̛If you check the surroundings, you will see the mess we are in. Virtually (?) there are 0 options for a viable IoT platform, mobile/desktop OS that doesn't annoy us. It could be privacy, locking, you name it, ...
Only recently, I happen to realise, we could have listened to this man little more. Too late then, we are sold already.
The Linux desktop is better than ever (in fact, I am switching to it the first time in a few days! I just need to setup some things first)
Sadly, the enshittification has also taken over parts of the Linux desktop as well though. For example, the mobile-first, flat-everything user-hostile design. (like gnome)
Dark themes were not common while skeuomorphism was mainstream, they are only in demand right now because viewing an extremely low-contrast white flat theme is an eyesore.
Luckily, KDE and the similar still exists and you can theme it:)
This is the 2nd time this word has been used to describe the Linux desktop in this thread, and it's disingenuous. FOSS doesn't do any of the things described with enshitification article: it doesn't sit between buyers and sellers and screw each of them in turn. That's not what's been happening. Maybe the software goes shit, but it is not "enshitification". At every turn there's been alternatives (GNOME3 -> Unity/Xfce, KDE4 -> Trinity, Pulseaudio -> Pulsewire, systemd -> upstart, Debian -> Devuian, etc)
Stop misusing the word, you're discrediting the good work of FOSS.
I am not misusing the word. I don't know what their incentives are, but they are just progressively making things worse and more locked down in the name of "progress".
Bland, corporate, utterly inoffensive and lifeless "design"? Check.
Trying to remove theming from users? Check.
etc....
The only reason why they haven't succeeded like other OSes is because it's FOSS. But they really want to take away the user's choice, shift the Overton window, and pretend like things were always bad.
I call this enshittification, and it is a seemingly inevitable consequence arising from the combination of the ease of changing how a platform allocates value, combined with the nature of a "two-sided market," where a platform sits between buyers and sellers, hold each hostage to the other, raking off an ever-larger share of the value that passes between them.
>shift the Overton window
You're all over the place. This has nothing to do with enshitification.
It is enshittification because they are making the software worse in opposition to the users. They don't care that the users hate the changes, they still do it anyway, deliberately.
Have you considered that many users do like the changes, and you just aren't one of them?
I first used desktop Linux in 1996 and liked it well enough. I remember using GNOME 2 and Xfce around 2010ish and liking both of them well enough, too. And they're still around if I want to use them.
But honestly? I don't use them because I like modern GNOME better. I use it every day, everything just works, and I like it more than the alternatives.
> But honestly? I don't use them because I like modern GNOME better. I use it every day, everything just works, and I like it more than the alternatives
Same here, but with KDE - and I'm a former XFCE user, and before that I was a Gnome user...
Yes, I have, but fundamentally, using old stuff is usually quite hard. If I could revert to it, it would be great, but those are usually not too compatible with today's technology.
I use those despite their horrible UI, not because of their horrible UI.
I respect that, and I hope my question didn't seem like criticism.
I suppose we're fortunate that options like Xfce and MATE are around, but I can relate re: compatibility. Truthfully I'd be just as happy with MATE as with GNOME 3.x, but last time I tried I ran into issues with high DPI and mixed DPI and didn't want to invest time trying to get it all working to my satisfaction.
Thank you very much, and you are amazing! Personally, I am just happy KDE exists and they actually listen to users. There is not much out there in terms of desktop environments like this, sadly:)
I disagree, use the word more to complain about GNOME and what they did to their UI. Even my 512GHz Windows XP Laptop runs more smoothly, and that without SSD.
I also think enshittification is a strong word, and a wrong one.
Because praise the shit. Praise the choice of shit. You like your shit. I like my shit. I enjoy my Gnome shit, I enjoyed Gnome 2 shit but I love Gnome 3 shit. I don't like KDE shit. I can't remember how to spell XKFG shit because it's not big enough to fit in my memory.
It's all choice. And praise choice. I do like Gnome. If you want to keep running Gnome 1 like it's 1999, I'm sure there's a way to do that too.
