I've already largely switched away from Windows except on my gaming machine, and I don't think that one's going to be updated any time soon (especially because my CPU is arbitrarily unsupported)....
I've already largely switched away from Windows except on my gaming machine, and I don't think that one's going to be updated any time soon (especially because my CPU is arbitrarily unsupported).
The visual redesign annoys me - the centered icons in particular make me feel that they finally gave up and started copying MacOS - but it's not anywhere near a dealbreaker. What is a dealbreaker is how they've ruined the context menu. Sure, there are a lot of cluttered options, but if the option I need isn't there, then it's completely useless. I'm also pissed off at how they're forcing Teams on everyone, but that's Microsoft's MO at this point.
More than anything, I really don't think that this version of Windows offers anything new or interesting. It feels more like a refresh to keep it 'fashionable' instead of an actual upgrade.
AFAIK you can move the start menu to the right and just remove Teams (it's a good app, makes sense to include it by default). But yeah it seems there's no point in switching right now since...
AFAIK you can move the start menu to the right and just remove Teams (it's a good app, makes sense to include it by default). But yeah it seems there's no point in switching right now since there's not many new features, maybe in a year or two. The android app integration is appealing but not there yet.
Reading the review, the option is still there, but hidden in the "Show more option menu". There's also a linked blog post that explain the rationale behind those changes. But with all that... why...
Sure, there are a lot of cluttered options, but if the option I need isn't there, then it's completely useless.
Reading the review, the option is still there, but hidden in the "Show more option menu". There's also a linked blog post that explain the rationale behind those changes.
But with all that... why can't we chose ourselves what to includes in the context menu ?
The context menus are a necessary change, and it will become much less of an issue as apps that extend it get updated to support the new system. There were fundamental problems with how extending...
The context menus are a necessary change, and it will become much less of an issue as apps that extend it get updated to support the new system.
There were fundamental problems with how extending the context menus in explorer worked dating back to Windows 95 that could not be fixed without breaking compatibility with everything that used it. If you have ever been annoyed at how long a right click often takes in the shell, this is why.
I haven't tried this myself yet, but I believe you can turn off most, if not all, Teams integration in the Settings app.
I don't really care so much about the ability for programs to add new items to the context menu as I do that they have simply removed options from the menu. I also looked up how to add new things...
The context menus are a necessary change, and it will become much less of an issue as apps that extend it get updated to support the new system.
I don't really care so much about the ability for programs to add new items to the context menu as I do that they have simply removed options from the menu.
I also looked up how to add new things to the new context menu and all I could find was a blog post that says that developers can do it, but doesn't actually say how, so it would appear that Microsoft hasn't yet built any form of external extensibility into it yet.
I do agree that the context menu needed a refresh though. I'm not upset that it's new, I'm upset that it's missing features of what it's replacing - to the point that there needs to be a way to circumvent the use of it. That's just bad design.
There were fundamental problems with how extending the context menus in explorer worked dating back to Windows 95 that could not be fixed without breaking compatibility with everything that used it. If you have ever been annoyed at how long a right click often takes in the shell, this is why.
I've never come across that issue before. (Edit: Nevermind, I have experienced this before, but generally it was because the storage media was not spinning at the time).
I've been using the Windows 11 Beta for about a month on my Framework laptop and I really like it overall. But it still has the same problems that Windows 10 does...like falling back to old,...
I've been using the Windows 11 Beta for about a month on my Framework laptop and I really like it overall.
But it still has the same problems that Windows 10 does...like falling back to old, Windows 95 style windows for advanced settings, missing features that were in Windows 10, etc.
I'm reminded of when TiVo moved from their original interface to the HD version. Even years after the transition, there were still menus that were in SD and never updated to the new menu style.
I think GNOME gets a bad rap because a) it looks pretty awful out of the box and b) it ships with Ubuntu which people perceive as being for non-power users and GOME gets lumped in with that. I...
GNOME? I have no idea. I neither use nor like GNOME, though I'd imagine they do a pretty good job at consistency as well, considering the popularity of the desktop environment.
I think GNOME gets a bad rap because a) it looks pretty awful out of the box and b) it ships with Ubuntu which people perceive as being for non-power users and GOME gets lumped in with that. I quite like GNOME and while there are some things I dislike about GNOME 40 overall I find it to be highly productive. But that's also going to be true for every DE you are comfortable with.
Honestly, I think the biggest reason why people don't like GNOME is probably because the design language keeps changing. I mean, KDE has been around for about as long, but if I know how to use a...
Honestly, I think the biggest reason why people don't like GNOME is probably because the design language keeps changing.
I mean, KDE has been around for about as long, but if I know how to use a KDE2 application, I'm not going to have an issue with the current iteration of that application. But the GNOME applications have drammatically changed; menus have disappeared, buttons are now in the window dressing, and everything is "clean" in a way that reminds you of cell phone apps.
