Frank Slootweg <this@ddress.is.invalid> wrote:

>> >User-Agent: tin/1.6.2-20030910 ("Pabbay") (UNIX) (CYGWIN_NT-10.0-WOW/2.8.0(0.309/5/3) (i686)) Hamster/2.0.2.2
>>
>> Fascinating that you prefer using that for Usenet, I'm the reverse,
>> using Forte Agent under Wine, lol.
>
> I started with 'notes' ('notesfiles', not Lotus Notes) on Unix/UNIX
>(HP-UX), so tin was the logical next step. Anyway, why would I want to
>use a GUI+'mouse' for a pure text medium? No need to go forwards and
>backwards between keyboard and mouse. But that's just my choice/
>preference.

Agent is the flagship (GUI) app other than the ones like you're using.

> As to Cygwin, for my personal use, Windows programs and all GNU tools
>is the best combination, ever since early 2003. WSL came way too late
>for me considering to switch to it.

Well, whether it's Cygwin or WSL 2 or Windows 7 Ultimate for that
matter, it's still an intriguing thing to be running in a rather more
profound sense than me running Agent under Wine.

>> >Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>> >> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>> >>
>> >> > Beta software is beta software. An OS is an OS.
>> >>
>> >> And Microsoft is the only one shipping a beta-quality OS and expecting its
>> >> users to rely on that for mission-critical production work.
>> >
>> > Don't be silly! Any such 'users' are total idiots. The consumer
>> >versions of Windows are totally unfit for "mission-critical production
>> >work". And even business/enterprise versions need competent IT staff and
>> >procedures to properly manage such systems.
>> >
>> > FYI, I've been supporting Five Nines systems (obviously not Windows
>> >ones), so remarks like yours tend to push my button.
>>
>> That's a good point - the business Winblows users will not be using
>> 24H2, in the first place, M$ is testing it in the wild with early-
>> adopter types, masochists one might surmise.
>
> Indeed. I only updated to 24H2 after some months passed and most of
>the dust had settled. Update went fine and the power efficiency
>enhancements indeed seem to make my laptop's battery to last longer.
>Who would have thought that! :-)

Good that it is functioning properly for your machine.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On 2025-01-01 14:34, Joel wrote:
> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>
>>> Beta software is beta software. An OS is an OS.
>>
>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>> users to rely on that for mission-critical production work.
>
>
> Yeah, weird, huh? It's like they can't make it right from the
> beginning, they need people to put up with bugs to test it in the
> wild, how is their internal testing so poor? Being on the cutting
> edge with M$ is just spinning one's wheels. Linux is the refuge.

If Windows 11 didn't routinely become unbootable from an update that
Microsoft didn't test under certain conditions, I'd disagree with Joel
here. However, it seems that it happens with every one of their
cumulative updates.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

Andrzej Matuch <andrzej@matu.ch> wrote:
>On 2025-01-01 14:34, Joel wrote:
>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>>
>>>> Beta software is beta software. An OS is an OS.
>>>
>>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>>> users to rely on that for mission-critical production work.
>>
>> Yeah, weird, huh? It's like they can't make it right from the
>> beginning, they need people to put up with bugs to test it in the
>> wild, how is their internal testing so poor? Being on the cutting
>> edge with M$ is just spinning one's wheels. Linux is the refuge.
>
>If Windows 11 didn't routinely become unbootable from an update that
>Microsoft didn't test under certain conditions, I'd disagree with Joel
>here. However, it seems that it happens with every one of their
>cumulative updates.

Seemingly, they aren't really testing the code, much less testing
deployment, because the kinds of bugs just don't reflect an odd
incompatibility but rather that specific PC systems are treated in an
erroneous manner, M$ can't manage its bloated OS, I guess. They do a
passable job on the whole, though, but certain customers are plagued
by the bugs.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On Wed, 1 Jan 2025 10:06:35 -0500, Andrzej Matuch wrote:

> I won't deny the symbiotic relationship between the two, but I don't
> believe that it was any kind of conspiracy behind it. Intel was making
> the x86 chips, and IBM and Microsoft's software ran on that platform.
> While AMD worked for Intel to produce x86 chips and even after they no
> longer worked together, Microsoft ran just as well on AMD's x86 chips.
> People have always criticized the x86 platform for not being as good a
> the RISC variants, but it doesn't mean that there wasn't a need for x86.

I agree. IBM chose the 8088 in part because they had experience with the
8085 on the System/23. The rest is history. Intel was late to the mobile
party which didn't do MS any favors. An example is the Atom processors.
Intel made a lot more money on Core processors and lost interest in the
Atom. That killed the cheap Surface line a Surface 3. The Pro series used
Core processors and had a higher price.

MS muddied the waters with the ARM powered Surface 1 and 2 and RT that
left a bad taste.

