On 6/17/2025 6:49 PM, Sergey Kubushyn wrote:

Especially when you are not even able to BUY that software, only rent. Not just it gets extremely expensive after couple of years like it is now with e.g. Altium Designer -- its cost PER YEAR is now the same as I paid for my PERPETUAL license when those were still available and "subscription" was (still "is" for me as mine only expires in 2027) something like 1/4..1/3 of that and not mandatory. The biggest problem is what are you going to do when the company goes bust (it is not "if", it is _ALWAYS_ "when" -- the bigger they are the harder they fall) or just decides they don't want to support that outdated weirdo anymore when they have a newer one or suddenly want 10x last year price...

PC-based CAD/EDA tools have always been a gamble. The market is too small
and the customers don't want to pay for ongoing support/development (the
"I can do all that stuff myself" econo-mentality).

I dropped ~$15K on the DASH suite in 1986. Where is it today? (where was
it in 1990??) PCAD, Protel, OrCAD, Cadence, DIPtrace, etc.

[Sad as STRIDES was one of the slickest tools I've used!]

OTOH, you could have purchased a Mentor Graphics workstation and still
be running PADS. Is Computervision still in business? (does anyone
maintain their hardware?).

OTOH, I bought AutoCAD w/AME in the same timeframe (about $3K) and it's
been there for me in the 40 years since! I'd still be waiting for
FOSS tools to have the capabilities I *bought* those many decades ago!

gEDA, KiCAD, FreeCAD... what happens when the developers move on
to the next "greatest" thing? Are you willing to take on maintenance
of the codebase -- even if only for yourself?

Ask yourself what your time is worth and how much the tool is
saving (or costing!) you. Simple business decision.

[E.g., I no longer maintain color printers as the cost far exceeds
what I have to pay if I print on a professionally maintained machine
at the end of the block!]

Whatever tools (not just EDA/CAD) you use, be sure you can keep running them (or, resurrect them, as needed) for as long as you need to support <whatever> you've used them for (hardware designs tend to be very transitory; though
I've a client who has somehow kept one of my designs in production for ~30 years... I have absolutely no idea how he finds the parts to do so! :< )

[VMs are the obvious choice for tools that can run on generic hardware.
Things get a bit more complicated when you need special hardware to host
the tools <frown>]

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Wed, 18 Jun 2025 01:57:33 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:

On 6/17/2025 6:49 PM, Sergey Kubushyn wrote:

Especially when you are not even able to BUY that software, only rent. Not >> just it gets extremely expensive after couple of years like it is now with >> e.g. Altium Designer -- its cost PER YEAR is now the same as I paid for my >> PERPETUAL license when those were still available and "subscription" was
(still "is" for me as mine only expires in 2027) something like 1/4..1/3 of >> that and not mandatory. The biggest problem is what are you going to do when >> the company goes bust (it is not "if", it is _ALWAYS_ "when" -- the bigger >> they are the harder they fall) or just decides they don't want to support
that outdated weirdo anymore when they have a newer one or suddenly want 10x >> last year price...

PC-based CAD/EDA tools have always been a gamble. The market is too small >and the customers don't want to pay for ongoing support/development (the
"I can do all that stuff myself" econo-mentality).

I dropped ~$15K on the DASH suite in 1986. Where is it today? (where was
it in 1990??) PCAD, Protel, OrCAD, Cadence, DIPtrace, etc.

[Sad as STRIDES was one of the slickest tools I've used!]

OTOH, you could have purchased a Mentor Graphics workstation and still
be running PADS. Is Computervision still in business? (does anyone
maintain their hardware?).

We use PADS, Logic and PCB. It works fine. It's made us a couple of
hundred million dollars sales, so its cost is in the noise.

We've used it since the DOS days when there was basically nothing
else, and never bothered to get something else. We will if we have to
some day.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/18/2025 10:23 AM, Crash Gordon wrote:

On 6/17/2025 8:49 PM, Sergey Kubushyn wrote:

Especially when you are not even able to BUY that software, only rent. Not >> just it gets extremely expensive after couple of years...

Back in the early aughts I worked for a company that bought me a Solidworks license.  At the time, one seat allowed for three installations: One on the work desktop, one on a home desktop, and one on a laptop -- with the proviso that only one instance would be active at any time.  This was explicitly spelled out in the license, and afaik was not enforced by any kind of "phone home" tech.

I'm in a similar situation (each client had their own idea of the "right"
tools so I have a sh*tload of them).

The company I was working for went bust, and I still have and use my "home desktop" version of Solidworks.  I have to keep an XP machine around to run it
on, which is not difficult for me (you should see my garage! :-) but I believe
I can convert it to a VM without breaking the license check, and that would run
on modern hardware.