And while most GNOME themes that are standard in popular distros are aesthetically pleasing, it's also very different in tone and texture from what you'll find anywhere else, so it has this unique "other" feeling that can be offputting.
This is my key issue. Getting GNOME into a state that I find acceptable requires way too much effort. Removing critical things like the minimize button also does not help with their reputation for...
it looks pretty awful out of the box
This is my key issue. Getting GNOME into a state that I find acceptable requires way too much effort.
Removing critical things like the minimize button also does not help with their reputation for catering to non-power users. It seems like the kind of decision that was made after focus testing in a nursing home, and a number of the participants accidentally clicked it and couldn't figure out where their email went. Another one is the GTK file picker's lack of a text box for pasting or manually typing a path, instead giving you a breadcrumb bar and making you click through folders. Even Windows figured out a better way to manage this 10 years ago.
I disagree there has been consistintency of desktop environment of GNOME and KDE environments (as someone who has used GNOME, KDE, Fedora, Ubuntu, Xubuntu and Kubuntu). Speaking of Linux...
I disagree there has been consistintency of desktop environment of GNOME and KDE environments (as someone who has used GNOME, KDE, Fedora, Ubuntu, Xubuntu and Kubuntu). Speaking of Linux distributions in general they are fragmented alternatives to more consistent desktop environments as MacOS and Windows.
And as with all other things in Linux-space, the styling of QT and GTK apps can be customized and unified into a theme that works with your desktop environment, if you want to take the time to learn how to do that.
Users who use Linux distributions as developers and Linux enthusiast has the time to learn to do that, but mostly Windows and MacOS end users rather pay a solution which does not need to learn how to configure things. As the old say goes Linux is free if you do not value your time.
The problem comes from the fact that you usually end up using software that conforms to multiple different UI toolkits. If you want to use KDE Plasma, you are more or less screwed. There is a lot...
The problem comes from the fact that you usually end up using software that conforms to multiple different UI toolkits.
If you want to use KDE Plasma, you are more or less screwed. There is a lot of good Qt software, but critical apps like Firefox are still GTK. And just because something is written in Qt does not necessarily mean it conforms to Plasma's design language.
You can get a pretty consistent system if you use GNOME or its derivatives and stick to GTK software designed for it. But you will still notice a difference in how "new" and "old" GTK apps are laid out. Modern apps eschew menu bars for hamburger menus or some custom buttons in the window dressing.
Yes, and they have the added problem of apps being gtk or qt, and not matching each other. It seems like apple is the only OS manufacturer that can change their design style and actually change...
Yes, and they have the added problem of apps being gtk or qt, and not matching each other. It seems like apple is the only OS manufacturer that can change their design style and actually change everything in a reasonable time (at least from what I have seen).
I consider this to be a pretty different problem. If you choose to install a KDE application on a GNOME desktop, yeah, it'll look different - but all the core GNOME apps, all the config UIs, all...
I consider this to be a pretty different problem. If you choose to install a KDE application on a GNOME desktop, yeah, it'll look different - but all the core GNOME apps, all the config UIs, all the shell menus, etc look consistent. The same is true with KDE, in reverse. On Windows 8, 10, and 11, the system is actually shipped with multiple different UI styles in core components. It's quite jarring.
Some themes exist that try to bridge the Qt / GTK stylistic divide, but font rendering is noticeably different between them. Setting Qt themes is also a little more painful than GTK themes in my...
Some themes exist that try to bridge the Qt / GTK stylistic divide, but font rendering is noticeably different between them.
Setting Qt themes is also a little more painful than GTK themes in my experience (KVantum? qt5ct? KDE themes != Qt themes?).
I'm a big fan of the new visual design. I have never liked how flat and pointy everything has been in Windows 8 and 10. Bringing back rounded corners, as well as subtle shading and other depth...
I'm a big fan of the new visual design. I have never liked how flat and pointy everything has been in Windows 8 and 10. Bringing back rounded corners, as well as subtle shading and other depth cues is a huge improvement in my book. Even Win32 apps look noticeably improved.
These aren't features I use or care about, but I know they are critical to other peoples workflows so I hope they come back: Moving the task bar to other sides of the screen, and "un-stacking" windows from the same app in the task bar. There is a registry setting to turn on what appears to be experimental support for the former, so I'm betting we will see it show up in a future update.
Back when Win10 first came out, there was a huge outcry/pushback against the privacy invasions that accompanied it -- constant, ubiquitous, un-disable-able user-tracking being sent back to the...
Back when Win10 first came out, there was a huge outcry/pushback against the privacy invasions that accompanied it -- constant, ubiquitous, un-disable-able user-tracking being sent back to the home servers at M$, coupled with often draconian, no-choice forced upgrades and yada. After-the-fact add-in buttons that claimed to give users the choice to opt-out and turn off some (but not all) user data-scraping, often, turned out to be fake buttons that didn't actually turn off anything ...