Well, I am open to changing my mind. What do you think, what's the reason then? From an outsider perspective, 10 years ago I've seen all those beautiful, perfectly usable UIs, and they have almost all disappeared by today. What happened? Why do you think GNOME does that?
I believe its maintainers truly believe they are taking the right decisions that are good for the project.
They limit the feature set so they can stabilize on a smaller basis.
They take UI/UX decisions they think are right.
That's what I think.
I also think they managed to build a beautiful UI that works quite well and that pleases many people.
I also think they removed useful features, lagged on important features (thumbnails in the file picker, which I always found bad anyway, including in the Gnome 2 days). I also think they believe they know better than their users on what they need when they really don't, that they should be more understanding of users trying to work around the flaws instead of despising them, and that they shouldn't both reduce the feature set to a minimum and break extensions in each release. And I do indeed think they made UI like it would be used on tablets, degrading the experience desktop, way too early when they didn't work well on tablet anyway for many reasons and KDE was the only bearable option on tablets at the time.
So I really believe they truly do what they think is best, but I also like the KDE approach way better: listening to the users, trying to polish things while not removing too many features, being humble in their decision, and acknowledging their users might have different needs / taste. (for instance, in Plasma 6, they are reverting to double click by default - many KDE devs prefer simple click and think it's objectively better, but they recognize many users are disturbed by this default.)
KDE is reall pretty good in listening to the users. They announced the removal of workspaces a while ago but then realized some people are actually using it. They reverted their decision on this. Of course people also stepped up to keep maintaining it. That's key in open source projects.
> I also think they believe they know better than their users on what they need when they really don't
This. They have this kind of Apple syndrome of belittling use feedback, minus the part of being a billion dollar megacorporation
Also, things would have gone differently if they forked gnome 2 into a new project and left the original sources intact, instead of dragging all the userbase and distros with them.
To me, this was the shittiest move. Instead of saying "hey, we want to remake gnome into a very opinionated DE so we will make a new project for it", they said "we will use the same gnome2 sources so we will not listen to you but you are forced to listen to us!"
It's possible that your subjective opinion of what makes a good UI differs from the subjective opinion of the people doing the work on Gnome.
I'd need some strong evidence that they are selling you to some other party in order to enrich their investors. Who are these investors? How do they even get profits from Gnome? What choices have been made that let 3rd party companies get better access to you/your data from Gnome itself?
Differences of opinion aren't evidence of intentionally screwing over users in order to get more money for investors.
Your opinion of what a good UI looks like is 10+ years old, GNOME is constantly trying to look like a modern relevant UI. I also hate it, that's why I don't use GNOME and instead use a desktop environment committed to looking like what was popular 20 years ago. I seem to be in the minority, I encounter many people that love GNOME
> the enshittification has also taken over parts of the Linux desktop as well though. For example, the mobile-first, flat-everything user-hostile design. (like gnome)
My first successful experiment with Linux was in 2003 (having flopped in the late 1990s, when Linux could be found on the shelf in stores everywhere). The desktop I used then was Gnome 2. My current desktop is Gnome 2, with more or less the exact same look and functionality, though it's called Mate. It's only enshittification if you have a way to force it on people against their will.
I love Gnome, so not everyone agrees on the "enshittification" claim. It's the most visually consistent DE for Linux, in my experience, and has an aesthetic that I appreciate. It mostly just stays out of my way, but has affordances for the things I need. Fortunately, there's plenty of choice in the Linux world, and we can hopefully coexist without too much hostility.
One big advantage of Linux is that if you don't like gnome, you can use (and contribute to) something else.
And the great thing about e.g. a window manager is that there is no lock-in effect. I must use GitHub because everyone is using GitHub (even thought I like the git e-mail workflow better), I must use WhatsApp because everyone is using WhatsApp, and I must use the Slack and their damn Electron app because my company chose Slack and I don't have a choice.
Gnome does nothing like that: I don't use it, and I don't even know how many people use it. I don't see why there would be a reason to complain about it; if you don't like it, don't use it.
