the server whenever they'd logged in with access to the server.
Exactly this.
On a plane with no free wifi? Its no use to you at all. Can't edit a document, can't watch a locally saved movie.
And thats one example. On a train last weekend and the "free wifi" was not working and phone signals dropped out along the countryside. Was a few grumpy parents who didn't prepare for offline with their kids.
And thats before you go to companies adopting a similar approach, where customer sites can have no phone reception and they don't want you on their wifi that only extends as far as the office block, not the production floor. Plus security risk of plugging yourself into unknown wifi points.
We need something we can install software on (even if its an emulator to get PC or Android apps working till it has native apps of its own)
The idea of it having its own app store for beginners, with options for advanced users to install non-store items, does appeal.
Although I upvoted this i don't entirely agree with it. There are use cases for the thin client approach but not necesarily with the likes pf Google or Microsoft as the back end.
One might be the corporate user who has to work remotely - not from home but from hotels, cleint sites or whatever, and especially internationally where equipment is liable to inspection by nosey border officials. In that situation would you want corporate materials to be on the laptop once a password is demanded? Or even worse, be exposed via a company VPN?
Another might be class work where the students log into their courswork remote desktops now and the next class an hour later logs into something different.
> There are a lot of desktop Linux distros out there. It's the number one thing non-Linux users complain about: too much choice.
True... but let's just look over the fence for a sec. There are, say 100 car manufacturers with say 25 models each, give or take. That makes 2500 models to chose from... and yet most people seem to be perfectly able to manage these many choices. There must be more to it than just too much choice. Lazyness? Can't-be-bothered-ness? Too much FUD? Fear of tech you don't understand?
Lack of knowledge?
When you go buy a car, there's someone holding your hand throughout the process explaining the differences. You can easily hop in and go for a test drive to experience what it is like, and when things don't go right, you go back to the garage and let them help you out. None of that is possible for less tech-savvy people when it comes to OSes (unless they know someone willing to do the hand-holding).
Also: people simply don't care. When you don't care about what car you drive, you go to a used car lot and just get whatever gets you from A to B, and this article posits that this option doesn't exist for Linux distros.
That's an entirely different problem. AFAIK Chromebooks are severely locked down and need more insider knowledge to change OS than you'd need to install any Linux flavor on a normal computer.
I might have misread the article, but it was more about having something ChromeOS-like, i.e. a system with reduced possibilities fit for the average browser/email/YouTube user (which admittedly is a large part of somewhat aged home users).
Even if such a distro existed, you would have just as much difficulties installing it on an old Chromebook.
"When you go buy a car, there's someone holding your hand"
They do that to take your money. When it comes to computing the taking your money is done through the hardware vendor and they will sell you something with either Windows, ChromeOS or, in one case, macOS installed. If you don;t know there's another option the hand-holding won't lead you to it.
> Also: people simply don't care. When you don't care about what car you drive, you go to a used car lot and just get whatever gets you from A to B, and this article posits that this option doesn't exist for Linux distros.
Isn't that what Linux Mint is for?
/s
(I kid, I love Linux Mint, it's a great and wonderful used car!)
When you don't care about what car you drive, you ...buy a Toyota.
Every Camry, Yaris or whatever I see around here is owned/driven by people who just want an appliance on wheels. It's testament to how successful Toyota have been in catering for that market that I couldn't think of more than two of their model names for this paragraph. I'd be hard pressed to enumerate the Hotpoint washing-machine range, for the same reason.
You can , using a simple USB key get a live distro on. Easy to do , allows to see if everything works before you install and if you're ok click an " install " icon that's already on the desktop. I don't see why one can't do this simple test first. Make several keys , try them and install what works for you. Mint is perfect in that respect and with timeshift the matter of backups is taken care of right out the box. Updates ? Automatic if you want them to be. It's not like the old days , all the objections have been answered with time and work from the community. Really , just use live usb keys / dvd's and install what you like. So simple that people forget this is not a stage 2 Gentoo install anymore. Anyone with a minimum of effort can get Mint or orher distros up and running.
Getting stuff onto the key needs a few steps - downloading it, installing whtever is the preferred S/W for copying it (Windows isn't going to have dd installed) and then running it. Keys with pre-installed live distros might be the way to go. It also solves the which distro? issue. Try before you buy - try several - and you can always reformat and reuse the keys afterwards. Set up the keys so ensure they only install dual boot.
In the case of cars, they are buying a product. If it was the same with computers you would be buying a computer with your OS of choice installed - and most laptop manufacturers offer few or no Linux options, and those which exist are not directed and the non-specialist channel.
An equivalent with cars would be if you were asked what engine management system you wanted on the car, and you would need to install it yourself in many cases.
But they all have the steering wheel, accelerator, brake and gear lever in the same place. And they all have the doors in the same place - mind you, the locking mechanisms can get a bit esoteric.
Minor controls and "infotainment" are another matter.
It's my latest motor which automatically decided it needed to roll the seat back as I opened the door and then gently bring me forward as I started the car which struck me things have gone too far.
I'm a couple or three years into middle age and don't need to be reminded of my tummy thank you very much.
Consider yourself relatively lucky. My father-in-law's just bought himself a brand new Mazda, which has a camera in the cockpit to constantly monitor the driver for signs of inattention - very Orwellian. But possibly in a desperate attempt to humanize the camera and give it a non-sinister purpose. Mazda's engineers also use it for facial recognition of the driver, to thus set personalized seat/steering-wheel/mirror positions.
Which would be great, except that the facial recognition algorithm is completely bobbins and fails to identify my FiL 95% of the time.
...Which would be harmless if it just retained the previous seat/wheel/mirror positions, but no, it reverts them all to "default driver" mode every single time - and the "default driver" is apparently someone of Warwick Davis' stature...
"But they all have the steering wheel, accelerator, brake and gear lever in the same place. "
Most computers I have used tend to have the keyboard in the same place (ok some of the keys are in the wrong location) all tend to have a mouse, or track pad.
How is this different?
I know far too many people that buy a new PC becuase there is a new version of M$ windows (or even Office), simply because they didn't think they could change the software that it came with....
Cars have something called "interoperability", to a degree that they don't even talk about it, it's just assumed.
I can drive almost any car on almost any road.
I can fuel it up almost anywhere.
I can get in a car I've never seen before and drive it safely and effectively.
My friend can drive my car, I can drive his without re-training.
Most people fit behind the wheel or the primary passenger seat of most cars.
I can get my car serviced at many places, both "official" and third-party.
When my car breaks down, the people rendering emergency assistance don't care what brand or model my car is.
I can generally assumed my car will fit in my driveway and garage.
My car can be brand new or 40 years old, and it will still do "the job", and all the above still applies.
All this would be a nice goal for the IT and OS world to strive towards, but I don't see it happening in my lifetime. And with the options we have currently, perhaps that's a good thing if none of the current crop becomes "The Standard".