These are my memories of Win10. I'm probably mis-remembering some of the details, but in general, that was a huge part of the complaints against it at the time. It was enough to finally push me over to all-Linux all-the-time. I have never used Win10, and only maintain 1/2 of one machine with Win7 for game-playing for a friend (who is Warcraft-ing on it, as I type).
All of those complaints may or may not have gone away; I don't know because I frankly quit paying attention to Windows after I switched to Linux. But right now, I'm not seeing much in the way of privacy concerns regarding Win11 ... is that because Win11 is better about this, or because everyone still using Windows has flatly given up the fight?
It's that one. On the Fediverse there has been much discussion of one particular violation - where Windows 10 allowed users to set up the OS without a Microsoft account by setting up the machine...
because everyone still using Windows has flatly given up the fight?
It's that one. On the Fediverse there has been much discussion of one particular violation - where Windows 10 allowed users to set up the OS without a Microsoft account by setting up the machine without an internet connection, Windows 11 requires both an internet connection and a Microsoft account, and that sucks - but I have yet to see it mentioned in mainstream reviews.
Nothing really changed. Win 11 Pro still allows local accounts. Apparently the ability to create local accounts was never intended for Win 10 Home, as they turned it off later. I do have to say,...
Windows 11 Home requires a Microsoft account and Internet connectivity; Windows 11 Pro can still be used with a local account. Windows 10 Home used to let you create a local account as long as you didn't connect to the Internet during setup, but that trick no longer works.
Nothing really changed. Win 11 Pro still allows local accounts. Apparently the ability to create local accounts was never intended for Win 10 Home, as they turned it off later. I do have to say, it doesn't help that Win 10 Home and Pro exist despite being almost the same.
I really think this thread is just going to be a lot of me saying "that don't make it right." That sucks! It sucks! You should be allowed to use your computer that you own without signing up for...
Apparently the ability to create local accounts was never intended for Win 10 Home, as they turned it off later.
I really think this thread is just going to be a lot of me saying "that don't make it right." That sucks! It sucks! You should be allowed to use your computer that you own without signing up for an account you don't want!
I don't know that the fact it was turned off implies that it was never intended for W10. Personally I think it mostly implies that they thought they could get away with it in this stage in the...
I don't know that the fact it was turned off implies that it was never intended for W10. Personally I think it mostly implies that they thought they could get away with it in this stage in the game. They've made several slow steps in that direction, first just requiring navigation of dark patterns, then false phone numbers, then disabling internet during setup, now it's just flat out impossible apparently, at least for the normal consumer.
Probably more because all the concerns about W10 were massively overblown and concerns have finally settled down to a reasonable timbre. Sure, there were reasons to be concerned, but people were...
Probably more because all the concerns about W10 were massively overblown and concerns have finally settled down to a reasonable timbre. Sure, there were reasons to be concerned, but people were addressing it like Windows 10 was Big Brother on steroids. A lot of what you're describing at the start there is basically from the overblown mania. There weren't constant, ubiquitous, draconian privacy invasions, no more than any other major OS or service anyway. No fake buttons either.
Really, Windows 10's big mistake was being upfront with people about all the telemetry and then asking people how much of it they wanted to turn off immediately upon setting up the OS. There's a weird dissonance people seem to have where when they get asked or shown all the choices and details upfront, they get overwhelmed and then overreact. Maybe they should have done what Facebook or Google does and bury that lede, because those companies are far more privacy invasive (and, wildly, are known to be) but somehow avoided the furor that Windows 10 received for so much less.
That whole Windows 10 furor is a big reason why I can't take many so-called privacy advocates without large grains of salt.
I guess I'm not sure what you mean by the sentence I quoted, then. Who care if it's "no more than any other major OS or service" if that's no justification for their behavior?
I guess I'm not sure what you mean by the sentence I quoted, then. Who care if it's "no more than any other major OS or service" if that's no justification for their behavior?
Where did you get the idea I was talking about justification of behaviour in the first place? You took one sentence out of my comment and divorced it from all context in order to treat it as if I...
Where did you get the idea I was talking about justification of behaviour in the first place?
You took one sentence out of my comment and divorced it from all context in order to treat it as if I said something else entirely. Why are you doing this?
I apologize - let me try to explain my understanding of your post. You say: I read this as your thesis - essentially, "the privacy violations in Win10 were not as big of a deal as everyone at...
I apologize - let me try to explain my understanding of your post. You say:
all the concerns about W10 were massively overblown and concerns have finally settled down to a reasonable timbre
I read this as your thesis - essentially, "the privacy violations in Win10 were not as big of a deal as everyone at first thought." You reinforce this by saying that:
there were reasons to be concerned, but people were addressing it like Windows 10 was Big Brother on steroids
You've established that you think the people who were saying Win10's telemetry was bad were wrong. That's great, and you go on to state your counter-point:
There weren't constant, ubiquitous, draconian privacy invasions
That's something that would be reasonable to disagree with, and I do disagree with it. You qualify it with the quote I took:
[the privacy violations were] no more than any other major OS or service anyway
My reading of this is that you're calibrating your idea of "constant, ubiquitous, draconian" (which, in the framework you give here, are the criteria which a privacy violation should satisfy in order for it to be OK for people to be upset about it) against the actions of "other major OS" vendors.
