From Newsgroup: comp.os.linux.advocacy

JJ <jj4public@gmail.com> wrote at 08:23 this Tuesday (GMT):

On Tue, 31 Mar 2026 06:52:29 -0000 (UTC), Lawrence D’Oliveiro wrote:

Ironic, indeed, that a Windows update that is supposed to make Windows
more stable and reliable fails to install reliably
<https://www.tomshardware.com/software/windows/latest-windows-11-update-is-broken-refuses-to-install-microsoft-pulls-latest-update-over-missing-files-error>.

I’m sure the AI they’re using for quality control nowadays told them
it was fine ...

Meh. Auto update feature has become a malware, just like antivirus. They all hinder daily computer use. Doing more bad than good.

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?
--
user <candycane> is generated from /dev/urandom
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?

Two way gamble: you risk making your box easy to crack by not updating,
or you risk bricking your computer by updating.

If you are developing an OS the average everyone uses, and your users
have this dilemma when prompted to update, you have a problem.

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-04-09 11:49 a.m., makendo wrote:

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?

Two way gamble: you risk making your box easy to crack by not updating,
or you risk bricking your computer by updating.

If you are developing an OS the average everyone uses, and your users
have this dilemma when prompted to update, you have a problem.

Which of the currently available operating systems has no chance of
breaking after an update?
--
CrudeSausage
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On Thu, 9 Apr 2026 12:11:37 -0400, CrudeSausage wrote:

On 2026-04-09 11:49 a.m., makendo wrote:

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?

Two way gamble: you risk making your box easy to crack by not updating,
or you risk bricking your computer by updating.

If you are developing an OS the average everyone uses, and your users
have this dilemma when prompted to update, you have a problem.

Which of the currently available operating systems has no chance of
breaking after an update?

That's not the question. Which OS has the greatest probability of
breaking?
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On Apr 9, 2026 at 10:39:11 AM MST, "rbowman" wrote <n3q6hvF8l0kU1@mid.individual.net>:

On Thu, 9 Apr 2026 12:11:37 -0400, CrudeSausage wrote:

On 2026-04-09 11:49 a.m., makendo wrote:

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?

Two way gamble: you risk making your box easy to crack by not updating,
or you risk bricking your computer by updating.

If you are developing an OS the average everyone uses, and your users
have this dilemma when prompted to update, you have a problem.

Which of the currently available operating systems has no chance of
breaking after an update?

That's not the question. Which OS has the greatest probability of
breaking?

Asked AI. Here is what it said:

----------------------------------------------------------------------
## Which OS Has the Highest Update-Break Risk?

Based on historical patterns and general reputation, here's a ranking from
most to least likely to break after an update:

### 1. 🐧 Linux (Rolling Release Distros) — *Highest Risk*
Distributions like **Arch Linux**, **Manjaro**, and **Garuda** push cutting-edge packages continuously. Because packages update independently (kernel, drivers, desktop environment, etc.), version mismatches can brick a system. Manjaro in particular has a reputation for botched updates breaking
GPU drivers or the bootloader.

### 2. 🪟 Windows — *High Risk (for a mainstream OS)*
Windows updates have a notably troubled history:
- **Feature Updates** (major version bumps) are frequent culprits — breaking drivers, removing features, or causing boot loops.
- The forced/automatic update model means users often can't delay problematic patches.
- Infamous examples: KB updates breaking printers, network adapters, or even causing BSODs.

### 3. 🐧 Linux (Stable Distros) — *Moderate Risk*
Distros like **Fedora** (cutting-edge but tested) or **Ubuntu** (LTS) are more stable, but major version upgrades (e.g., Ubuntu 22.04 → 24.04) can still break proprietary drivers (Nvidia especially), PPAs, or custom configurations.

### 4. 🍎 macOS — *Lower Risk*
Apple controls both hardware and software tightly, which reduces breakage. However, major version upgrades (e.g., Ventura → Sonoma) can still break:
- Third-party apps with kernel extensions
- Older hardware performance
- Enterprise security tools

### 5. 📱 iOS / Android — *Lowest Risk*
Mobile OSes are sandboxed and tightly controlled. App-level breakage can
occur, but full system instability after an update is rare.