In Britain, maybe, where garages - if you're lucky enough to have one - are still designed to accommodate an Austin 7, the standard dimensions having been set many decades ago.
In the States - well, here I have room for 2 large cars in my house garage, and space for another 5 in the garages up at the barn.
All of that is entirely theoretical, alas, because every single garage is full of my wife's crap carefully curated artifacts and priceless heirlooms.
Comprehension problems?
I was saying, backing up John Robson's point, that in the UK, too many garages are still designed to an obsolete standard car size (probably dictated by national planning regs). The home I had there in the late 90s was built in '92, yet the only time I managed to park my Vauxhall Astra estate in it - hardly the largest of cars - I had to climb, Dukes-of-Hazzard-style, out of the window because I couldn't even open the door.
Here in the States, builders are pragmatic and build to the size that's needed and practical, rather than what Sir Humphrey wrote in "Planning Document 1947/1/a-subsection(iii)".
Let me know if you need a more simplified version.
I can drive just about any computer in the world
A Dell, a Toshiba, HP, Lenovo - lets call these stick shift
I can drive a Macbook - lets call that an automatic
and I can drive a tablet or mobile phone - ok things are getting a bit harder with the car analogy now - so lets add in motorcycle, I can drive one of those two.
I can also drive a semi-truck, and road roller, now we're are getting into specialist driving - so that will do for most Linus distros.
(I had great fun learning to drive a steam traction engine)
I have never had a play with any of the BSD variants out there, but I sure as hell 'recon I could do most 'office' type work on one if required, and would have a fair go at basic systems admin.
This is something called "interoperability", to a degree that they don't even talk about it, it's just assumed.
Most people could do this but for some reason have been told that it is terrifying to even try.
/Rattus
p.s. never did learn to oporate a crane though, one day....
Exactly Liam's point. Most people only do the basics in a car and no idea of a lot of the features
All they want from a device is to get on and surf.
In a car they don't care if it is manual, a slush of, CVT or all electric, select a gear or put it into D and go
Only some people care about the above
>> Exactly Liam's point. Most people only do the basics in a car and no idea of a lot of the features
Most people nowadays, their primary compute device is a phone or tablet. That's just going to increase. A lot more convenient than even a Chromebook, never mind a laptop of any type.
Phones already have 63% of all end user computer devices. So this entire debate is kind of pointless.
There aren't anything like that many. Not any more.
But honestly in my experience the same thing applies with cars as computers. Most people manage the choice by not really making it. They just buy what they're told to buy by someone who they trust to know. I just asked my office mate what he's driving these days. His answer? "It's a blue SUV thing. Hybrid. Good seats."
He doesn't know or care. He went out with his other half and got the thing they got offered the best finance deal on. He doesn't even remember who made it.
As this pertains to Linux I imagine the vast majority of people don't chose Linux for the same reason the vast majority of people have never tried to drive a Mitsuoka. They have no idea there is such a thing.
SteamOS and Bazzite are great for gamers, I've run both on a custom-built gaming rig. For the purposes described in this article, though, I don't think Valve will save the day for the masses. They are concentrating on one use-case: gaming. They don't really have the incentive to produce an OS optimized for grandma's old laptop.
There is a partial fix for that - Edge is available on Linux, at least as a .deb, it works fine on MX linux and LMDE. At least all the web based MS stuff is available
The real issue, as has been pointed out, for a large number is getting it installed. LMDE looks more like Windows 10 than windows 11 does out of the box.
Those actually stuck with Windows (rather than unable to get an alternate) are those dependent on software that does not have non Windows options.
This includes includes a lot of multiplayer gamers, where anticheat acts as an artificial block, (the only fix for that would be legislation) or productivity apps (where a VM would often be a solution, but likely to be an unacceptable one).
I personally suggest to friends that before they spend a load of money on replacing their computer which has an unsupported OS they try LMDE. (Providing assistance if need be).
I absolutely second your point.
I was once involved, as IT support, in a project that was supposed to be a non-profit organizing IT trainings for the poor in order to help them get positions in the workplace. A laudable goal, I thought, so I participated gladly.
At one point, the lead responsible made an off-hand remark about how expensive MS Office was and I immediately offered to install LibreOffice for them, explaining that it was free and had all the functionality they needed. I was showered in thanks and installed LO on their five admin laptops (admin because they were the non-profit management, not because they had any particular IT skills).
A few weeks later I went back to configure this and that, par for the course for a non-profit project, and the lead responsible commented on the fact that everyone had gone back to Excel and Word because LibreOffice was "too complicated".
I kept my mouth shut and made a noncommittal "hm hmm" noise.
From what I've heard since, that non-profit has dissolved into nothing.
When a person has years of experience with a particular tool, why would they want to spend time working their way back up to speed on something different when they could hit the ground running on the familiar?
I love learning new stuff but many people are about getting stuff done and they will take the short route to their objective.
Sneering about it is just arrogant, and crying because they don’t want to play your game is infantile. They are not stupid they just don’t have the time to start again.
Another elephant in that room would be 'how do I ensure all my data, files, pictures and so on are still there', let alone any apps that they might use besides a browser. Even ignoring MS Office it's going to be a big challenge to overcome that, even if it's only a perception of difficulty. Pisses me off every time I have to refresh the home computer like for like.
Download the operating system you want to install. Search for Linux distributions for beginners to get some suggestions.
They are asking a user, who presumably knows zilch about Linux, to search and do not give any clues what to search for or what makes a good distribution.
They should have chosen one (I suggest Linux Mint, Mate desktop) but if they had they would have been flamed by all the other distros and desktop users.
Yup.
That's the problem I find with the penguin community. They all laugh at you when you're on Windows, but choose one distro and you get yourself a barrage of insults because you didn't choose their preferred distro.
For all the holier-than-thou behavior of the Linux community, there is a whole lot of pre-teen testosterone behavior in a lot of them.
There is certainly a subset of immature, insecure, defensive people who cannot cope with anyone having a different point of view to their own, but they are a subset, and not the majority—at least in my experience.
Also, sometimes we fragile petals need to have enough determination and good sense to move on when one of the above types decides to vent on us.
Having said all the above, changing OS is NOT a trivial matter, especially for anyone one without much of an IT clue, which is the majority of the population.
Give them a pre-loaded OS, however, and many of them would have a fighting chance of getting going—assuming the necessary 'determination and good sense'.
Something like VanillaOS, with a little bit more polish, would probably serve a lot of folks very well.
I actually have trouble making up my mind which linux to use. The only place where it is obvious is where the distribution is optimised for particular hardware, as is the case with Raspberry Pi, their tweaked Debian is the obvious choice. And if is for that reason that I tend to favour Debian derived stuff and when I find myself messing with RedHat derived distros, I become a bit lost and frustrated.
But everytime I touch linux I find myself swearing about something. Increasingly I am looking for alternatives. Last night it was KVM, just running the same OS as the host as a guest, had to stop and have a cup of tea to calm down.