I apologize if I'm misunderstanding, but I'm not sure how to read any other meaning into this post.
There's a misunderstanding. I also said "Sure, there were reasons to be concerned". It's not that people were wrong to be concerned or think telemetry is bad, but that the negative response to it...
You've established that you think the people who were saying Win10's telemetry was bad were wrong.
There's a misunderstanding. I also said "Sure, there were reasons to be concerned". It's not that people were wrong to be concerned or think telemetry is bad, but that the negative response to it was too strong. Like reacting to a papercut with yells and screams of anguish—a papercut is bad, but it's not that bad.
That's something that would be reasonable to disagree with, and I do disagree with it. You qualify it with the quote I took:
My reading of this is that you're calibrating your idea of "constant, ubiquitous, draconian" (which, in the framework you give here, are the criteria which a privacy violation should satisfy in order for it to be OK for people to be upset about it) against the actions of "other major OS" vendors.
Those are actually the adjectives used by the person I was replying to. The idea I'm addressing is that there were not invasions of throughout a usage lifespan of W10 that exceeded what was already done by other OS' and services that would warrant W10 being ascribed those things while others are not. My privacy settings now are the same as I set them as five years ago. The permissions they handle are explicitly described and are adhered to strictly, it seems. I can't see how this telemetry can be described by any of those adjectives unless I am so privacy-protective that any exposure of it is treated at a high DEFCON level.
What I'm getting at is that people treated Windows 10 as if it was going above and beyond the usual amount and thus it had a huge amount of privacy furor, but that it was an unwarranted level. Some privacy furor made sense but level of it, and the consequent reputation W10 got, was not really justified.
And even if one were to describe this regular level of telemetry that all OS' use with the restricted settings I have right now in W10 as those adjectives, I'd still say the language is too strong for the level of privacy that has been yielded. It's a major OS from a major corporation but it's not Big Brother on steroids.
We'll have to agree to disagree. We're clearly interested in different definitions of privacy, and that's fine - but please recognize that I was not taking your words out of context, and in fact I...
We'll have to agree to disagree. We're clearly interested in different definitions of privacy, and that's fine - but please recognize that I was not taking your words out of context, and in fact I don't think you've said anything different in this post than I was originally responding to.
I wonder how they are ever going to do this? I don't see anyone signing up to run proprietary closed source kernel modules and AFAIK pretty much every successful anti-cheat on Windows is running...
I wonder how they are ever going to do this? I don't see anyone signing up to run proprietary closed source kernel modules and AFAIK pretty much every successful anti-cheat on Windows is running at some sort of privileged level.
Every client-side anti-cheat is necessarily flawed and vulnerable to manipulation, it's always going to be a matter of making reasonable trade-offs. Valve has confirmed that they're working...
Every client-side anti-cheat is necessarily flawed and vulnerable to manipulation, it's always going to be a matter of making reasonable trade-offs.
Valve has confirmed that they're working directly with the EAC and (IIRC) BattlEye developers to support Proton, and at least EAC already has a working implementation for native Linux games.
I don't have any particular insider knowledge of how the existing native version works, but I'd expect it to use something fairly simple with code signing and file/memory integrity checks. I wouldn't be surprised if it only works with specific signed versions of Proton, for example.
Any of those checks can theoretically be bypassed, but as with DRM, that's true of anything with enough time.
I doubt they'd go as far as signed kernel modules and requiring secure boot/TPM, at least at this stage, but I think that the people for whom that'd be a hard limit are still a minority among the audience for the Steam Deck.
Yea that's a good point I imagine most people with Steam Deck probably want care even if they do go with a kernel module, and they will probably dwarf the number of people gaming on Linux desktop....
Yea that's a good point I imagine most people with Steam Deck probably want care even if they do go with a kernel module, and they will probably dwarf the number of people gaming on Linux desktop.
Valve has confirmed that they're working directly with the EAC and (IIRC) BattlEye developers to support Proton, and at least EAC already has a working implementation for native Linux games.
Anti-cheat: We recommend using user-space anti-cheat components for best results, as they can typically run in the Wine environment and provide the same level of functionality. Kernel-space solutions are not currently supported and are not recommended. Note: we have been working with most anti-cheat technology providers to provide Proton compatibility. If your solution currently isn't working, please reach out to both your vendor and Valve for support.
It seems to me that the user-space solutions that have been previously recommended are insufficient (hence the platform's reputation for being easier to cheat on) so they are working with EAC and BattleEye to develop a kernel-space solution. At least that's how I'm reading it.
I interpreted that quote as Valve saying they were working to support anti-cheat devs in implementing support for user-space validation of Proton games, more than that they've started work on...