At the moment Intel seems to be imploding as MS sails on.

On Wed, 01 Jan 2025 14:34:06 -0500, Joel wrote:

> Yeah, weird, huh? It's like they can't make it right from the
> beginning, they need people to put up with bugs to test it in the wild,
> how is their internal testing so poor? Being on the cutting edge with
> M$ is just spinning one's wheels. Linux is the refuge.

I really hate defending Microsoft but when they do a release how are they
supposed to test every possible combination of hardware and software?
Linux has gotten better but it was problematic on laptops and there were/
are lists of Linux friendly laptops. The Windows world assumes it's going
to work.

Apple has a much easier task since they know exactly what the software is
going to run on.

On Wed, 01 Jan 2025 15:24:22 -0500, Joel wrote:

> That's a good point - the business Winblows users will not be using
> 24H2, in the first place, M$ is testing it in the wild with early-
> adopter types, masochists one might surmise.

Don't bet on it. I updated both of my work Windows machines to 24H2 last
month. IT isn't forcing it yet but they get antsy if machines aren't
updated. Windows 10 is completely verboten. Any box that couldn't run 11
was wiped. I got a new Linux box out of it. The head of IT does his best
to pretend Linux doesn't exist.

rbowman <bowman@montana.com> wrote:

>> Yeah, weird, huh? It's like they can't make it right from the
>> beginning, they need people to put up with bugs to test it in the wild,
>> how is their internal testing so poor? Being on the cutting edge with
>> M$ is just spinning one's wheels. Linux is the refuge.
>
>I really hate defending Microsoft but when they do a release how are they
>supposed to test every possible combination of hardware and software?
>Linux has gotten better but it was problematic on laptops and there were/
>are lists of Linux friendly laptops. The Windows world assumes it's going
>to work.
>
>Apple has a much easier task since they know exactly what the software is
>going to run on.

In part that is true analysis, yeah.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On 1 Jan 2025 21:14:06 GMT, Frank Slootweg wrote:

> As to Cygwin, for my personal use, Windows programs and all GNU tools
> is the best combination, ever since early 2003. WSL came way too late
> for me considering to switch to it.

I don't think it has been maintained in a while but there was a native
Windows build of Unix utilities that didn't depend on Cygwin or MSYS2. I
get tired of typing ls and finding nothing there.

I haven't used Cygwin recently. Speaking of way too late I started with
DJGPP which ported gcc to Windows. There were two threads growing out of
that. Corrina Vinschen and others took the Cygwin branch which more to
create a Unix environment of Windows while Colin Peters, and later Mumit
Khan took the mingw32 branch to use gcc to build native Windows programs
that became MSYS2 over the years. It was an interesting era.

At work we used the MKS NutCracker tools to port what were originally AIX
programs to Windows. While Cygwin would have probably worked it would have
gotten into the whole GPL limitations on commercial software morass.

On 2025-01-01 18:30, Joel wrote:
> Andrzej Matuch <andrzej@matu.ch> wrote:
>> On 2025-01-01 14:34, Joel wrote:
>>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>>>
>>>>> Beta software is beta software. An OS is an OS.
>>>>
>>>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>>>> users to rely on that for mission-critical production work.
>>>
>>> Yeah, weird, huh? It's like they can't make it right from the
>>> beginning, they need people to put up with bugs to test it in the
>>> wild, how is their internal testing so poor? Being on the cutting
>>> edge with M$ is just spinning one's wheels. Linux is the refuge.
>>
>> If Windows 11 didn't routinely become unbootable from an update that
>> Microsoft didn't test under certain conditions, I'd disagree with Joel
>> here. However, it seems that it happens with every one of their
>> cumulative updates.
>
>
> Seemingly, they aren't really testing the code, much less testing
> deployment, because the kinds of bugs just don't reflect an odd
> incompatibility but rather that specific PC systems are treated in an
> erroneous manner, M$ can't manage its bloated OS, I guess. They do a
> passable job on the whole, though, but certain customers are plagued
> by the bugs.

The bug that finally caused me to stop bothering with them isn't
Microsoft's fault, but I do fault them for continuing to demand TPM when
the requirement was causing stuttering with anyone who owned an AMD
processor. Like I said, if you were aware of what was causing the
problem and your motherboard allowed you to install a hardware TPM, you
were clear as long as you didn't mind paying to buy one, assuming that a
BIOS update wasn't available. However, most people could _at best_
disable it and lose the ability to install 11. If they knew about Rufus,
they could try to circumvent the problem and hope that Microsoft didn't
screw them over in an update. For laptop owners though, it's an absolute
mess that the manufacturers have no interest in fixing. There is no
excuse for any manufacturer to be aware of such an issue for years and
not do a thing about it. Additionally, there is no excuse for Microsoft
to acknowledge this reality and not at least allow AMD users to use 11
without TPM if not disable the hwrng causing the issue. Anyways, I'm
tired of being treated like I don't matter because they already got my
money.

