From Newsgroup: comp.misc

The healthcare sector is hitting some hiccups trying to upgrade its
equipment to Windows 11 <https://www.computerworld.com/article/4082710/industrial-and-medical-devices-struggle-to-upgrade-to-windows-11.html>.

But then, this has happened with every major upgrade of Windows before
this; why are the same mistakes still being made?

Surely when you buy an expensive piece of equipment with an expected
useful lifetime measurable in decades and a computer at its heart,
common sense dictates that you will nail down a support contract that
will cover hardware *and* software maintenance over that lifetime,
will you not?

We have had about half a century of experience of gear like this,
since long before Microsoft came along. I wonder what lessons have
been learned ...
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence D’Oliveiro <[email protected]d> wrote at 04:10 this Saturday (GMT):

The healthcare sector is hitting some hiccups trying to upgrade its
equipment to Windows 11
<https://www.computerworld.com/article/4082710/industrial-and-medical-devices-struggle-to-upgrade-to-windows-11.html>.

But then, this has happened with every major upgrade of Windows before
this; why are the same mistakes still being made?

Surely when you buy an expensive piece of equipment with an expected
useful lifetime measurable in decades and a computer at its heart,
common sense dictates that you will nail down a support contract that
will cover hardware *and* software maintenance over that lifetime,
will you not?

We have had about half a century of experience of gear like this,
since long before Microsoft came along. I wonder what lessons have
been learned ...

and this is why the windows 11 force upgrade is so annoying, there have
been so many issues and the upgrade treadmill in general is infuriating
having stuff break for no fault of your own
--
user <candycane> is generated from /dev/urandom
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

candycanearter07 <[email protected]> wrote
in news:[email protected]d:

Lawrence D’Oliveiro <[email protected]d> wrote at 04:10 this Saturday
(GMT):

The healthcare sector is hitting some hiccups trying to upgrade its
equipment to Windows 11 >><https://www.computerworld.com/article/4082710/industrial-and-medical-dev >>ices-struggle-to-upgrade-to-windows-11.html>.

But then, this has happened with every major upgrade of Windows before
this; why are the same mistakes still being made?

Surely when you buy an expensive piece of equipment with an expected
useful lifetime measurable in decades and a computer at its heart,
common sense dictates that you will nail down a support contract that
will cover hardware *and* software maintenance over that lifetime,
will you not?

We have had about half a century of experience of gear like this,
since long before Microsoft came along. I wonder what lessons have
been learned ...

and this is why the windows 11 force upgrade is so annoying, there have
been so many issues and the upgrade treadmill in general is infuriating having stuff break for no fault of your own

If the tech builders weren't forced to use upgraded technology the support could stay stable for many generations of products. Often it was suggested/demanded by the developers because some tool in the newer cycle
eased the job of the developer.

If the developers were competant and willing to work with the old
development tools upgrades are usually easy. The downside of course is
some of the desires of the product's end users that demand features only
in newer systems they are familiar with.

This my experience from 44 years of system development and dealing with
younger developers. Managers and product leaders can also fall into the
trap of newer is better. AI stuff just makes this issue worse. Even the
new developers barely know what is going on and they will likely be laid
off before real user feedback comes in.

-- David LaRue
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often it
was suggested/demanded by the developers because some tool in the
newer cycle eased the job of the developer.

Even if the developers of the product didn’t want to use Windows 11,
what are they supposed to do once Microsoft says it’s not going to
support Windows 10 any more?
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote in news:10ec126$3dipv$[email protected]:

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often it
was suggested/demanded by the developers because some tool in the
newer cycle eased the job of the developer.

Even if the developers of the product didn’t want to use Windows 11,
what are they supposed to do once Microsoft says it’s not going to
support Windows 10 any more?

You first need to know what your products depend on in say Win 10. Then
use only those Win 10 features or older that are supported going into Win
11. Yes this becomes a pain after many generations but only if you depend
on the features of the OS. If you look at Win 10 Development there are features and programming methods still there from 3.1 and NT. You
obviously don't need to go back that far. Then structure your product
using only the chosen features for a while.

You need to decide what is more important going forward. Windows 11 and
back really didn't add much. Yes your tool chains evolved. Can you
develop for the current Win 10 product and keep using it on 11/12/beyond?
If you can the existing product could be supported for a long time using
just Microsoft's and your existing tool chain and use case environment.

A little hint for you. What do you really need that is in 11? Could you
make something like that with your 10 tools?

I am using Win 7 Pro here at home and refuse to upgrade past that for my
own reasons. Microsoft stopped supporting 7 years ago... yet I still get daily updates from Microsoft for Defender. All my development tool chains still work for 7 and below. When I really need to upgrade to a new
computer I could load it onto a VM and use the VM system on 11/12/Linux or whatever I want.

All my employers since 2000 or so saw this as the only long term approach
for them. They are all still doing it!