I interpreted that quote as Valve saying they were working to support anti-cheat devs in implementing support for user-space validation of Proton games, more than that they've started work on kernel-space solutions, but I can see how you'd read it that way.
Like I said above, I'm of the view that all client-side anti-cheat is necessarily flawed, and that ring0/kernel-space is unreasonably invasive and a security risk, but that's an area reasonable people can differ.
I do think that Valve has enough weight to convince devs to accept a user-space solution, at least in exchange for listing on the Deck, but I'm sure that the AC platforms will be able to report a different/lower integrity level than something like Riots system for Vanguard on Windows. The whole discussion strikes me as closely analogous to DRM for video content, eg the widevine modules that get used for decrypting streaming content from Netflix/Amazon/etc. If server-browsers were still a thing, I could see a world where, similar to the way VAC works/used-to-work, the server operators could set a required integrity level for clients connecting to their lobby. Matchmaking games could prefer to pair players with similar integrity levels.
There's an argument that you'd potentially end up segregating legitimate Linux users into a pool with more cheaters, but A) dealing with cheaters is as much a human problem as a technical one, B) in my experience Valve at least seems to do a good enough job detecting and removing cheaters from Dota without the use of kernel modules, and C) ultimately I don't really care that much about the occasional cheater when I just want to join my buddies for a couple rounds of Fall Guys.
Yea that makes sense too and could very well be the case. As I outlined in my other reply (and you hint at as well) I'm worried that would make Proton into the cheater platform. Very much agree...
I interpreted that quote as Valve saying they were working to support anti-cheat devs in implementing support for user-space validation of Proton games, more than that they've started work on kernel-space solutions, but I can see how you'd read it that way.
Yea that makes sense too and could very well be the case. As I outlined in my other reply (and you hint at as well) I'm worried that would make Proton into the cheater platform.
Like I said above, I'm of the view that all client-side anti-cheat is necessarily flawed, and that ring0/kernel-space is unreasonably invasive and a security risk, but that's an area reasonable people can differ.
Very much agree with you there. I have no interest in running any of these on my machine. Even if they are well meaning they will increase your attack surface and a closed source kernel module probably has a handful of people auditing it instead of hundreds.
They probably just work with what they got. They clearly aren't invading kernel space and yet both BattleEye and EAC have Proton support now. Not sure if they're running at a reduced confidence...
They probably just work with what they got. They clearly aren't invading kernel space and yet both BattleEye and EAC have Proton support now. Not sure if they're running at a reduced confidence level or what, but they're figuring out out.
If they end up settling on a user-space implementation for EAC and BattleEye while Windows uses kernel-space protections, all the cheaters and cheat-makers are going to migrate to Linux if they...
If they end up settling on a user-space implementation for EAC and BattleEye while Windows uses kernel-space protections, all the cheaters and cheat-makers are going to migrate to Linux if they haven't already. If I were developing a competitive multiplayer game this would make me less inclined to support Linux and perhaps even actively detect wine and prevent people from running it in Proton.
To be clear, I'm not asking for an anti-cheat kernel module. I certainly wouldn't run one on my machine. But I don't see how you're going to have competitive cross-platform multiplayer games on Linux without one.
You wouldn't even need Linux. If there's a codepath in there for win32 via Proton that doesn't demand kernel access then it shouldn't be impossible to hop on that codepath from windows itself.
You wouldn't even need Linux. If there's a codepath in there for win32 via Proton that doesn't demand kernel access then it shouldn't be impossible to hop on that codepath from windows itself.
I've already largely switched away from Windows except on my gaming machine, and I don't think that one's going to be updated any time soon (especially because my CPU is arbitrarily unsupported).
The visual redesign annoys me - the centered icons in particular make me feel that they finally gave up and started copying MacOS - but it's not anywhere near a dealbreaker. What is a dealbreaker is how they've ruined the context menu. Sure, there are a lot of cluttered options, but if the option I need isn't there, then it's completely useless. I'm also pissed off at how they're forcing Teams on everyone, but that's Microsoft's MO at this point.
More than anything, I really don't think that this version of Windows offers anything new or interesting. It feels more like a refresh to keep it 'fashionable' instead of an actual upgrade.
AFAIK you can move the start menu to the right and just remove Teams (it's a good app, makes sense to include it by default). But yeah it seems there's no point in switching right now since there's not many new features, maybe in a year or two. The android app integration is appealing but not there yet.
Reading the review, the option is still there, but hidden in the "Show more option menu". There's also a linked blog post that explain the rationale behind those changes.
But with all that... why can't we chose ourselves what to includes in the context menu ?
I get that. But that reads to me "here's an extra layer of frustration just to annoy you", which is why it's a dealbreaker for me.
The context menus are a necessary change, and it will become much less of an issue as apps that extend it get updated to support the new system.
There were fundamental problems with how extending the context menus in explorer worked dating back to Windows 95 that could not be fixed without breaking compatibility with everything that used it. If you have ever been annoyed at how long a right click often takes in the shell, this is why.