At least the manufacturer replaced my motherboard when I intentionally
fucked it up on the day before my warranty ended. It was becoming clear
that the fingerprint reader wasn't working because the technician
shorted the motherboard during the first repair when they replaced my
battery, so I wanted a replacement and I wouldn't get it unless I killed
it.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

On 2025-01-01 18:31, rbowman wrote:
> On Wed, 1 Jan 2025 10:06:35 -0500, Andrzej Matuch wrote:
>
>> I won't deny the symbiotic relationship between the two, but I don't
>> believe that it was any kind of conspiracy behind it. Intel was making
>> the x86 chips, and IBM and Microsoft's software ran on that platform.
>> While AMD worked for Intel to produce x86 chips and even after they no
>> longer worked together, Microsoft ran just as well on AMD's x86 chips.
>> People have always criticized the x86 platform for not being as good a
>> the RISC variants, but it doesn't mean that there wasn't a need for x86.
>
> I agree. IBM chose the 8088 in part because they had experience with the
> 8085 on the System/23. The rest is history. Intel was late to the mobile
> party which didn't do MS any favors. An example is the Atom processors.
> Intel made a lot more money on Core processors and lost interest in the
> Atom. That killed the cheap Surface line a Surface 3. The Pro series used
> Core processors and had a higher price.
>
> MS muddied the waters with the ARM powered Surface 1 and 2 and RT that
> left a bad taste.
>
> At the moment Intel seems to be imploding as MS sails on.

Yeah, I hear you about Intel. I wasn't aware of how bad it was doing
until I watched a video on YouTube a while ago where the producer showed
that the company had neglected most categories and had their crown
stolen from them. They're not in consoles, nobody wants them for mobile
computing and gamers seem to prefer AMD.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

On 2025-01-01 18:38, rbowman wrote:
> On Wed, 01 Jan 2025 14:34:06 -0500, Joel wrote:
>
>> Yeah, weird, huh? It's like they can't make it right from the
>> beginning, they need people to put up with bugs to test it in the wild,
>> how is their internal testing so poor? Being on the cutting edge with
>> M$ is just spinning one's wheels. Linux is the refuge.
>
> I really hate defending Microsoft but when they do a release how are they
> supposed to test every possible combination of hardware and software?

No, but they should give people the option to take a wait and see
approach rather than force updates on everyone who didn't pay for a Pro
license. Some people want every update the moment it's available; others
like my wife want to watch the other computer crash and let the
corporation fix the issue before she installs them.

> Linux has gotten better but it was problematic on laptops and there were/
> are lists of Linux friendly laptops. The Windows world assumes it's going
> to work.
>
> Apple has a much easier task since they know exactly what the software is
> going to run on.

I have yet to see an Apple update screw anything up but it is definitely
easier for them since there aren't as many MacOS configurations to consider.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

Andrzej Matuch <andrzej@matu.ch> wrote:
>On 2025-01-01 18:30, Joel wrote:
>> Andrzej Matuch <andrzej@matu.ch> wrote:
>>> On 2025-01-01 14:34, Joel wrote:
>>>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>>>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>>>>
>>>>>> Beta software is beta software. An OS is an OS.
>>>>>
>>>>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>>>>> users to rely on that for mission-critical production work.
>>>>
>>>> Yeah, weird, huh? It's like they can't make it right from the
>>>> beginning, they need people to put up with bugs to test it in the
>>>> wild, how is their internal testing so poor? Being on the cutting
>>>> edge with M$ is just spinning one's wheels. Linux is the refuge.
>>>
>>> If Windows 11 didn't routinely become unbootable from an update that
>>> Microsoft didn't test under certain conditions, I'd disagree with Joel
>>> here. However, it seems that it happens with every one of their
>>> cumulative updates.
>>
>> Seemingly, they aren't really testing the code, much less testing
>> deployment, because the kinds of bugs just don't reflect an odd
>> incompatibility but rather that specific PC systems are treated in an
>> erroneous manner, M$ can't manage its bloated OS, I guess. They do a
>> passable job on the whole, though, but certain customers are plagued
>> by the bugs.
>
>The bug that finally caused me to stop bothering with them isn't
>Microsoft's fault, but I do fault them for continuing to demand TPM when
>the requirement was causing stuttering with anyone who owned an AMD
>processor. Like I said, if you were aware of what was causing the
>problem and your motherboard allowed you to install a hardware TPM, you
>were clear as long as you didn't mind paying to buy one, assuming that a
>BIOS update wasn't available. However, most people could _at best_
>disable it and lose the ability to install 11. If they knew about Rufus,
>they could try to circumvent the problem and hope that Microsoft didn't
>screw them over in an update. For laptop owners though, it's an absolute
>mess that the manufacturers have no interest in fixing. There is no
>excuse for any manufacturer to be aware of such an issue for years and
>not do a thing about it. Additionally, there is no excuse for Microsoft
>to acknowledge this reality and not at least allow AMD users to use 11
>without TPM if not disable the hwrng causing the issue. Anyways, I'm
>tired of being treated like I don't matter because they already got my
>money.