Linux zealots often suffer from the curse of knowledge and because something is plain clear obvious to them, they think it is easy for the uninitiated. Comes across as patronising and arrogant.
I'll trade your anecdotal evidence for mine: I've been running KVM on my Linux setups (currently, and for recent years, Miint) for at least ten years and I can't remember ever having any issues.
Possibly you have an unusual/flaky hardware setup, or your wicked fairy godmother cursed your Linux experience.
Interesting take.
How about: over ten years (regardless of IT 'experience') no issues using KVM, i.e. the possibility that KVM on Linux is basically 'plug and play'—no experience required. That has been my experience. ;-)
Other experiences may be available, but on the basis of mine I would expect that 'the problem' is neither KVM, nor a standard Linux setup.
I suppose there are some wrong choices about. If you want a desktop then a command line driven server distro might be wrong. If you want something easy to install Linux from Scratch or Gentoo might be bad choices. But for the average use Mint, Zorin or Ubuntu would all be reasonable and aethetics might be a determining factor. Nobody should get flamed for choosing any of fthose.
Everyone yammers on about `Linux` like the second coming, but there are far too many choices (false prophets) and niche distros to choose from. There are ***other*** FOSS OSes beyond `Linux`. The BSDs for example have far ***fewer*** choices yet equally capable OSes:
* FreeBSD for more current main stream hardware.
* OpenBSD for the security conscious.
* NetBSD for a variety of current, extremely old, or odd hardware.
`FreeBSD` is often used as base for desktop distros such as `GhostBSD`; there is `NomadBSD` a live desktop distro on a USB stick; also `DragonFly BSD` and `MidnightBSD`.
`NetBSD` might not have any desktop distros, but it has all the packages and instructions necessary to setup an X11 desktop. Similarly for `OpenBSD`.
Wait, what now? When the BSD licence is literally one of the most permissive ones around, even more so than the GPL?
Quoting Wikipedia (yes yes, always the last refuge of the scoundrel, I know), "Most of the current BSD operating systems are open source and available for download, free of charge, under the BSD License."
On what grounds do you consider xBSD "not free software"?
I thought violence was the last refuge of the scoundrel, or was it the incompetent? I'll need to look that up on Wikipedia...
The BSDs, and I've run them here and there since the early 2000s, are all like raw steak. Yes, you can eat it right off of the foam tray, but it takes a skilled cook to get the best out of it. And some butter.
I thought violence was the last refuge of the scoundrel,
"Patriotism", in the original version of the quote. But then you see, Dr Johnson didn't know about Wikipedia, and surely wouldn't have approved of such effortless acquisition of knowledge ("...the dictionary that has taken 18 hours of my every day for the last 10 years. My mother died; I hardly noticed. My father cut off his head and fried it in garlic in the hope of attracting my attention; I scarcely looked up from my work. My wife brought armies of lovers to the house, who worked in droves so that she might bring up a huge family of bastards. I cared not..."), so I merely updated his words for him :)
The problem with the BSDs is, they are distributed under a licence which allows anyone to take your hard work -- which you did so everyone could benefit from it -- and turn it into a proprietary product, by making a small but compatibility-busting change. You then have at least as hard a job keeping it free as you had creating it in the first place.
At least the GPL is militant about not sharing being as bad as stealing.
Read this very carefully and think abut it:
Assuming that the licence is delibertely chosen it reflects the intentions of the licensor.
Those licencing under GPL have particular intentions - that the recipient of a binary should have the right ti receive a copy of the source code, that if (note that word "if" - there's no compulsio to ask for it) they ask for it and duly receive it they have a right to redistribute it under the same terms. By distributing a binary without the source and those conditions might well be regarded as stealing.
Those licencing under BSD simply give away the source without such strings, requiring only attribution and a disclaimer of liability. The inent is different. It is their right to give away their work under those terms if they choose to do so and they do so choose.
Accepting a gift, including BSD's gift is not stealing.
GPLed S/W can be stolen. BSD's is a free gift and cannot be stolen.
This may be difficult to get your head around if you're deeply immersed in the GPL philosphy but please try to see other people's PoV and become a little ore agnostic because the deeply committed attitude is one thing that really puts people off FOSS and does the wider movement no good at all.
My sister's elderly laptop stopped working the other day, and she emailed me to ask what replacement she should buy. I pointed out that Windows 10 was on its way out, and Windows 11 laptops would likely be more expensive htan she was expecting. "Would a Chromebook suit your purposes?" I asked.
A hour later, she had gone to Currys, bought a Lenovo Chromebook, and was on line. She is completely clueless about technology, and has only ever used Windows.
I, on the other hand, have been using Linux almost exclusively for 25 years, and I bought a Raspberry Pi Zero 2 and camera module to use as a security camera. Apalled at the amount of bloat in Raspberry Pi OS when I did a test install, I decided to use Alpine instead. It took me 4 goes to get it right.
That's the difference between ChromeOS and Linux.
Sure. ChromeOS is just about allowing people to sign into Google offerings, so they had to keep it ultra-simple and limited by design (it's Google or nothing).
I think the article was about something which would be simple to use as ChromeOS, but not tied to Google, which is the big problem and would immediately open a fun-sized can of worms: What do you include in that stripped-down OS? For an example, Firefox or Chrome (or maybe some clone of either)?
Whatever you do, it will be "not the right thing" for somebody. Cue 87 different versions with slightly different content/look, and we're back at the "too many Linux distros" problem.
Obligatory xkcd.
Wouldn't that be the difference between buying hardware with the OS already installed, and doing it yourself?
All this talk about what people are going to do when Win10 goes out of support.... They are going to keep using it until they buy a new computer. My parents have never updated the OS on any of their computers. They had Windows 98 (or XP) on 10 year old computers. The only reason my mom uses Win11 is because her Win10 laptop died and had to get a new one. But she also has a Win7 desktop.
To get a ChromeOS 'free' linux equivalent you need a manufacturer to sell it pre-installed. Someone could polish up a lightweight Linux desktop, add LibreOffice and a good browser. But I don't know how you get the average consumer to buy it. It would be like launching a web site with a revenue model that isn't advertisements.
Who's this"you" you're addressing?
Obviously not me. I'm typing this on an Asus laptop. Is that obscure tinkerer H/W? It came with W10 on it - in fact it still it's still there and it's an unpredictable mess that in my judgemetn Only Just Works except when it Just Doesn't Work. I installed Linux on it and it Just Worked.
I use the Linux option because I really can't be arsed tinkering with Windows on a monthly basis to keep it working or reinstalling it when tinkering doesn't succeed.
That's enlightening.
The who is clear. They're referring to Steve Graham, and they are defending Linux against the last part of that comment. Specifically, the part where difficulty installing Alpine on a Raspberry Pi is used to explain problems others would have with Linux when they're completely unrelated. They were right to make that defense.