I haven't tried this myself yet, but I believe you can turn off most, if not all, Teams integration in the Settings app.
I don't really care so much about the ability for programs to add new items to the context menu as I do that they have simply removed options from the menu.
I also looked up how to add new things to the new context menu and all I could find was a blog post that says that developers can do it, but doesn't actually say how, so it would appear that Microsoft hasn't yet built any form of external extensibility into it yet.
I do agree that the context menu needed a refresh though. I'm not upset that it's new, I'm upset that it's missing features of what it's replacing - to the point that there needs to be a way to circumvent the use of it. That's just bad design.
I've never come across that issue before. (Edit: Nevermind, I have experienced this before, but generally it was because the storage media was not spinning at the time).
I've come to the realisation that I'm not the target audience of Windows any more and that's okay.
I kind of wish that Microsoft would release Windows for Professionals.
Which is, of course, a Linux distribution with Windows in a VM just like WSL.
I've been using the Windows 11 Beta for about a month on my Framework laptop and I really like it overall.
But it still has the same problems that Windows 10 does...like falling back to old, Windows 95 style windows for advanced settings, missing features that were in Windows 10, etc.
I'm reminded of when TiVo moved from their original interface to the HD version. Even years after the transition, there were still menus that were in SD and never updated to the new menu style.
Isn't this the case for many Linux distributions?
I think GNOME gets a bad rap because a) it looks pretty awful out of the box and b) it ships with Ubuntu which people perceive as being for non-power users and GOME gets lumped in with that. I quite like GNOME and while there are some things I dislike about GNOME 40 overall I find it to be highly productive. But that's also going to be true for every DE you are comfortable with.
Honestly, I think the biggest reason why people don't like GNOME is probably because the design language keeps changing.
I mean, KDE has been around for about as long, but if I know how to use a KDE2 application, I'm not going to have an issue with the current iteration of that application. But the GNOME applications have drammatically changed; menus have disappeared, buttons are now in the window dressing, and everything is "clean" in a way that reminds you of cell phone apps.
And while most GNOME themes that are standard in popular distros are aesthetically pleasing, it's also very different in tone and texture from what you'll find anywhere else, so it has this unique "other" feeling that can be offputting.
This is my key issue. Getting GNOME into a state that I find acceptable requires way too much effort.
Removing critical things like the minimize button also does not help with their reputation for catering to non-power users. It seems like the kind of decision that was made after focus testing in a nursing home, and a number of the participants accidentally clicked it and couldn't figure out where their email went. Another one is the GTK file picker's lack of a text box for pasting or manually typing a path, instead giving you a breadcrumb bar and making you click through folders. Even Windows figured out a better way to manage this 10 years ago.
I disagree there has been consistintency of desktop environment of GNOME and KDE environments (as someone who has used GNOME, KDE, Fedora, Ubuntu, Xubuntu and Kubuntu). Speaking of Linux distributions in general they are fragmented alternatives to more consistent desktop environments as MacOS and Windows.
Users who use Linux distributions as developers and Linux enthusiast has the time to learn to do that, but mostly Windows and MacOS end users rather pay a solution which does not need to learn how to configure things. As the old say goes Linux is free if you do not value your time.
I did not try to insult. Next time i will try to rephrase my opinion in a more polite manner.
The problem comes from the fact that you usually end up using software that conforms to multiple different UI toolkits.
If you want to use KDE Plasma, you are more or less screwed. There is a lot of good Qt software, but critical apps like Firefox are still GTK. And just because something is written in Qt does not necessarily mean it conforms to Plasma's design language.
You can get a pretty consistent system if you use GNOME or its derivatives and stick to GTK software designed for it. But you will still notice a difference in how "new" and "old" GTK apps are laid out. Modern apps eschew menu bars for hamburger menus or some custom buttons in the window dressing.
Yes, and they have the added problem of apps being gtk or qt, and not matching each other. It seems like apple is the only OS manufacturer that can change their design style and actually change everything in a reasonable time (at least from what I have seen).
I consider this to be a pretty different problem. If you choose to install a KDE application on a GNOME desktop, yeah, it'll look different - but all the core GNOME apps, all the config UIs, all the shell menus, etc look consistent. The same is true with KDE, in reverse. On Windows 8, 10, and 11, the system is actually shipped with multiple different UI styles in core components. It's quite jarring.
Some themes exist that try to bridge the Qt / GTK stylistic divide, but font rendering is noticeably different between them.
Setting Qt themes is also a little more painful than GTK themes in my experience (KVantum? qt5ct? KDE themes != Qt themes?).
I'm a big fan of the new visual design. I have never liked how flat and pointy everything has been in Windows 8 and 10. Bringing back rounded corners, as well as subtle shading and other depth cues is a huge improvement in my book. Even Win32 apps look noticeably improved.