I find the criticism of GNOME a bit weird - personally, I love its user interface because it virtually isn't there. It gets out of the way and when I do see it, it feels clean and well organised. I'm old enough to remember the windows-esque start menu type setups (which I know many DEs still use) and am a bit of a convert to the convergence mindset of GNOME.
To be honest, the only thing that bothers me about Linux desktop environments is that there isn't a simple way of getting a 4G dongle and using my laptop as a phone (calls, SMS, etc).
I agree completely! I ditched Ubuntu for this very fact. GNOME and snaps are deal breakers for me. I switched to Fedora KDE a year ago and I use it every day. And not just for coding. It's a real treat and stuff just works and gets out of my way.
I use KDE because Gnome's multimonitor is always problematic in my environment but I spent a few days on it trying to use it the way they intended it to be used and now have a lot of sympathy for it. It's a really nice simple workflow. This is mostly applicable to work though. I don't know that it would work so well for general purpose home use or even non-technical work for that matter.
you know, I hear this sentiment a lot (linux desktop is now awesome!).
And it may even be true this time around (I love me some KDE).
But for me, it doesn't matter. Once I discovered i3 I just never went back, there is just no desktop that can compare to it.
Obviously I'm not your typical user, so this is an aside from a "nerd", but it's sometimes interesting not having a horse in that race anymore (I've been using i3 for 10-15 years I think at this point?)
Other then in most stuff migrating to the browser how so?
Browser + terminal has been a good combo for a long time, the desktop enviornment only have suffered enshittification due to a push to touch screen oriented UI conventions.
PipeWire made sound finally work flawlessly for me on Linux, Proton made it possible to play a ton of games, graphics drivers being stable, especially for Vulcan and Firefox enabling hardware acceleration.
Some things are worse as you note, but there have been som important improvements.
ASUS - ROG Zephyrus M16 16" 240Hz Gaming Laptop runs Ubuntu 22.04 very well in my opinion: CUDA, Vulkan, Wifi, bluetooth, sleep etc. I installed Linux on a second SSD next to Windows on the original SSD. The version with NVIDIA RTX 4070 costs 1700 USD at Best Buy (the one with RTX 4090 may exceed your budget)
The Linux support for ThinkPads is pretty good I think. I've been using a T14 AMD Gen 1 for three years now with Ubuntu. The hardware is supported by the LVFS too. I'm upgrading to a T14 AMD Gen 4 shortly and will be running Linux (Ubuntu) again.
Any recent Thinkpad will do fine. The Apple Silicon laptops can't run Windows, but can run linux, so there's that, too. Also, Dell XPS laptops do very well with linux.
- wide HW support. 10 years ago it was not guaranteed that suspend will work on a random machine, 15 years ago we were struggling with ndiswrapper to get wifi working.
- professional domain-specific software now available, both free (such as KiCad) and commercial (such as DaVinci Resolve) and supported by vendors on Linux.
Since 1997 here -- still my OS of choice, for desktops, laptops, and servers. I enjoy FreeBSD and NetBSD too, but Debian feels like "home". No surprises, no gotchas, no telemetry, no nonsense.
Every time I have to use my Windows 10 corporate laptop, I become even more bewildered that people actually tolerate such a farce as their daily computing environment.
I used to keep a Windows installation to hand on a disk, but after I realised that I hadn't booted into it for over two years I started using FLOSS operating systems exclusively.
Re. photography, do any specific features of Photoshop come to mind that you miss in free alternatives?
Most of my games don't really work with Proton very well. Also nvidia gpu which is very weird to fix. On this particular computer I run windows with an MX VM where I do most of my work. On other computers I run Debian or MX.
Yet many software developers have personally benefited a lot from FOSS and are aware of the principles behind it.
But when someone is willing to pay us a lot of money, many of us will willingly become deaf to the free software ideals and submit to the corporations in exchange for stable employment.
> many of us will willingly become deaf to the free software ideals
Genuine question: what proportion of developers actually understand the free software ideals? Have you ever tried to go to your colleagues and ask what [choose your open source license] implies?
My experience is that most people think that GPL means that you need to publicly distribute all the code and that BSD means that you can just use it without any attribution.