Of course Alpine is harder to configure than the general Linux distro. It's specifically built to exclude almost everything, and you put back what parts you need. That requires you to know what parts you need and what parts those parts need. A lot of distros don't make you do that, and Steve was not using Alpine because Raspberry Pi OS didn't work, but simply because he wanted something smaller. The average user would not install Alpine, and their choice of Ubuntu or something built on it would include most of the things they'd need from the first installation, with a relatively convenient way of installing more. Not everything, certainly, and for some users there would be some problems getting some functions they couldn''t do without, but a lot of users would not need to tinker with anything to get that experience.
I don't think that's the difference you claim it is. You probably could have obtained a similar effect by installing most stable desktop Linux distros on a former Windows machine and, if necessary which it probably wouldn't be, showing her where the browser icon is. If your sister's laptop had a hardware failure, I can understand why you didn't, but that does not prove whether it would have worked or not.
I've installed Endless OS on a i5/8Gb RAM/256Gb SSD recycled laptop dating from 2014. I installed the full ISO (26Gb iso size).
This full version of Endless OS is a maximalist solution aimed at people without Internet or with metered Internet using refurbished computers and it includes reference materials and some learning resources. Good Dr Proven did a review a while back. Its OSTree based with a modified Gnome as UI which is a bit different to Windows 11. I found it very solid and functional.
I think this new article is after a minimalist solution, 'enough to get on the Internet with a Web browser' to paraphrase OA.
Endless is an amazing distro. It's been my go to for friends and family and any who've asked what they could try instead of windows. It just works and keeps working. My parents have been running it for 4 years now and I've had to intervene once. Most everyone else I don't get support calls from, because it uses chrome browser be default and has loads of app preinstalled. Printing of course can be a challenge, but that's the same for windows users too, just not as much, and printing is less of a thing now anyway. I was surprised that it didn't get a mention in the article.
What seems to being asked for is a ChromeOS replacement that still leverages Google's infrastructure. It's possible that might work in the rather short term, but it's basically unsustainable when your project depends on a service that isn't under your control.
The only way to have a sustainable OS and ecosystem is to have free, locally installed applications. You could *potentially* also do the same with free, non corporate controlled server hosted applications but that has a cost that needs to be accounted for, and access needs to work seamlessly.
Then, even if there's a straightforward desktop with Firefox (obviously not Chrome), LibreOffice, GIMP etc users will still not want to switch because they have a friction free experience with Google Apps, Reddit, Discord, and Whatsapp. All of those are at the risk of being unusable at any moment, and for messaging apps such as Whatsapp the web interface needs linking to a proprietary mobile app on a corporate controlled platform.
You'll need to do better than a 'potential revenue model'. It needs to be a *definite* revenue model, and note that people have extreme reticence to paying for something they currently do not pay money for even when it's in their future best interests. As ChromeOS is 'free' to an end user that's a hard thing to ask.
There needs to be a sea change in end user attitude, but that ship has long sailed. If you use something, pay for it, even (especially) if it is free as in beer. The other alternative is finding a friendly multi millionaire, but as that doesn't seem happen often I'm not about to hold my breath.
"What seems to being asked for is a ChromeOS replacement that still leverages Google's infrastructure."
Not necessarily. There are a couple of possible approaches. One would be a login that asks for a service and a user ID. The service could be Google or Microsoft. It could be a corporate's private server. It could just be your preferred home page, DDG or whatever (in which case you wouldn't need a UID) or your webmail. It could open up a market for new service providers. Liam mentioned NextCloud as if it were a webmail server - it's much more than that: https://nextcloud.com/hub/ and could be the basis for that corporate private service or an independent service provider.
It might even be possible to install local applications but provide the home directory over the net.
It could be several of those depending on what service is chose at login. It would certainly be a benefit to corporate users whoc have to travel to countries with over-curious border officials.
Yes, but not really. Technically you're correct but that doesn't help. First forget about corporate private servers, this is very clearly aimed at end users.
Then, forget about Google and Microsoft because *the entire point* is removing dependence on a third party[1]. If the huge corporate can turn off their service at any point, and there's nothing you can do about it, it's fundamentally not a sane alternative to a corporate driven service such as ChromeOS.
So the only thing left is something such as NextCloud, which is not free. So the money needs to come from somewhere.
'overly curious border officials' is a non point. If your data is that important customs officials shouldn't look at it, then sidestep the issue by not taking your laptop, and use one when you get there.
[1] Yes, you might get away with using a large corporate's resources for a while, and some may consider it a reasonable tradeoff but it's ultimately not a sustainable, guaranteed solution.
> The service could be Google or Microsoft.
Whichever it is, it would make sure the OS is short-lived, as any of them will eventually make their service incompatible with the existing (or shut it down). On the other hand, a "private server" spells "sorry, IT professionals only"... In that case you could as well take any Linux/BSD distro and tailor it to your needs.
To be long-lived a system needs to be independent, so you can keep using that computer till it falls apart or you decide to change it. That also means programs should be local, not in some volatile cloud which any breeze could disperse at any time. (I'm still using some versions of programs I had bought 20 years ago. They are perfectly up to the task, no point in paying for new exciting features I neither need nor want.)
TL;DR: Do you want a KISS-type OS, or just a front end to some short-lived commercial service?
> What seems to being asked for is a ChromeOS replacement that still leverages Google's infrastructure.
Not by my reading of the article; the author's asking for something that Windows 10 users can migrate their PCs to, easily and without any "techy stuff" - something with the usability of ChromeOS. In fact the article's first sentence expressly says, "without a Google account".
The plethora of Linux distributions is overwhelming. However, a large part of the problem is that Windows comes pre-installed, whereas advice to "use Linux" usually has the first step, "choose a distro, download it, write it to a USB drive and install it".
Arguing ad absurdum, if PCs came completely bare and the first instruction to the user was "download Windows after deciding which version you want - Home, Pro, Enterprise - write it to a USB disk and then install, remembering to slipstream network drivers so that you can get online to install any other drivers you'll need"... how popular would Windows be?
Once installed, modern Linux is as powerful, performant, & user-friendly as Windows.
But those words "once installed" are critical! The "dead simple bulletproof all-free OS" that the author seeks may or may not exist among all those Linux distros, but it's not installed on the PCs of the users who need it and they generally don't have the skills to do so.
^ This, 100%, David 132 is absolutely on the money here. That's my understanding of what Liam was driving at, & also my experience as a non-techy (relative to El Reg readership) person making the change from Windows 7 to Mint firstly, & now to MX Linux, where I'll stay as long as it exists. (SystemD free version, for those that care.)
I think that is what Liam should have been driving at, but I think he took a wrong turning. The problem is, as David correctly stated, getting the thing installed. To some extent, it's also getting Linux to do the things they used to do on Windows in such a way that they feel confident about achieving the same goals. Neither of those will be fixed by making Linux more like Chrome OS. The installation problem won't be fixed in the slightest, and it's a pure surrender on the functionality angle because it degrades the machine to the least common feature set of anything else, rather than normal desktop Linux which can do all of that plus lots of functional desktop applications even if we ignore virtualization and emulation.