These aren't features I use or care about, but I know they are critical to other peoples workflows so I hope they come back: Moving the task bar to other sides of the screen, and "un-stacking" windows from the same app in the task bar. There is a registry setting to turn on what appears to be experimental support for the former, so I'm betting we will see it show up in a future update.
Back when Win10 first came out, there was a huge outcry/pushback against the privacy invasions that accompanied it -- constant, ubiquitous, un-disable-able user-tracking being sent back to the home servers at M$, coupled with often draconian, no-choice forced upgrades and yada. After-the-fact add-in buttons that claimed to give users the choice to opt-out and turn off some (but not all) user data-scraping, often, turned out to be fake buttons that didn't actually turn off anything ...
These are my memories of Win10. I'm probably mis-remembering some of the details, but in general, that was a huge part of the complaints against it at the time. It was enough to finally push me over to all-Linux all-the-time. I have never used Win10, and only maintain 1/2 of one machine with Win7 for game-playing for a friend (who is Warcraft-ing on it, as I type).
All of those complaints may or may not have gone away; I don't know because I frankly quit paying attention to Windows after I switched to Linux. But right now, I'm not seeing much in the way of privacy concerns regarding Win11 ... is that because Win11 is better about this, or because everyone still using Windows has flatly given up the fight?
It's that one. On the Fediverse there has been much discussion of one particular violation - where Windows 10 allowed users to set up the OS without a Microsoft account by setting up the machine without an internet connection, Windows 11 requires both an internet connection and a Microsoft account, and that sucks - but I have yet to see it mentioned in mainstream reviews.
Nothing really changed. Win 11 Pro still allows local accounts. Apparently the ability to create local accounts was never intended for Win 10 Home, as they turned it off later. I do have to say, it doesn't help that Win 10 Home and Pro exist despite being almost the same.
I really think this thread is just going to be a lot of me saying "that don't make it right." That sucks! It sucks! You should be allowed to use your computer that you own without signing up for an account you don't want!
I don't know that the fact it was turned off implies that it was never intended for W10. Personally I think it mostly implies that they thought they could get away with it in this stage in the game. They've made several slow steps in that direction, first just requiring navigation of dark patterns, then false phone numbers, then disabling internet during setup, now it's just flat out impossible apparently, at least for the normal consumer.
Probably more because all the concerns about W10 were massively overblown and concerns have finally settled down to a reasonable timbre. Sure, there were reasons to be concerned, but people were addressing it like Windows 10 was Big Brother on steroids. A lot of what you're describing at the start there is basically from the overblown mania. There weren't constant, ubiquitous, draconian privacy invasions, no more than any other major OS or service anyway. No fake buttons either.
Really, Windows 10's big mistake was being upfront with people about all the telemetry and then asking people how much of it they wanted to turn off immediately upon setting up the OS. There's a weird dissonance people seem to have where when they get asked or shown all the choices and details upfront, they get overwhelmed and then overreact. Maybe they should have done what Facebook or Google does and bury that lede, because those companies are far more privacy invasive (and, wildly, are known to be) but somehow avoided the furor that Windows 10 received for so much less.
That whole Windows 10 furor is a big reason why I can't take many so-called privacy advocates without large grains of salt.
Just because the three big OS companies do it, doesn't make it right.
Nobody said it did.
I guess I'm not sure what you mean by the sentence I quoted, then. Who care if it's "no more than any other major OS or service" if that's no justification for their behavior?
Where did you get the idea I was talking about justification of behaviour in the first place?
You took one sentence out of my comment and divorced it from all context in order to treat it as if I said something else entirely. Why are you doing this?
I apologize - let me try to explain my understanding of your post. You say:
I read this as your thesis - essentially, "the privacy violations in Win10 were not as big of a deal as everyone at first thought." You reinforce this by saying that:
You've established that you think the people who were saying Win10's telemetry was bad were wrong. That's great, and you go on to state your counter-point:
That's something that would be reasonable to disagree with, and I do disagree with it. You qualify it with the quote I took:
My reading of this is that you're calibrating your idea of "constant, ubiquitous, draconian" (which, in the framework you give here, are the criteria which a privacy violation should satisfy in order for it to be OK for people to be upset about it) against the actions of "other major OS" vendors.
I apologize if I'm misunderstanding, but I'm not sure how to read any other meaning into this post.
There's a misunderstanding. I also said "Sure, there were reasons to be concerned". It's not that people were wrong to be concerned or think telemetry is bad, but that the negative response to it was too strong. Like reacting to a papercut with yells and screams of anguish—a papercut is bad, but it's not that bad.
Those are actually the adjectives used by the person I was replying to. The idea I'm addressing is that there were not invasions of throughout a usage lifespan of W10 that exceeded what was already done by other OS' and services that would warrant W10 being ascribed those things while others are not. My privacy settings now are the same as I set them as five years ago. The permissions they handle are explicitly described and are adhered to strictly, it seems. I can't see how this telemetry can be described by any of those adjectives unless I am so privacy-protective that any exposure of it is treated at a high DEFCON level.