There are some things to plan for, like keeping the older VMs and support hardware working.

Do you need to be tethered to Microsoft's price model forever?

Don't even get me started on Win 10's spyware. Win 11 is 98% hackable all
the way through right now by script kiddies. Nothing is secure or reliable unless you plan for it and keep trying to fix problems before they happen
to real users. Look at all the systems around the world that failed due to the recent cloud outage? Were you affected? I didn't notice and had to
look for sites/products that were reported down just to see what was going
on.

If you were affected or your product was, has that problem already been
fixed for the next time it will happen?

Glad to meet you Lawrence!

Keep thinking of ways to improve your tools/products. Keep your employers
out of the news.

--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

On Tue, 4 Nov 2025 06:18:02 -0000 (UTC), David LaRue wrote:

On Tue, 4 Nov 2025 04:56:39 -0000 (UTC), Lawrence D’Oliveiro wrote:

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often it
was suggested/demanded by the developers because some tool in the
newer cycle eased the job of the developer.

Even if the developers of the product didn’t want to use Windows 11,
what are they supposed to do once Microsoft says it’s not going to
support Windows 10 any more?

You first need to know what your products depend on in say Win 10. Then
use only those Win 10 features or older that are supported going into
Win 11.

Doesn’t matter what features you do or don’t use, you still have to test everything just as thoroughly. Which, in heavily-regulated, safety-
critical fields like industrial control and healthcare, can be a pretty expensive business.

Which brings us right back to where we came in: the companies offering
these upgrades want to charge hefty fees for them.

The users could have seen this coming, decades in advance, given this is
not the first time such a situation has occurred with some version of Microsoft Windows. And plan their long-term support contracts accordingly. That’s the way to cushion the blow.
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote in news:10ecgp5$3hi5o$[email protected]:

On Tue, 4 Nov 2025 06:18:02 -0000 (UTC), David LaRue wrote:

On Tue, 4 Nov 2025 04:56:39 -0000 (UTC), Lawrence D’Oliveiro wrote:

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often it
was suggested/demanded by the developers because some tool in the
newer cycle eased the job of the developer.

Even if the developers of the product didn’t want to use Windows 11,
what are they supposed to do once Microsoft says it’s not going to
support Windows 10 any more?

You first need to know what your products depend on in say Win 10.
Then use only those Win 10 features or older that are supported going
into Win 11.

Doesn’t matter what features you do or don’t use, you still have to
test everything just as thoroughly. Which, in heavily-regulated, safety- critical fields like industrial control and healthcare, can be a pretty expensive business.

Which brings us right back to where we came in: the companies offering
these upgrades want to charge hefty fees for them.

The users could have seen this coming, decades in advance, given this is
not the first time such a situation has occurred with some version of Microsoft Windows. And plan their long-term support contracts
accordingly. That’s the way to cushion the blow.

Testing should always include everything possible. Including some portion
of the developers spending time activiely trying to hack/disable/interfere
with the products to find issues that can be addressed/mitigrated by development. With small teams you sometimes need to focus on this for a
while.

Most of my career has been with companies whose products are mission
critical and must be reliable for 20 years or more at a minimum. Such
systems should never fail. At worst they should fail and recover so
silently and quickly that users can't even observe issues during the
failures. It takes a huge amount of testing effort too. It is
challenging work and quite enjoyable if you can work under such pressure.

Yes such products can be expensive to maintain when you are dependant on limited resources or timespans. Minimize costs where you can. Hardware
and software improvements should only be used if they really help the
product. Such products require 95+% testing to actual development time.
The public or end user might see different products but the underlying
support hardware and software is usually reused in all the products or
even across many industries.

Such reliability costs the developers who pass that cost on to the end
user. If supporting the next generation of anything worries you think
about ways to minimize the impact. Sometimes we do have to pay Microsoft
for longer lifespans when they are embedded in our products. There are
always options. Think of the product as multiple systems and then remake
your product to fit it. Perhaps the engine inside your product can be
reused or left unchanged. Then it is just the user side of the system
that needs to upgrade to the latest (or perhaps multiple levels of) the
user interface (in Windows).

Think outside the current box you or your company are in. How would you improve things? Can a simple example be made to try out your ideas?
Pitch the good stuff to appropriate people in your company.

I can't really count the number of times I've done this. Some companies
are blessed with wonderful managers. Others you might need to go higher.
Just be ready for rejection or (sometimes worse) acceptance that can
really force you to up your game.

Have fun!
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

On Tue, 4 Nov 2025 12:27:27 -0000 (UTC), David LaRue wrote:

Most of my career has been with companies whose products are mission
critical and must be reliable for 20 years or more at a minimum. Such systems should never fail.

And surely, they should not continue to depend on critical components
which have become obsolete or unsupported, should they?
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote:

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often it
was suggested/demanded by the developers because some tool in the
newer cycle eased the job of the developer.