"I've always attributed that quote to Einstein, but St Ex is as good an author for that."
It's from Terres des Hommes.
See post "St Exupéry biography geofenced"
Einstein did say something similar (presumably in German) and is often (mis?)quoted. Unfortunately I don't have any definitive Einstein reference and may been from a conversation later reported by others.
Who's going to pay for it? Not the development, or the packaging and distribution/hosting aspects, but the cloud aspect.
If you want the decentralised aspects of Chrome OS (i.e. things like remote password storage) without using one of the large providers (who all have other revenue streams.), who do you use?
You have to be signed up with someone if you want the centralisation that makes these devices convenient.
I wouldn't want to run a system like that. Even if it just stored passwords (let alone any other PI) you'll have a lot of regulations and legal obligations that won't be cheap to police, that's before you've started paying for the hosting or the hardware. (and whatever infrastructure you've got to set up). This is not the kind of thing you can set up without a known revenue stream. Will people actually pay for it? Will you get enough donations if you go down the charitable foundation route?
Conversely I wouldn't want to trust my information to a provider that I could see just vanishing one day due to lack of funds.
"who do you use?"
If you're in a corporate environment that could be your own service. ChromeOS is only a thin client and its disadvantage is that it's tied to Google. If you want any sort of control over your own material it really shouldn't be for you. (Neither, of course, should any other remote office provider.)
"I wouldn't want to run a system like that."
Others don't seem so inhibited. Numerous businesses offer NextCloud instances as a service.
"I wouldn't want to run a system like that."
Others don't seem so inhibited. Numerous businesses offer NextCloud instances as a service.
For free?
I may have missed the point, but I read the article as requesting a Google free alternative to Chrome OS for people who are scuppered by the Windows 10 expiration and don't want to have to fork out on new hardware.
Google and MS can provide these services at no upfront cost to the consumer due to their other revenue streams. If someone wants to provide an open source alternative for the average home user they're going to incur costs, so my question is still 'Who pays?'
The system I wouldn't want to run is a 'no upfront cost to user but also no obvious guaranteed revenue' based one.
"If you're in a corporate environment that could be your own service."
If you're in a corporate environment most of this is easier. Choosing a Linux distro is easier because you don't have to ask for opinions. Upgrading Windows versions is easier because there's a budget for replacing older machines. None of this is about a corporate environment, whether or not they choose to use the thing after it's developed. This is for the home user. The home user who needs an operating system that's locked down to only give them a browser tends not to be the one running the services for that OS to connect to. Hence the question of where you would run that and how you would deal with the administrative problems, both financial and regulatory, that would be involved.
In practice, the way that will be handled is that it won't. Some of that will be built in as local software, a client for some server software might get added, and then the devs will leave, assuming that someone else will run that server somewhere and figure out how to do it. Not because the devs can't, but that smart ones realize that this is going to be expensive to run and support and take away time from improving the software.
I am the only technology savvy person connected with a local countryside youth organisation.
We have two problems:
* Windows 10 EOL (no money to replace a perfectly good laptop, bought when "Windows 10 was going to be the last version")
* Microsoft business for charities EOL (the replacement 'basic' is no good as it only seems to work online, which is no good at an event on a farm or in a village hall without broadband or even mobile phone signal!)
The first has been postponed due to the new Oct 26 deadline as I set the laptop up to backup to OneDrive when connected to broadband.
The second is going to be a deal-breaker, but no-one else in the organisation has ever used anything but Microsoft (all learnt it at school), not even Google Docs. I've tried to sell LibreOffice as "like the version of Microsoft tools we all liked 5 years ago", but getting no traction.
I can see it all getting "MEH!" until it all stops and then blame and resentment and constant calls about trivia from the staff member etc.
Not in a position to walk away either.
Windows 10 is only going EOL if you don't pay for extended support. You can purchase up to three years additional security fixes, by which time for the vast majority of businesses any hardware has been written down.
Rather sucks for consumers, of course, but then again it also offers a buffer so that you can move to a free (as in speech) OS and avoid this issue in the future.
Indeed. To Caver_Dave I think the best approach would be to say to the higher-ups in the charity something like,
"After October, there won't be any more free security patches for our computers from Microsoft, which means we'll be at increased risk of ransomware, data loss, or worse. Can you imagine the shame and panic if all our clients' personal data ends up leaked?
Our choices are,
1) pay Microsoft non-trivial amounts per PC per year for patches (which only kicks the can down the road),
2) buy all new PCs, or
3) Switch to Linux and and LibreOffice. I've set up a spare laptop here with them, [skinned to look very similar to Windows, with the ribbon mode turned on in LO - Ed.] and I want each of you to try using it for 24 hours in place of your Windows PC. Yes, it'll be a learning exercise and despite my best efforts, some things will be in slightly unfamiliar places. But you're intelligent adults and I'll be surprised if such minor things flummox you. The alternative options mean spending significant amounts of money that is needed for our primary charitable purpose..."
Or something like that anyway. A charity, of all organizations, should recognise the value of saving every penny. Or maybe I have an old-fashioned view of charities :)
Thank you David 123
There are no higher-ups in the charity. There is the "advisory board" on which I sit as a Vice-President of the charity, made up of the trustees i.e. older ex-members. There is one part-time staff member. And the "executive committee" run by the youngsters, for the youngsters. As the "advisory board" we can make recommendations to the "executive committee" to stop them getting into trouble with the law or Charity Commission, to be the trustees of the charity and employers of the staff member. The members numbered around 300 last year.
Yes, your suggestions 1 and 2 are in line with mine, but either would consume a large percentage of our working capital. Option 3 is a problem when the staff member only works 20 hours a week and struggles to keep up with the admin within that time in addition to the work with the youngsters.
I was hoping that someone would say "this is what we did" or "this is a tutorial online for something similar". Anyway, thank you all that replied.
I'm speculating, but I suspect hard-up charities of the sort described are not the kind of places to "write down" hardware.
Although given the grief laptops get, you may be correct that three years is enough to sort it out through wear and tear, if they can afford the support fees.
Parish and town councils too.
Local government is not allowed to "write down" assets. The audit rules explicitly say that the purchase value goes in the books until it's finally disposed of.
On top of that, most parish councils have a tiny precept (surcharge on council tax) that only just pays the legally-required Proper Officer for a couple of hours work a week.
A replacement laptop in 2025 means a significant tax increase in 2025-2026, which was set in stone in November 2024.
So it's not in the 2025 budget, because they didn't know back in July-September 2025 when that budget was decided.
On top of that, MS charge full whack to local government as they don't qualify for the charity or home editions.
How tech savvy? Can you set up a VM to play with stuff?
If you only have 1 laptop a single licence for Server 2024 is 19 quid as I write this. (You can get trial ISOs to test first, so you could check compatibility).