Well Linux is usable as a desktop operating system. For IoT there are multiple open source project, Home Assistant, ESP Home, just to name a few.
The real problem are mobile operating system. Android phones are nowadays even more and more locked down, and using ROMs without Google Services is nearly impossible.
There's SailfishOS. It still uses Android kernel+drivers, but above that it's a "real" GNU/Linux system (glibc, systemd, bash, Qt, connman+ofono, zypp/packagekit, Gecko). It's not completely FOSS, but it is usable as a daily driver, and has been for at least 10 years (based on personal experience).
> and using ROMs without Google Services is nearly impossible.
Ever hear of GrapheneOS, CalyxOS, /e/ OS, LineageOS, divestOS? I have been using one of those for 2 years now, just like a "normal" Android. I bought my phone with it pre-installed, I didn't have to do anything.
Of course I can't use the apps that require the Google Services, but in my experience that's mostly just stuff like Google Maps (there are great alternatives) and YouTube (there are apps like NewPipe that work really well).
I think it's deeper than that; I think the problem is mobile devices. The OS has to somehow paper-over the fact that there's no mouse, and that everything has to be done with finger-stabs on a 3"x5" screen. That doesn't work with the traditional desktop widgets, so a variety of OS-level widgets and Javascripty plugins is layered on top. But (a) they're not consistent with one-another, and (b) they're not consistent with the desktop metaphor (which isn't going to go away).
Basically, I don't think a phone is suitable for user-input of any complexity. It's a device for selecting content that you then consume passively. It can't be used as a replacement for a desktop. "Mobile first" sounds all very well, but nearly all mobile-first projects have the desktop portion permanently stubbed.
> I think it's deeper than that; I think the problem is mobile devices.
I think it's deeper than that; I think the problem is app ecosystem. More and more apps that you need to use (e.g. the only way to perform 2FA with your bank) are dependent on Google/Apple services, use anti-root, anti-tampering and remote attestation techniques, making them impossible to run on free (libre) mobile OS alternatives.
I just want an IoT plug so I can switch things off and on. It should expose a tcp port and allow me to send packets to it via wifi with some level of authentication. I don't want any cloud shit with it. Just a simple (okay, it's not actually that simple) device that connects to wifi and lets me throw packets at it to control it.
Without GNU, that's just a fever dream.
Many cancers are survivable these days. I hope he has one of those.
You get an ESP8266 micro with wifi plus a power supply, relay, momentary button, current and voltage sense, and a couple LEDs all for about $8. Serial debug and flash headers are broken out for easy access on the PCB.
They ship with chinese firmware but the headers and standard hardware make them dead simple to flash with your own firmware, or ESPhome or Tasmota if you prefer.
> I just want an IoT plug so I can switch things off and on. It should expose a tcp port and allow me to send packets to it via wifi with some level of authentication. I don't want any cloud shit with it. Just a simple (okay, it's not actually that simple) device that connects to wifi and lets me throw packets at it to control it.
Shelly devices offer a firmware that can be controlled trough a REST API locally. Unfortunately it's still proprietary and not open, but it doesn't require a cloud connection.
Otherwise you can buy a device and replace the firmware, there are number of open alternatives, such as ESP Home, Tasmota, etc.
Or... you can build it yourself. Building a smart plug is an easy task, if you have some practice on electrics. You will likely build a better product in terms of safety and capabilities that one you can buy.
All the tools exist to build your own. For example with a esp32-s3 and a custom pcb at one of the low priced fab houses in China. Sure you have to do it yourself.
The primary reason off the shelve products are cloud etc. is because these companies spent the time and money to do the above and since no on wants to pay 100+ for an iot switch they add cloud garbage etc. These products are now sold to the masses and if you have to support them you need control over them or your costs go through the roof.
I am working on a hardware iot product (no cloud) and I have to tape off the USB service port not because there is anything that could go wrong but because people don't read instructions and think the thing will power over USB when there is a power supply included with a barrel plug...
Only recently, I happen to realise, we could have listened to this man little more. Too late then, we are sold already.
Recover well Mr Stallman! I wish you the best!