I think Microsoft genuinely feels that their barely supporting
hardware, in any real sense, is the natural way, that if people are
too stupid or stubborn to try Linux on a somewhat older system, they
get what they deserve, bloat that overloads their gear. It's really
amazing, how attached people are to Winblows software, and how
intimidated they are by anything Linux (I mean they can try Mint,
FFS), M$ and the OEMs just suck their money year after year.

>At least the manufacturer replaced my motherboard when I intentionally
>fucked it up on the day before my warranty ended. It was becoming clear
>that the fingerprint reader wasn't working because the technician
>shorted the motherboard during the first repair when they replaced my
>battery, so I wanted a replacement and I wouldn't get it unless I killed
>it.

You mean that you made an already malfunctioning board completely
dead, because you didn't want the fingerprint reader alone repaired,
but a replacement of the board? I guess I can't fault you, for that,
to me the idea of a warranty claim is the ultimate in drudgery, which
is why I purchased the wireless charger for my phone when its USB-C
stopped working under warranty, no fucking way am I playing that game
where they issue me a replacement and refurbish and profit from my
original one, fuck them, I'd rather give Amazon the few bucks for an
alternative charging mechanism.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On Wed, 1/1/2025 7:19 PM, rbowman wrote:
> On 1 Jan 2025 21:14:06 GMT, Frank Slootweg wrote:
>
>> As to Cygwin, for my personal use, Windows programs and all GNU tools
>> is the best combination, ever since early 2003. WSL came way too late
>> for me considering to switch to it.
>
> I don't think it has been maintained in a while but there was a native
> Windows build of Unix utilities that didn't depend on Cygwin or MSYS2. I
> get tired of typing ls and finding nothing there.
>
> I haven't used Cygwin recently. Speaking of way too late I started with
> DJGPP which ported gcc to Windows. There were two threads growing out of
> that. Corrina Vinschen and others took the Cygwin branch which more to
> create a Unix environment of Windows while Colin Peters, and later Mumit
> Khan took the mingw32 branch to use gcc to build native Windows programs
> that became MSYS2 over the years. It was an interesting era.
>
> At work we used the MKS NutCracker tools to port what were originally AIX
> programs to Windows. While Cygwin would have probably worked it would have
> gotten into the whole GPL limitations on commercial software morass.
>

That's gnuwin32.

https://gnuwin32.sourceforge.net/packages.html

Example of a package. Has multiple download components, and
you have to do something about %PATH% if you expect full integration.
I run these in portable mode (English translation "I don't follow
the instructions") .

https://gnuwin32.sourceforge.net/packages/coreutils.htm

Requires a bit of skill in terms of getting them set up to run.

Note that on Windows 11, the nice dyndll response of the past is missing.
On previous Windows, running an EXE that was missing a dynDLL, the loader
would tell you the DLL that was missing. You'd put it next to the EXE,
and it would tell you the next DLL that was missing. In this way,
you could assemble all the dependencies. Windows 11 seems to be all or nothing.
But it is easy enough to test, by moving a DLL and see what happens.

Static compiling seems to be better for consumers, but not many do it.
And look at SNAPS, if you want to see the ultimate level of bloating
-- as the size of the package increases (gnome desktop SNAP), the
wasted download bandwidth increases enormously.

I still use at least one gnuwin32 regularly. That's gawk.exe .
I have other things I can run, like bash shell. The difference
between gnuwin32 gawk.exe and bash shell gawk, is the line endings.
Gnuwin32 uses native Windows line termination, while bash shell
uses Linux line termination, and a tiny mod must be made to
your script, if moving it.

Cygwin is yet another environment, with a decidedly mixed bag of
namespaces. But it's nothing to freak out about. That
gives you a Windows.exe and you refer to /dev/sda when
picking the first disk drive :-) It depends on the person
doing the port, as to what they support or what throws
a warning so you know what just happened.

Paul

On Wed, 1/1/2025 4:54 PM, Lawrence D'Oliveiro wrote:
> On 1 Jan 2025 20:19:01 GMT, Frank Slootweg wrote:
>
>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>
>>> And Microsoft is the only one shipping a beta-quality OS and expecting
>>> its users to rely on that for mission-critical production work.
>>
>> Any such 'users' are total idiots.
>
> You said it, I didn’t ...
>

There are multiple streams. We don't have experience
with all of them, so cannot comment on the practical
differences. I bet the US military version, doesn't
have the "location tracker" turned on :-) And at least
one other SKU, doesn't have Edge. There are differences
in the lineup, and a different set of policies in
GPEDIT.