If you install that with the Desktop Experience you get basically current Windows 11. (It's likely the laptop will be supported). You will probably have to faf with group policy to remove things like the requirement to hit CTRL-ALT-DEL to login, and you'll want to setup non Administrator accounts. (and hide the Admin one too if possible).
Office 2024 lifetime licence is 19 quid too. 2021 is about 12.
That would be somewhere between 30 and 40 quid, depending on the office version you want, (+ your time) to remain supported for the lifetime of Server 2025.
Seconded. I'm very leery of those resellers that have "totally legit" "second hand" software licenses. I might be doing them a grave disservice, but there's always somethng of the allofmp3.com (remember that site? "totally legitimate" MP3 downloads for only pennies each) about the whole business.
Wow. I used to think Fear Uncertanty and Doubt was an MS exclusive thing.
Do either of you have any assertions to back up those claims? You're both essentially accusing a business of theft and deception, maybe a single shred of evidence would help here?
The licences are unused licences as part of a larger purchase. Reselling these is 100% legal in the EU, and I've not heard anything to suggest the UK hasn't adopted this aspect of EU law.
The site I linked to is listed as a partner on the MS site here. This is a little different from the dodgy MP3 sites of old.
The only other thing I can tell you is a friend has recently bought a server licence that has activated fine. I'm intending to do the same to keep my mums machine up-to date. (I'm also seriously considering a 3 licence server key to rid myself of Win 11 without leaving Windows).
I offer nothing to back up my suspicions and I'm delighted to learn that this vendor appears to be quite legit. I just don't think that's the most reasonable assumption when someone offers you a valuable thing at 95% off.
Fair enough, but you could've discovered they were legit by taking 30 seconds to check. The 'Verify us on Microsoft' link is quite prominent.
You didn't do this though, you just libelled them instead. (This isn't the pub, it's an international high traffic website).
It's also not a valuable commodity, it's not hard to get or in short supply, it's just a piece of enterprise software, which have much lower usage rates so tend to command higher prices. Also, volume licencing means companies tend to have the odd licence spare. These are effectively worthless, as if you're buying more than a few seats it makes more sense to take up your own volume licence, and if not you're probably going to use Linux anyway.
Therefore it makes sense to offer these for sale to hobbyists and small businesses at a reduced cost. It keeps the tech savvy windows users happy, and it keeps the MS skill set alive in people who are happy to fiddle for fun.
And at the end of the day I still think it's a good solution for people who want to avoid the cost of new hardware but want to continue to run Windows, especially if you have limited funds like the OP.
If ChromiumOS dies, that will be because nobody steps up to keep it going. So there wouldn't be anybody to step up and find an alternative, either.
Or, turn that around.
If there is enough community to deliver an alternative to ChromiumOS, then there is enough community to fork/maintain the existing.
IMHO what we really need is a standard API layer, so any app will just download and work on anything.
IMHO what we really need is a standard API layer, so any app will just download and work on anything.
You do, you're looking at it. ;-)
Oh... you wanted a non bloated one that hasn't piecemeal evolved over time... yeah.. that's not happening any time soon.
Otherwise, the browser is your standard layer and that will only increase.
I used to boggle at a text editor that could use 700+MB memory for 4 tabs (just checked), but VSCode is my go to text editor and fallback IDE of choice these days, as it's good enough. (It starts promptly, and seems to work well).
Further thoughts in reply to comments:
POSIX would be one possible basis for progress, yes. But it would need some revisiting, as would compatibility layers such as UWIN.
MS 365 and the like notwithstanding, a web browser is not a desktop GUI. People have long tried to absorb it into the host GUI as an HTML desktop, but I am not aware of any success stories in that department. Or have I missed something? There is also the inherent insecurity of rogue cloud apps pwning the whole thing.
There are other possible starting points, but none as yet is any where near a universal baseline.
MS 365 and the like notwithstanding, a web browser is not a desktop GUI.
A web browser isn't a desktop GUI, but the components of a web browser power an increasing number of cross platform applications.
If you want an app that you can write once and run on a mobile phone, an embedded device or a desktop then the arguments for basing it on anything other than an embedded web stack is becoming increasingly hard I'd argue.
Qt is probably the other contender, IIRC you can target QML based stuff at most things these days.
The reality is, things are 'fast enough' now. The company I work for produces hardware. The old hardware was a small 386 machine with a DOS bootloader running embedded linux. The latest hardware (which we've been selling for a good 4 or 5 years now) is based on a raspberry PI and is quite capable of running a web derived stack for it's touch based UI, so it does. This means the same devs can work on all aspects of the app from the embedded UI to the web.
There's so many factors at play here, the biggest one?
I'd say the VAST majority of people who use desktop computers - including laptops - don't know what FOSS is and/or don't care.
The VAST majority of people will become aware of something like the Chromebook due to huge marketing budgets.
Online and brick and mortar.
As far as they are concerned, they switch it on, log in and have a full office suite and a browser - and are good to go.
They'll trust Google with their privacy and data - just like they trust Apple or Samsung - this is the default.
So who does that leave?
People who DO care about these things.
Is that a big enough market to sustain a FOSS solution and smaller laptop manufacturers willing to run that FOSS solution on their hardware and try and market it?
Who knows.
The way I see it, anyone aware of FOSS, aware of privacy concerns ... is probably already running Linux.
I've formed an opinion on this - there's zero market for it, so it will never happen.
"They'll trust Google with their privacy and data"
Sigh.
Of course there are signs in the none-US part of the world that this depencence might not be a good thing. If senior management were to come to their IT departments and ask what the alternatives are the smart IT departments will have done enough precautionary research to have some answers. I wonder how mnay IT departments are that smart.
This is not about IT departments. This is about the home user. IT departments can answer this question because they can state all the options with cost numbers next to them. Figuring out the cost of switching all user desktops to Linux may be tricky, but it's calculable with some effort.
This is for home users, where they don't have someone who can give them the Linux options with a full list of changes they have to make, tradeoffs that apply, and costs of the alternatives. The home user has to figure it out themselves and they don't know all that information. The suggestion in the article is intended to fix that anyway, but it doesn't actually fix it because, instead of trying to help with those issues, it tries to pretend they don't exist and remove the problems instead of solving them. The users' problems tend to be things like "How do I manage my music library which is one of the things I happen to want to do with my computer", and the article's version answers questions they have often never faced like "what if an operating system update breaks the filesystem" (they usually don't).
> At the UI and functionality level, there is nothing in ChromeOS that would be difficult to replicate for any distro vendor.
Not exactly.
The author forgot about security. ChromeOS is the most secure operating system after Android and iOS. Everything is siloed. There's no way for one app (website) to access another's data. (Also, Chromebooks are locked down fully, more than just using the TPM). Linux unfortunately isn't there yet and that's what's missing: a fully locked down distro where the user doesn't have access to their home directory. This prevents things like cookie stealing and more.