What I'm getting at is that people treated Windows 10 as if it was going above and beyond the usual amount and thus it had a huge amount of privacy furor, but that it was an unwarranted level. Some privacy furor made sense but level of it, and the consequent reputation W10 got, was not really justified.
And even if one were to describe this regular level of telemetry that all OS' use with the restricted settings I have right now in W10 as those adjectives, I'd still say the language is too strong for the level of privacy that has been yielded. It's a major OS from a major corporation but it's not Big Brother on steroids.
We'll have to agree to disagree. We're clearly interested in different definitions of privacy, and that's fine - but please recognize that I was not taking your words out of context, and in fact I don't think you've said anything different in this post than I was originally responding to.
As soon as I can run games with anti-cheat in Linux I'll wipe Windows 10 off of my laptop.
I wonder how they are ever going to do this? I don't see anyone signing up to run proprietary closed source kernel modules and AFAIK pretty much every successful anti-cheat on Windows is running at some sort of privileged level.
Every client-side anti-cheat is necessarily flawed and vulnerable to manipulation, it's always going to be a matter of making reasonable trade-offs.
Valve has confirmed that they're working directly with the EAC and (IIRC) BattlEye developers to support Proton, and at least EAC already has a working implementation for native Linux games.
I don't have any particular insider knowledge of how the existing native version works, but I'd expect it to use something fairly simple with code signing and file/memory integrity checks. I wouldn't be surprised if it only works with specific signed versions of Proton, for example.
Any of those checks can theoretically be bypassed, but as with DRM, that's true of anything with enough time.
I doubt they'd go as far as signed kernel modules and requiring secure boot/TPM, at least at this stage, but I think that the people for whom that'd be a hard limit are still a minority among the audience for the Steam Deck.
Yea that's a good point I imagine most people with Steam Deck probably want care even if they do go with a kernel module, and they will probably dwarf the number of people gaming on Linux desktop.
Both of those execute in ring 0 AFAIK? That's where I'm struggling to see how they will work in Linux. Steam has this to say in their documentation: https://partner.steamgames.com/doc/steamdeck/proton
It seems to me that the user-space solutions that have been previously recommended are insufficient (hence the platform's reputation for being easier to cheat on) so they are working with EAC and BattleEye to develop a kernel-space solution. At least that's how I'm reading it.
I interpreted that quote as Valve saying they were working to support anti-cheat devs in implementing support for user-space validation of Proton games, more than that they've started work on kernel-space solutions, but I can see how you'd read it that way.
Like I said above, I'm of the view that all client-side anti-cheat is necessarily flawed, and that ring0/kernel-space is unreasonably invasive and a security risk, but that's an area reasonable people can differ.
I do think that Valve has enough weight to convince devs to accept a user-space solution, at least in exchange for listing on the Deck, but I'm sure that the AC platforms will be able to report a different/lower integrity level than something like Riots system for Vanguard on Windows. The whole discussion strikes me as closely analogous to DRM for video content, eg the widevine modules that get used for decrypting streaming content from Netflix/Amazon/etc. If server-browsers were still a thing, I could see a world where, similar to the way VAC works/used-to-work, the server operators could set a required integrity level for clients connecting to their lobby. Matchmaking games could prefer to pair players with similar integrity levels.
There's an argument that you'd potentially end up segregating legitimate Linux users into a pool with more cheaters, but A) dealing with cheaters is as much a human problem as a technical one, B) in my experience Valve at least seems to do a good enough job detecting and removing cheaters from Dota without the use of kernel modules, and C) ultimately I don't really care that much about the occasional cheater when I just want to join my buddies for a couple rounds of Fall Guys.
Yea that makes sense too and could very well be the case. As I outlined in my other reply (and you hint at as well) I'm worried that would make Proton into the cheater platform.
Very much agree with you there. I have no interest in running any of these on my machine. Even if they are well meaning they will increase your attack surface and a closed source kernel module probably has a handful of people auditing it instead of hundreds.
They probably just work with what they got. They clearly aren't invading kernel space and yet both BattleEye and EAC have Proton support now. Not sure if they're running at a reduced confidence level or what, but they're figuring out out.
If they end up settling on a user-space implementation for EAC and BattleEye while Windows uses kernel-space protections, all the cheaters and cheat-makers are going to migrate to Linux if they haven't already. If I were developing a competitive multiplayer game this would make me less inclined to support Linux and perhaps even actively detect wine and prevent people from running it in Proton.
To be clear, I'm not asking for an anti-cheat kernel module. I certainly wouldn't run one on my machine. But I don't see how you're going to have competitive cross-platform multiplayer games on Linux without one.
You wouldn't even need Linux. If there's a codepath in there for win32 via Proton that doesn't demand kernel access then it shouldn't be impossible to hop on that codepath from windows itself.
Another article about Win11 posted by Ars:
The best part of Windows 11 is a revamped Windows Subsystem for Linux