---

### Bottom Line

| OS | Break Risk | Main Cause |
|---|---|---|
| Arch/Manjaro Linux | 🔴 Highest | Rolling releases, dependency conflicts | | Windows | 🟠 High | Forced updates, driver conflicts |
| Fedora/Ubuntu Linux | 🟡 Moderate | Driver & PPA issues on upgrades |
| macOS | 🟢 Low | Tight HW/SW integration |
| iOS/Android | 🟢 Lowest | Sandboxing, controlled ecosystem |

If **stability** is your priority, **macOS** or an **Ubuntu LTS** release are your safest bets. If you love living dangerously (and learning a lot), run **Arch Linux**. 😄 ----------------------------------------------------------------------

I then asked for evidence... after a bit of back and forth:

----------------------------------------------------------------------
My original ranking — particularly placing Arch Linux at the top — is still largely based on community reputation, not hard data. The academic studies focused almost exclusively on Windows, so a rigorous cross-OS comparison
simply doesn't exist in the public literature. The honest conclusion is: Windows has well-documented, verifiable update failures, but whether it's objectively worse than rolling-release Linux distros remains unproven by empirical study. ----------------------------------------------------------------------

I think that is fair. I pushed it on macOS specifically:

----------------------------------------------------------------------
Bottom line: The original "low risk" label for macOS was too generous. It breaks less catastrophically than Windows, but it breaks often enough that power users routinely advise waiting 2–4 weeks before applying major macOS updates. There is still no rigorous cross-platform empirical study that quantifies this difference objectively. ----------------------------------------------------------------------

Again, all seems reasonable by my experience.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-04-09 1:39 p.m., rbowman wrote:

On Thu, 9 Apr 2026 12:11:37 -0400, CrudeSausage wrote:

On 2026-04-09 11:49 a.m., makendo wrote:

I'm sure it gives a rush for all the gambling guys, though. Will my
computer be broken today?

Two way gamble: you risk making your box easy to crack by not updating,
or you risk bricking your computer by updating.

If you are developing an OS the average everyone uses, and your users
have this dilemma when prompted to update, you have a problem.

Which of the currently available operating systems has no chance of
breaking after an update?

That's not the question. Which OS has the greatest probability of
breaking?

Linux. I would suggest that's why distributions like Pop_OS! make it
very easy to fix your installation when it inevitably fails. In Linux
Mint, the possibility is minimized because their updating system makes
it clear which updates are essential and which can be ignored. While I
would be lying in saying that Windows never broke on me, the reality is
that Linux broke on me way more. Modern Fedora broke on me _every_ time
I would update it because of how it handles the NVIDIA drivers.
Distributions like Ubuntu break because a library was updated here for
on piece of software you installed, but a specific version was needed
for something being used by the system or vice versa. That's probably
part of why a lot of people are happy to switch to Flatpak over what the distribution itself offers.
--
CrudeSausage
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 09 Apr 2026 18:30:58 GMT, Brock McNuggets wrote:

Which OS has the greatest probability of breaking?

Asked AI. Here is what it said:

So what do *you* believe?
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On Apr 9, 2026 at 4:12:06 PM MST, "Lawrence D´Oliveiro" wrote <10r9bo5$m58a$4@dont-email.me>:

On 09 Apr 2026 18:30:58 GMT, Brock McNuggets wrote:

Which OS has the greatest probability of breaking?

Asked AI. Here is what it said:

So what do *you* believe?

...

Again, all seems reasonable by my experience.
--
It's impossible for someone who is at war with themselves to be at peace with you.
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On Thu, 9 Apr 2026 15:49:50 -0400, CrudeSausage wrote:

Linux. I would suggest that's why distributions like Pop_OS! make it
very easy to fix your installation when it inevitably fails. In Linux
Mint, the possibility is minimized because their updating system makes
it clear which updates are essential and which can be ignored. While I
would be lying in saying that Windows never broke on me, the reality is
that Linux broke on me way more. Modern Fedora broke on me _every_ time
I would update it because of how it handles the NVIDIA drivers.
Distributions like Ubuntu break because a library was updated here for
on piece of software you installed, but a specific version was needed
for something being used by the system or vice versa. That's probably
part of why a lot of people are happy to switch to Flatpak over what the distribution itself offers.