You could say "yeah, that's easy, just allow only the browser". But that logic fails when you need to update the OS. Because most Linux updates are per package, not atomic with a dual partition scheme like in ChromeOS. And package updates can and will fail.
A lesser version of ChromeOS is possible, like ChromeOS Flex. But it still needs quite some love. The devil is in the details.
... in exchange of your data.
How could a FOSS OS like that work? Who would pay for the "cloud" infrastructure to keep it running?
What user would need is a real desktop OS that could run just like Windows or macOS, and not coded by people who still believe that punched cards and teletypes were the best devices, and vi is all you need for whatever task.
It's not the year of Desktop Linux, and it will never come, until Linux understand desktop needs...
For Google, it's probably feeding Users + Data into their ad network and AI training.
Because functionally, I can see Mozilla being in the right place to take this on (if they could just get their ethics back in place).
But right now what Mozilla doesn't need is a plan without a solid business case.
And it had nothing to do with the ease of use stuff you're talking about. ChromeOS's big advantage was that was sold PRE-INSTALLED on widely available laptops. The biggest hurdle for the average person to adopt Linux is not the ease of use stuff. Yes that matters but that's not what is holding it back. Because people never get to that point anyway.
The first big hurdle is that there are so many distros. There isn't "Linux" in the way there is "Windows". There is Fedora, Ubuntu, and on and on. People have no idea what "Linux" they should download. Reading articles or talking to friends is no help, because they are looking for ONE answer but they will get many. Articles will list the top 3, top 5 or even top 10 Linux distros. That just turns people off. Ask a friend and you'll probably get multiple suggestions, even if the best answer would be to suggest the same distro they use so that they can easily help their friend if they decide to try it.
Even if they get past that point, having to install it is a huge hurdle. Just the idea is intimidating to the average person. You get into the installer and no matter how easy they make it you still are told you're carving up your Windows install and using its free space for Linux. You worry that it might screw up your Windows. You worry it might take too much space you will need later. You worry it will violate your PC's warranty. And that's if the install itself isn't giving you more choices or asking more questions that make you more nervous. If you think installation is a breeze, try having a friend who isn't a computer whiz install Windows from scratch on a typical PC you buy from Dell or HP or whoever. Windows Update is better than it used to be about finding drivers but too often you still have to dig up drivers to make things work correctly, tweak settings etc. - all stuff that's done for you when you buy a PC with Windows preinstalled. Yes us techies like to install from scratch to avoid the bloatware, but we're alone in that.
Being pre-installed is the main reason why Windows is entrenched. It is the main reason why ChromeOS usage was able to outpace the (desktop) usage of all Linuxes combined without breaking a sweat. An easier to use Linux distro won't change that.
But this isn't about making "Something like GoogleOS without sign-in to Google" a dominant OS - It's about people with an existing PC or laptop having somewhere to go to when Windows 10 reaches end of life. It may become a dominant OS if it serves its purpose, but no one should care if it doesn't.
If installation can be made as simple as download setup.exe and run it, job done.
"If installation can be made as simple as download setup.exe and run it, job done."
That doesn't support the concept in the article, which made no mention whatsoever of ease of installation but did spend time advocating for Chrome OS features (I.E. lock down everything) as the solution to this problem. If installation were as easy as you describe, you could install anything using that method.
And it never will be that simple to install an OS. You have to make choices like:
Do you keep Windows and try to set up a working dual boot, or just destroy it?
Do you copy all the user files to the user's new home directory? What if you don't have enough disk space to manage that?
Do you try to set up emulation for software they have installed? Do you try to find alternatives. Admittedly, the article's proposed feature set makes that simple: you can't run any of those, whether native versions or under emulation.
And you have technical challenges as well, because a program running on Windows will have some annoying workarounds needed when entirely erasing Windows and still having the tools to easily write something else without starting any new sessions. It's possible, as anyone who has ever let "rm -rf /" run to completion can attest, but it increases the workload for writing the installer which booting to the installer and just writing it like a normal program doesn't have.
" If you think installation is a breeze, try having a friend who isn't a computer whiz install Windows from scratch on a typical PC you buy from Dell or HP or whoever. Windows Update is better than it used to be about finding drivers but too often you still have to dig up drivers to make things work correctly,"
On pretty much all recent Dell laptops, the generic Windows USB installer (created using the official Microsoft tool) won't find the SSD - you need to download the storage controller driver package from the Dell support site, extract it then put the driver files on another SSD and manually load them when you get to the select drive section. This is way beyond the average user.
Ah yes, the solution to the windows crowd who asks "Will Linux run my apps?" is surely a Linux distro that runs NO apps. Brilliant.
If you want a distro that runs on old hardware and only needs to run a web browser, there are plenty of options, like Debian, Mint, and Fedora atomic. Speaking of which, atomic distros generally provide rollback capabilities, but those aren't sufficient because... you're scared of git? You don't have to worry about the underlying technology, it's seemless. That's the whole point.
And somehow, "fancy add-on cross-distro package managers like Flatpak" are too complicated, but using a Debian VM in ChromeOS without an app store isn't? Make it make sense.
You want there to be ONE Linux distro for everyone to tell people to use, so that there's "no choice of desktop, no dual boot", etc. Not only is that goal completely unrealistic since there are many good options currently, a new distro that can't do anything other than running a browser would NEVER become "the" Linux distro.
As others have said, ChromeOS became widespread because it came preinstalled on cheap laptops, not because it's so locked down.
> As others have said, ChromeOS became widespread because it came preinstalled on cheap laptops, not because it's so locked down.
This. Windows isn't winning on Desktop computers because its better than Linux. It wins because it's what's on the machine when people buy it.
"If only there was a simpler OS"
"If only it had a more accessible appstore"
"If only the Desktop was better"
"If only it was as simple as Chrome OS"
...if only [argument-goes here] changed about the FOSS OS offerings, surely the masses would flock to it, right? Right?
No. No, they wont.
This isn't about the OS folks. It's not about the appstores. It's not about the usability. Ask yourselves, how many of those non-tech people using windows, have EVER installed windows themselves? Any guesses? Hmm? BARELY ANYONE. How did they set up their computer then? Easy: By virtue of it coming pre-installed from the store.
This is the secret sauce. This is the reason. And while the answer to "What does this come installed with?" is "Windows _", windows will remain the dominant desktop operating system. Period, end of sentence.
Windows doesn't win this because its so much simpler to use than, e.g. Linux. At the surface level, sure. Anything beyond that, and a modern Linux Desktop Env wins easily. Linux also wins on ease of install...I know its en vogue to criticise Manjaro, but have you tried the installer of their KDE version? It's pure bliss, and so simple, my Granny could set up LUKS herself with it.
And even at the surface level, a well put-together KDE desktop, or dare I say it, even GNOME these days, is easily as simple to handle, or even easier, than windows. I mean, short of actually inventing J.A.R.V.I.S from the MCU, how much simpler than Desktop-Metaphore, + start Menu, plus clickable-everything + searchable settings menu + appstore + preinstalled everything + super easy installer can it get?