Paul

On Wed, 1/1/2025 6:17 PM, Andrzej Matuch wrote:
> On 2025-01-01 14:34, Joel wrote:
>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>>
>>>> Beta software is beta software. An OS is an OS.
>>>
>>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>>> users to rely on that for mission-critical production work.
>>
>>
>> Yeah, weird, huh?  It's like they can't make it right from the
>> beginning, they need people to put up with bugs to test it in the
>> wild, how is their internal testing so poor?  Being on the cutting
>> edge with M$ is just spinning one's wheels.  Linux is the refuge.
>
> If Windows 11 didn't routinely become unbootable from an update that
> Microsoft didn't test under certain conditions, I'd disagree with Joel here.
> However, it seems that it happens with every one of their cumulative updates.

What does this "unbootable" mean exactly ?

Do you have just Windows on a disk by itself, and after
every Cumulative, the OS does not boot ? If it does not
boot, does it try three times until concluding the
repair procedures did not work ? And if the OS did not boot,
I presume at some point it did not boot and you were stuck.

Did you have a multiboot, GRUB was in control of the menu,
GRUB would no longer start (but the Windows Boot Manager entry
in the popup boot did work?). Generally, the only time Windows
breaks GRUB, is when adding a C: to the disk via Clean Install.

*******

Did you "Clean Install" to resume operation ? Every time ?
If multi-booting, this would require the (linux) Boot Repair CD, for
easiest GRUB repair. I've used that a few times, and for simple
setups, it can work.

The thing I have trouble with, is seeming damage to the Microsoft
folder in UEFI partition. I don't know of a simple way to 're-pave"
the files in there, like while running from a DVD. It may be
possible to restore /EFI/Microsoft from a backup (Macrium could
put the whole partition back), but for people without a backup,
I don't know of a solution that is direct and to the point.

Paul

On Wed, 1/1/2025 7:28 PM, Andrzej Matuch wrote:
> On 2025-01-01 18:38, rbowman wrote:
>> On Wed, 01 Jan 2025 14:34:06 -0500, Joel wrote:
>>
>>> Yeah, weird, huh?  It's like they can't make it right from the
>>> beginning, they need people to put up with bugs to test it in the wild,
>>> how is their internal testing so poor?  Being on the cutting edge with
>>> M$ is just spinning one's wheels.  Linux is the refuge.
>>
>> I really hate defending Microsoft but when they do a release how are they
>> supposed to test every possible combination of hardware and software?
>
> No, but they should give people the option to take a wait and see approach rather than force updates on everyone who didn't pay for a Pro license. Some people want every update the moment it's available; others like my wife want to watch the other computer crash and let the corporation fix the issue before she installs them.
>
>> Linux has gotten better but it was problematic on laptops and there were/
>> are lists of Linux friendly laptops. The Windows world assumes it's going
>> to work.
>>
>> Apple has a much easier task since they know exactly what the software is
>> going to run on.
>
> I have yet to see an Apple update screw anything up but it is definitely easier for them since there aren't as many MacOS configurations to consider.
>

There is a button in Windows Update, to delay the installation of patches.

"Pause Updates" Pull-down menu == "Up to 5 weeks (35 days)"

This should be sufficient, you can resume updates when your desktop is
free and a block of time is available for the update to happen.

The purpose of this, is to control the "arbitrariness" of updating.
It does not stop updating altogether. People here have also achieved
that (no updates) by breaking stuff :-) The "Update Orchestrator"
exists, to try to keep updates flowing, by noticing damage to
the update system.

Windows used to have a control variable between 0..4, and one of the
values made it so updates only showed up if you clicked the button
to check for updates. That made the control of it, perfectly manual,
and if you wanted to stop updating for three years, you could. The
design intent of the current scheme, is to not do it like that,
and (long term average) always be receiving updates.

Paul

Paul <nospam@needed.invalid> wrote:
>On Wed, 1/1/2025 6:17 PM, Andrzej Matuch wrote:
>>
>> If Windows 11 didn't routinely become unbootable from an update that
>> Microsoft didn't test under certain conditions, I'd disagree with Joel here.
>> However, it seems that it happens with every one of their cumulative updates.
>
>What does this "unbootable" mean exactly ?
>
>Do you have just Windows on a disk by itself, and after
>every Cumulative, the OS does not boot ? If it does not
>boot, does it try three times until concluding the
>repair procedures did not work ? And if the OS did not boot,
>I presume at some point it did not boot and you were stuck.
>
>Did you have a multiboot, GRUB was in control of the menu,
>GRUB would no longer start (but the Windows Boot Manager entry
>in the popup boot did work?). Generally, the only time Windows
>breaks GRUB, is when adding a C: to the disk via Clean Install.
>
>*******
>
>Did you "Clean Install" to resume operation ? Every time ?
>If multi-booting, this would require the (linux) Boot Repair CD, for
>easiest GRUB repair. I've used that a few times, and for simple
>setups, it can work.
>
>The thing I have trouble with, is seeming damage to the Microsoft
>folder in UEFI partition. I don't know of a simple way to 're-pave"
>the files in there, like while running from a DVD. It may be
>possible to restore /EFI/Microsoft from a backup (Macrium could
>put the whole partition back), but for people without a backup,
>I don't know of a solution that is direct and to the point.