That's your experience. Mine is different partially because I have no
Nvidia devices.
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

At 10 Apr 2026 03:23:17 GMT, rbowman <bowman@montana.com> wrote:

On Thu, 9 Apr 2026 15:49:50 -0400, CrudeSausage wrote:

Linux. I would suggest that's why distributions like Pop_OS! make
it very easy to fix your installation when it inevitably fails. In
Linux Mint, the possibility is minimized because their updating
system makes it clear which updates are essential and which can be
ignored. While I would be lying in saying that Windows never broke
on me, the reality is that Linux broke on me way more. Modern
Fedora broke on me _every_ time I would update it because of how it
handles the NVIDIA drivers. Distributions like Ubuntu break because
a library was updated here for on piece of software you installed,
but a specific version was needed for something being used by the
system or vice versa. That's probably part of why a lot of people
are happy to switch to Flatpak over what the distribution itself
offers.

That's your experience. Mine is different partially because I have no
Nvidia devices.

Linux Mint has a driver manager that takes the guesswork out of
enabling NVIDIA's drivers, or switching back to nouveau.

The only reason you hear me talk about NVIDIA driver issues: I've
departed from the Linux Mint distro to run test kernels, and they
need the latest drivers to work. (And sometimes, drivers won't
run on rc kernels without modification...thankfully, that's not
going on now.)

What _is_ going on with the newest drivers is the interference
with XFCE compositing, which is solved by turning off vblank.
--
-v ASUS TUF DASH F15 x86_64 Mem: 15.9G
OS: Linux 6.17.0-20-generic D: Mint 22.3 DE: Xfce 4.18 (X11)
NVIDIA GeForce RTX 3060 Mobile (6G) 580.126.09
"Never test for an error you don't know how to handle."
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

On 2026-04-09 11:23 p.m., rbowman wrote:

On Thu, 9 Apr 2026 15:49:50 -0400, CrudeSausage wrote:

Linux. I would suggest that's why distributions like Pop_OS! make it
very easy to fix your installation when it inevitably fails. In Linux
Mint, the possibility is minimized because their updating system makes
it clear which updates are essential and which can be ignored. While I
would be lying in saying that Windows never broke on me, the reality is
that Linux broke on me way more. Modern Fedora broke on me _every_ time
I would update it because of how it handles the NVIDIA drivers.
Distributions like Ubuntu break because a library was updated here for
on piece of software you installed, but a specific version was needed
for something being used by the system or vice versa. That's probably
part of why a lot of people are happy to switch to Flatpak over what the
distribution itself offers.

That's your experience. Mine is different partially because I have no
Nvidia devices.

It worked well on my AMD-only laptop. However, if you're looking to buy
a gaming laptop, there are absolutely no choices that include an AMD GPU.
--
CrudeSausage
M4 Air
--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.os.linux.advocacy

Even Windows Server is not immune <https://www.tomshardware.com/software/windows/microsofts-april-patch-puts-windows-domain-controllers-into-reboot-loops>. It’s not just one bug:

KB5082063 now has three acknowledged bugs within a week of
release, and Microsoft has warned separately that the same update
prompts some Windows Server 2025 machines for a BitLocker recovery
key after installation. The company is investigating reports that
KB5082063 fails to install entirely on a subset of Windows Server
2025 systems.

April security updates have disrupted Windows Server domain
controllers for three consecutive years. In March 2024, Microsoft
shipped an emergency out-of-band fix after that month's Patch
Tuesday caused DC crashes outright. The April 2024 patch cycle
then broke NTLM auth across Windows Servers and forced unplanned
DC restarts, which Microsoft corrected in a May 2024 rollout.

In June last year, the company released another correction for
Active Directory authentication problems introduced by the April
2025 security update. This month's LSASS crash follows the same MO
for the third year running: a general Patch release followed by
post-deployment failure reports from enterprise admins, and a
scramble for mitigation while the fix is prepared.

Aren’t they charging enough for that high-priced product to make
quality support worthwhile?

My guess is, the answer is now “no”. Windows Server sales have been so impacted by Linux that the return on investment just isn’t there any
more.
--- Synchronet 3.21f-Linux NewsLink 1.2