But all that is completely beside the point. People don't use windows because it so great. People use windows because its what was on the machine when they bought it. Most people aren't in tech, and doing an "OS install" is, to most people, a much scarier prospect than changing the tires on their car...despite the latter being a lot more difficult and dangerous activity.
So, if the FOSS world wants to change that, they need to change how peoples access to consumer hardware works.
This and thousand times this!
MS Windows is incomprehensible. No regular user actually understands how to do anything in MS Windows (and it applies equally to Apple users, if not more). At best they fumble about, achieve something randomly by googling and Brownian motion and then memorise the resulting completely stupid way of doing something as the only way of doing that thing…
The root cause is not that Linux (or any other OS) is complicated. The solution is not to make it simpler.
I know that Dell and Lenovo made half-hearted attempts to support Linux users, but it was just that, a sop. Until we get some sort of major buy-in from the hardware vendors, Linux will remain a fragmented, reluctantly supported option that won't get much traction at the Big Box retailers. That's where it needs to be to shift users away from Windows.
ChromeOS is what Netscape* woulda been shoulda been until - we all know the story - MS decided to kill it. We lost at least 10 more like 20 years.
*Alright I KNOW it was not an OS but it was a start. I mean, look at the thing it was mounted on. How difficult would it have been back then to build the bottom couple of layers for crissake?
https://www.biography.com/authors-writers/antoine-de-saint-exupery
Gives this old AU git: "Sorry, this content is not available in your region."
Via a VPN with a EU end point, works fine.
The Wikipedia article should work for everyone.
https://en.m.wikipedia.org/wiki/Antoine_de_Saint-Exupéry
The quote is from near the end of chapter III, "L'Avion" Terre des Hommes:
Il semble que la perfection soit atteinte non quand il n’y a plus rien à ajouter, mais quand il n’y a plus rien à retrancher.
As a name not exactly marketing coup but does capture essence of the specification.
A lot of Windows users really do fall into the web browser, email and very basic document creation category even those that are big social media users which is mostly accessed via smartphone or tablet devices.
Even the fairly computer illiterate fairly grok ChromeOS.
Technically RetardOS isn't a great challenge. The Linux kernel provides the very broad and ongoing hardware support. The software layers above that become more problematic partly because having everyone agreeing on a single specification makes cat herding appear simple.
ChromeOS and Android to an extent, are so prevalent because a single mega corporation could, and did define and construct the complete hardware and software which following the path of least resistance other parties adopted. But beware of Geeks bearing gifts. ;)
You go out of your way to bash btrfs and recommend avoiding it, so maybe your not aware that when you enable the Linux environment (Crostini) on a Chromebook, the user's files, applications, and settings within that environment are stored within a disk image that uses btrfs.
I used ChromeOS for about 3 months. It's terrible. Settings are split between Android like settings in ChromeOS and settings in the Chrome Browser, which seems to provide the desktop.
Linux Mint with Mate desktop is faster, longer battery life, more flexible and no harder to use on the same Samsung hardware.
The entire point of it is a handful of Google Apps and an Emulator for Android. The Linux support is crippled in a container. The underlying Linux is inaccessible.
The success of ChromeOS is that it is pre-installed on cheap HW and mandated for some Schools and 100% integrated to Google (which is bad).
Nothing to do with how good it is, ease of use or security. It's simple because it's totally inflexible. The Security is more about Google's control.
I think that the demand for a non-W11 replacement for W10 is likely to be rather less than this article, and many others, suggests. Business users will replace, in general, but most private owners will shrug their shoulders and keep on using W10 until they replace the computer, probably because it's broken. Just as they did with XP, for example.
I like my Pixel 5 very much, thank you, and don't give a toss that I no longer get OS updates.
The author seems to have a clear vision and scope in mind, the author should set about doing it.
F/OSS isn't there to do their bidding, but it will enable them to get things done.
If they don't have the technical skills required, they can collect a team of like minded individuals or even pay for people's time.
The result?
There were N Linux distros vying for attention, now there is N+1 Linux distrosvying for attention; and didn't they start out by saying there are too many Linux distros?
ChromeOS works because Google dictates the hardware, manages the updates and offers a useful suite of cloud apps. Hobby crap like ChromeOS Flex and derivatives is for "propeller heads" and not guaranteed to work. Also, that stuff needs to be installed like any other distribution and that is already too much. Anything else than dedicated hardware (with broad IHV buy in) and a centrally managed, pre-loaded OS is going to be a disaster. That is where this Valhalla FOSS OS for the masses is dead in the water.
To offer a ChromeOS experience you have to replicate Google's business model with this. A range of machines with predefined, known to be good hardware. The FOSS world doesn't have the clout to get hardware vendors to buy in. Then you need the resources to manage all the update images for this fleet of hardware. There needs to be lifecycle management. The biggest task will be getting a suite of reliable cloud services going for storage, documents, e-mail, etc. And that needs to be centrally managed as well. No pie-in-the-sky ownCloud stuff. No user of a ChromeOS like machine will accept these roll-your-own, hacky solutions. Anything other than turnkey and management free is off the table. Then the biggest obstacle, this needs to be next to free of cost as much as possible. You get all of this included with a Chromebook purchase.
The absolute biggest hurdle though is Google. They are not killing Chromebooks. They are shifting the ChromeOS desktop on top of Android and you can bet you won't be able to smell the difference. Android is a Linux distribution. So is ChromeOS. What you'll get is the same dead simple desktop OS, but now it can run Android apps natively. On managed hardware. At low cost. Fool proof.
Linux users: "Linux gives you so much choice! It's great! The more the better!"
WIndows users: "Linuxg gives you too much choice! It's terrifying! Can't we just have one single option and no alternatives?"
Liam Proven: "Windows users are right; there are too many Linux options, so let's create another one just for them!" *
Patrick Henry: "Give me libery or give me death!"
Valin Hess (commenting on the Evil Empire in Star Wars): "You see, boys, everybody thinks they want freedom, but what they really want is order."
Douglas Adams: "To summarize the summary of the summary: people are a problem."
Of course they're all right, especialy DNA...
* (The obligatory XKCD reference goes without saying.)
Tons of interesting comments on this article.
Trying not to be TLDR: Windows has become the Edsel/Model-T Ford of Software. Win11 does not provide the neccessary security enhancements that are essential. MacOS is better but too pricey unless on a near or past end of support machine. Linux with a GUI has improved massively but there is something about it that scares the public off. ChromeOS is great but there is one big obstacle namely you cannot plaster the desktop with links to folders, files and apps like using a noticeboard. This practice has been disapproved of since time immemorial but many of the Windows and MacOS junkies absolutely love it. Instead in ChomeOS they have to grapple with Google Drive and the Files app, both of which are nothing like intuative and in fact are close to plain awful.
What is the poor Windows junkie to do? Answers please on a postcard to...