There have been a couple too many media reports, about Winblows
fucking up to that degree. And several more about severe bugs on
certain OS install scenarios, yada yada. (And we're pretending not to
remember the various security and Internet holes there were, beginning
with NT 4, including 2000 and XP. Vista of course was probably pretty
great in that regard, though, oddly enough, as its overall code was a
mess.)

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

Paul <nospam@needed.invalid> wrote:

>There is a button in Windows Update, to delay the installation of patches.
>
> "Pause Updates" Pull-down menu == "Up to 5 weeks (35 days)"
>
>This should be sufficient, you can resume updates when your desktop is
>free and a block of time is available for the update to happen.
>
>The purpose of this, is to control the "arbitrariness" of updating.
>It does not stop updating altogether. People here have also achieved
>that (no updates) by breaking stuff :-) The "Update Orchestrator"
>exists, to try to keep updates flowing, by noticing damage to
>the update system.
>
>Windows used to have a control variable between 0..4, and one of the
>values made it so updates only showed up if you clicked the button
>to check for updates. That made the control of it, perfectly manual,
>and if you wanted to stop updating for three years, you could. The
>design intent of the current scheme, is to not do it like that,
>and (long term average) always be receiving updates.

If Windows Update is so dependent on this delay feature, that tells me
M$ is being ridiculous. Either they expect me to still be running
Win10, or to have clung to 23H2 at least, I guess, I mean WTF is going
on with them? Why would they imply that 24H2 was something "seekers"
would be cool with, having *severe* bugs?

In any event, it would seem evident people running 23H2 should wait
for Windows Update to offer a version upgrade, at least - if not
engage your delay "feature".

--
Joel W. Crump

Yeah, (supported versions of) Microsoft Windows is a piece of shit*.

* Unless you're running very high-end or fresh hardware..

On Wed, 1 Jan 2025 22:42:23 -0500, Paul wrote:

> I bet the US military version, doesn't have the "location tracker"
> turned on :-)

But how would they be sure? Looking at source code provides no guarantees
that those binaries were actually built from that code. They would have to
build it themselves to be sure. You can only do that with Free Software.

On Wed, 1 Jan 2025 23:59:53 -0500, Paul wrote:

> People here have also achieved that (no updates) by breaking stuff :-)

Microsoft itself has now come up with an update that does this, too--kills
the ability to receive further updates. As I mentioned in the posting that
started this thread.

On Tue, 31 Dec 2024 19:08:22 -0000 (UTC), Lawrence D'Oliveiro wrote:

> Newsgroups: alt.comp.os.windows-11,comp.os.linux.advocacy

OS WAR! <G>

--
s|b

On 2025-01-01 20:10, Joel wrote:
> Andrzej Matuch <andrzej@matu.ch> wrote:
>> On 2025-01-01 18:30, Joel wrote:
>>> Andrzej Matuch <andrzej@matu.ch> wrote:
>>>> On 2025-01-01 14:34, Joel wrote:
>>>>> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>>>>>> On Wed, 01 Jan 2025 14:45:44 GMT, Gregg Fowler wrote:
>>>>>>
>>>>>>> Beta software is beta software. An OS is an OS.
>>>>>>
>>>>>> And Microsoft is the only one shipping a beta-quality OS and expecting its
>>>>>> users to rely on that for mission-critical production work.
>>>>>
>>>>> Yeah, weird, huh? It's like they can't make it right from the
>>>>> beginning, they need people to put up with bugs to test it in the
>>>>> wild, how is their internal testing so poor? Being on the cutting
>>>>> edge with M$ is just spinning one's wheels. Linux is the refuge.
>>>>
>>>> If Windows 11 didn't routinely become unbootable from an update that
>>>> Microsoft didn't test under certain conditions, I'd disagree with Joel
>>>> here. However, it seems that it happens with every one of their
>>>> cumulative updates.
>>>
>>> Seemingly, they aren't really testing the code, much less testing
>>> deployment, because the kinds of bugs just don't reflect an odd
>>> incompatibility but rather that specific PC systems are treated in an
>>> erroneous manner, M$ can't manage its bloated OS, I guess. They do a
>>> passable job on the whole, though, but certain customers are plagued
>>> by the bugs.
>>
>> The bug that finally caused me to stop bothering with them isn't
>> Microsoft's fault, but I do fault them for continuing to demand TPM when
>> the requirement was causing stuttering with anyone who owned an AMD
>> processor. Like I said, if you were aware of what was causing the
>> problem and your motherboard allowed you to install a hardware TPM, you
>> were clear as long as you didn't mind paying to buy one, assuming that a
>> BIOS update wasn't available. However, most people could _at best_
>> disable it and lose the ability to install 11. If they knew about Rufus,
>> they could try to circumvent the problem and hope that Microsoft didn't
>> screw them over in an update. For laptop owners though, it's an absolute
>> mess that the manufacturers have no interest in fixing. There is no
>> excuse for any manufacturer to be aware of such an issue for years and
>> not do a thing about it. Additionally, there is no excuse for Microsoft
>> to acknowledge this reality and not at least allow AMD users to use 11
>> without TPM if not disable the hwrng causing the issue. Anyways, I'm
>> tired of being treated like I don't matter because they already got my
>> money.
>
>
> I think Microsoft genuinely feels that their barely supporting
> hardware, in any real sense, is the natural way, that if people are
> too stupid or stubborn to try Linux on a somewhat older system, they
> get what they deserve, bloat that overloads their gear. It's really
> amazing, how attached people are to Winblows software, and how
> intimidated they are by anything Linux (I mean they can try Mint,
> FFS), M$ and the OEMs just suck their money year after year.

As simple as the process of creating a USB thumdrive loaded with Linux
and booting from it is, this is considered too complicated by a large
amount of computer users. A lot of them already find it difficult to
uninstall software that they clearly don't need or to run a command like
sfc periodically to check if their system files are corrupt. They need
someone to hold their hand to install Linux and even then, they'll need
a hand just to figure out how to install and remove software. It seems
ridiculous for me to say this but it's not a joke; people really are
this stupid.

I've grown up surrounded by these people. I spent a lot of my life
thinking that everyone around me was at least as smart as I was.
Instead, I'm realizing that most people are either book smart or street
smart, and very few have the curiosity required to learn how things work
on their own.

>> At least the manufacturer replaced my motherboard when I intentionally
>> fucked it up on the day before my warranty ended. It was becoming clear
>> that the fingerprint reader wasn't working because the technician
>> shorted the motherboard during the first repair when they replaced my
>> battery, so I wanted a replacement and I wouldn't get it unless I killed
>> it.
>
> You mean that you made an already malfunctioning board completely
> dead, because you didn't want the fingerprint reader alone repaired,
> but a replacement of the board? I guess I can't fault you, for that,
> to me the idea of a warranty claim is the ultimate in drudgery, which
> is why I purchased the wireless charger for my phone when its USB-C
> stopped working under warranty, no fucking way am I playing that game
> where they issue me a replacement and refurbish and profit from my
> original one, fuck them, I'd rather give Amazon the few bucks for an
> alternative charging mechanism.

When they first repaired the laptop because the keyboard was faulty, I
asked them to replace the battery while they were at it because the wear
was at 17% or so. I was told that replacing it myself was risky because
one wrong move results in a spark that might kill your motherboard, so I
figured a professional would do a cleaner job. When I got the unit back,
the keyboard worked fine but suddenly the fingerprint reader was broken
so I asked them to look into it. They did nothing, as far as I know, but
sent me the unit back with some laptop sleeve and an apology. From that
point on, I tried literally everything to get that fingerprint reader to
work right but nothing resulted in it working consistently. With a day
left to my warranty, I figured they had gotten a spark when they changed
the battery, so I killed the motherboard (without it receiving any kind
of obvious damage) to force them to remedy the situation they had
ignored the previous time around. I was right: the motherboard was
indeed the issue and they had most likely gotten a spark because it
works perfectly now (in Windows, since it is not supported under Linux).

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

On 2025-01-01 23:59, Paul wrote:
> On Wed, 1/1/2025 7:28 PM, Andrzej Matuch wrote:
>> On 2025-01-01 18:38, rbowman wrote:
>>> On Wed, 01 Jan 2025 14:34:06 -0500, Joel wrote:
>>>
>>>> Yeah, weird, huh?  It's like they can't make it right from the
>>>> beginning, they need people to put up with bugs to test it in the wild,
>>>> how is their internal testing so poor?  Being on the cutting edge with
>>>> M$ is just spinning one's wheels.  Linux is the refuge.
>>>
>>> I really hate defending Microsoft but when they do a release how are they
>>> supposed to test every possible combination of hardware and software?
>>
>> No, but they should give people the option to take a wait and see approach rather than force updates on everyone who didn't pay for a Pro license. Some people want every update the moment it's available; others like my wife want to watch the other computer crash and let the corporation fix the issue before she installs them.
>>
>>> Linux has gotten better but it was problematic on laptops and there were/
>>> are lists of Linux friendly laptops. The Windows world assumes it's going
>>> to work.
>>>
>>> Apple has a much easier task since they know exactly what the software is
>>> going to run on.
>>
>> I have yet to see an Apple update screw anything up but it is definitely easier for them since there aren't as many MacOS configurations to consider.
>>
>
> There is a button in Windows Update, to delay the installation of patches.
>
> "Pause Updates" Pull-down menu == "Up to 5 weeks (35 days)"
>
> This should be sufficient, you can resume updates when your desktop is
> free and a block of time is available for the update to happen.