Even if the developers of the product didn't want to use Windows 11,
what are they supposed to do once Microsoft says its not going to
support Windows 10 any more?

Do you want an appliance or an IT device? They need to be designed very differently and people who try to use a general purpose desktop operating system to run an embedded system are going to be disappointed sooner or
later.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote:

On Tue, 4 Nov 2025 12:27:27 -0000 (UTC), David LaRue wrote:

Most of my career has been with companies whose products are mission
critical and must be reliable for 20 years or more at a minimum. Such
systems should never fail.

And surely, they should not continue to depend on critical components
which have become obsolete or unsupported, should they?

We have plenty of systems around here that are based on RT-11. Just got
rid of the last PDP-8e in 2022 and replaced it with an Allen-Bradley PLC.
I expect the PLC to last as long as the PDP-8 did because that's what
it's for.

The embedded control world is not the IT world. Embedded systems often
outlast their support and their manufacturer. So you plan for it.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

On Tue, 4 Nov 2025 19:02:49 -0500 (EST), Scott Dorsey wrote:

On Tue, 4 Nov 2025 04:56:39 -0000 (UTC), Lawrence D’Oliveiro wrote:

On Tue, 4 Nov 2025 04:41:48 -0000 (UTC), David LaRue wrote:

If the tech builders weren't forced to use upgraded technology the
support could stay stable for many generations of products. Often
it was suggested/demanded by the developers because some tool in
the newer cycle eased the job of the developer.

Even if the developers of the product didn't want to use Windows
11, what are they supposed to do once Microsoft says its not going
to support Windows 10 any more?

Do you want an appliance or an IT device?

This is about expensive pieces of gear, often safety-critical, but
certainly business-mission-critical, and often with an expected
operational lifetime measurable not just in years, but in decades.

Surely, when planning such purchases, common sense dictates that you
will nail down a support contract that will cover hardware *and*
software maintenance over that lifetime, will you not?
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

On Tue, 4 Nov 2025 19:05:12 -0500 (EST), Scott Dorsey wrote:

On Tue, 4 Nov 2025 20:34:15 -0000 (UTC), Lawrence D’Oliveiro wrote:

On Tue, 4 Nov 2025 12:27:27 -0000 (UTC), David LaRue wrote:

Most of my career has been with companies whose products are
mission critical and must be reliable for 20 years or more at a
minimum. Such systems should never fail.

And surely, they should not continue to depend on critical
components which have become obsolete or unsupported, should they?

We have plenty of systems around here that are based on RT-11.

So that’s a “no”? You *should* continue to depend on critical
components which have become obsolete or unsupported?

The embedded control world is not the IT world. Embedded systems
often outlast their support and their manufacturer. So you plan for
it.

How is the customer supposed to do that? Are you able to offer them
solid support contracts for the duration of the expected life of the
product in question? What happens if that duration exceeds the
supported lifetime of some upstream proprietary product that *you*
depend on? Do you take on the necessary support burden on behalf of
your customer? Or do you just tell them that’s not within the scope of
the support contract?
--- Synchronet 3.21a-Linux NewsLink 1.2

From Newsgroup: comp.misc

Lawrence =?iso-8859-13?q?D=FFOliveiro?= <[email protected]d> wrote:

On Tue, 4 Nov 2025 19:05:12 -0500 (EST), Scott Dorsey wrote:

On Tue, 4 Nov 2025 20:34:15 -0000 (UTC), Lawrence D’Oliveiro wrote:

On Tue, 4 Nov 2025 12:27:27 -0000 (UTC), David LaRue wrote:

Most of my career has been with companies whose products are
mission critical and must be reliable for 20 years or more at a
minimum. Such systems should never fail.

And surely, they should not continue to depend on critical
components which have become obsolete or unsupported, should they?

We have plenty of systems around here that are based on RT-11.

So that’s a “no”? You *should* continue to depend on critical >components which have become obsolete or unsupported?

The problem is that I can get better support for RT-11 than for Windows 11.

The embedded control world is not the IT world. Embedded systems
often outlast their support and their manufacturer. So you plan for
it.

How is the customer supposed to do that? Are you able to offer them
solid support contracts for the duration of the expected life of the
product in question? What happens if that duration exceeds the
supported lifetime of some upstream proprietary product that *you*
depend on? Do you take on the necessary support burden on behalf of
your customer? Or do you just tell them that’s not within the scope of
the support contract?

I'm used to getting a ten-year guarantee for parts availability, and
full schematics with test equipment. Source code would be nice but
usually isn't available for production systems anymore. Although since
it was often available for older systems when they were shipped, it is
easier for me to fix the 1980s GenRad 2615 analyzer than the latest Keysight. --scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
--- Synchronet 3.21a-Linux NewsLink 1.2