Is that actually available to Home users too? I know that people using
Pro can do so but I'm not so sure with the default version.

> The purpose of this, is to control the "arbitrariness" of updating.
> It does not stop updating altogether. People here have also achieved
> that (no updates) by breaking stuff :-) The "Update Orchestrator"
> exists, to try to keep updates flowing, by noticing damage to
> the update system.
>
> Windows used to have a control variable between 0..4, and one of the
> values made it so updates only showed up if you clicked the button
> to check for updates. That made the control of it, perfectly manual,
> and if you wanted to stop updating for three years, you could. The
> design intent of the current scheme, is to not do it like that,
> and (long term average) always be receiving updates.

I used to believe that it was imperative for people to get updates as
quickly as possible to prevent hackers from exploiting known issues.
However, it seems that there is no winning solution here. Either you
apply updates immediately and risk having your system go down because
Microsoft didn't test things properly or you risk having the system
taken over by bad agents. With Linux, you'll get a quick patch but the
likelihood that the system will go down as a a result of it is minimal
compared to Windows and the second situation is not likely to happen
even if you don't apply it. At some point, people need to consider these
two scenarios and ask themselves whether Windows truly is better for
them than Linux is.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

On 2025-01-02 00:15, Joel wrote:
> Paul <nospam@needed.invalid> wrote:
>> On Wed, 1/1/2025 6:17 PM, Andrzej Matuch wrote:
>>>
>>> If Windows 11 didn't routinely become unbootable from an update that
>>> Microsoft didn't test under certain conditions, I'd disagree with Joel here.
>>> However, it seems that it happens with every one of their cumulative updates.
>>
>> What does this "unbootable" mean exactly ?
>>
>> Do you have just Windows on a disk by itself, and after
>> every Cumulative, the OS does not boot ? If it does not
>> boot, does it try three times until concluding the
>> repair procedures did not work ? And if the OS did not boot,
>> I presume at some point it did not boot and you were stuck.
>>
>> Did you have a multiboot, GRUB was in control of the menu,
>> GRUB would no longer start (but the Windows Boot Manager entry
>> in the popup boot did work?). Generally, the only time Windows
>> breaks GRUB, is when adding a C: to the disk via Clean Install.
>>
>> *******
>>
>> Did you "Clean Install" to resume operation ? Every time ?
>> If multi-booting, this would require the (linux) Boot Repair CD, for
>> easiest GRUB repair. I've used that a few times, and for simple
>> setups, it can work.
>>
>> The thing I have trouble with, is seeming damage to the Microsoft
>> folder in UEFI partition. I don't know of a simple way to 're-pave"
>> the files in there, like while running from a DVD. It may be
>> possible to restore /EFI/Microsoft from a backup (Macrium could
>> put the whole partition back), but for people without a backup,
>> I don't know of a solution that is direct and to the point.
>
>
> There have been a couple too many media reports, about Winblows
> fucking up to that degree. And several more about severe bugs on
> certain OS install scenarios, yada yada. (And we're pretending not to
> remember the various security and Internet holes there were, beginning
> with NT 4, including 2000 and XP. Vista of course was probably pretty
> great in that regard, though, oddly enough, as its overall code was a
> mess.)

Its code was a mess because it was a complete rewrite of what already
existed. The reason Vista was bad and 7 was good was because the same
code that had gone into the former was optimized by the time of the
former's release. Windows is still using that code today and I would
imagine that there is no longer a need for a rewrite.

--
Andrzej (Andre) Matuch
Telegram: @AndrzejMatuch
Zephyrus G14 GA401QM on Fedora 41
KDE supporting member

"s|b" <me@privacy.invalid> wrote:
>On Tue, 31 Dec 2024 19:08:22 -0000 (UTC), Lawrence D'Oliveiro wrote:
>
>> Newsgroups: alt.comp.os.windows-11,comp.os.linux.advocacy
>
>OS WAR! <G>

I frequently talk to Copilot in a Linux Web app about how sleek Linux
is compared to Winblows, and it agrees with me. M$ is laughing at the
people using their products on aging hardware.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.