You may need to spoof a MAC address if nodelocked to that. It's possible to really fingerprint the hosting system and make portability difficult. But, they want you to USE their products, not throw your hands up in frustration because you added RAM to a box or upgraded the CPU, replaced a disk drive,
etc.

I'm sure SW has been extensively updated in the past couple of decades, but the
200x version I have does everything I need.

Exactly. Find something that does the job, that has a set of bugs that
you know and can work-around, and leave the feeping-creaturism to folks
with more time than brains! ("Oooooo! A new update! Let's see what
THIS one does...")

By far, the biggest bit of inertia comes from the libraries (electronic components, packages, mechanisms, etc.). These are often not portable
from one vendor/product to another.

OTOH, its likely that noone here uses a huge variety of "parts" on
a given design so rebuilding a library from scratch is more of a
nuisance than an impediment. You can incrementally rebuild a library
as you take on new designs.

[Big companies with "standards" are more likely to benefit from library portability; if they've many thousand parts built (to satisfy a variety
of users and projects), then recreating them each time a new tool comes
/en vogue/ can be a significant time $ink.]

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

On 6/17/2025 6:49 PM, Sergey Kubushyn wrote:

Especially when you are not even able to BUY that software, only rent. Not >> just it gets extremely expensive after couple of years like it is now with >> e.g. Altium Designer -- its cost PER YEAR is now the same as I paid for my >> PERPETUAL license when those were still available and "subscription" was
(still "is" for me as mine only expires in 2027) something like 1/4..1/3 of >> that and not mandatory. The biggest problem is what are you going to do when >> the company goes bust (it is not "if", it is _ALWAYS_ "when" -- the bigger >> they are the harder they fall) or just decides they don't want to support
that outdated weirdo anymore when they have a newer one or suddenly want 10x >> last year price...

PC-based CAD/EDA tools have always been a gamble. The market is too small and the customers don't want to pay for ongoing support/development (the
"I can do all that stuff myself" econo-mentality).

I dropped ~$15K on the DASH suite in 1986. Where is it today? (where was
it in 1990??) PCAD, Protel, OrCAD, Cadence, DIPtrace, etc.

[Sad as STRIDES was one of the slickest tools I've used!]

OTOH, you could have purchased a Mentor Graphics workstation and still
be running PADS. Is Computervision still in business? (does anyone
maintain their hardware?).

OTOH, I bought AutoCAD w/AME in the same timeframe (about $3K) and it's
been there for me in the 40 years since! I'd still be waiting for
FOSS tools to have the capabilities I *bought* those many decades ago!

gEDA, KiCAD, FreeCAD... what happens when the developers move on
to the next "greatest" thing? Are you willing to take on maintenance
of the codebase -- even if only for yourself?

The problem is not the codebase maintenance. You can happily live without forever as long as you still have hardware it runs on.

The FUNDAMENTAL difference is that unlike [almost] ALL modern commercial software the older software had PERPETUAL licenses and OSS doesn't have any
at all. Those will NOT stop working in a year because your license expired
and it is too expensive or impossible at all to renew it for another year.
If this happens you'll end up holding a bag -- all your work done with those tools is useless now and you can't do anything to something you still have
to support. Not just some ancient stuff but something one year old or even newer because your tools stopped working and there is no way to make them
work again.

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. * ******************************************************************

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/18/2025 2:22 PM, Sergey Kubushyn wrote:

The problem is not the codebase maintenance. You can happily live without forever as long as you still have hardware it runs on.

Then you just pick an OLDER "perpetual" version and live without the
"new features" that were added when the license became time-sensitive.

Ask me why I run W7 on my machines...

The FUNDAMENTAL difference is that unlike [almost] ALL modern commercial software the older software had PERPETUAL licenses and OSS doesn't have any at all. Those will NOT stop working in a year because your license expired and it is too expensive or impossible at all to renew it for another year.

But, they may contain flaws that make them unusable until someone
"volunteers" to fix them.

I own two copies of Brief. But, can't use either of them because
the code is "too fast" for modern hardware. So, the money spent
on them is no longer yielding benefits for me.

Ditto for the DASH suite. Wonderful if I have a need to return to
one of those old designs for some (paid!) maintenance. But, otherwise,
an unperforming investment.

If this happens you'll end up holding a bag -- all your work done with those tools is useless now and you can't do anything to something you still have

You still have the IP. You just don't have it in a compatible tool
that you WANT to pay for.

Are you planning on making any money "supporting" the design for
which you used these tools? If so, then the tool still has financial
value to you -- even if you gripe about not wanting to have to keep
"renting" it.

Would you prefer they charge you $100K for a perpetual license?
Are you sure you will have future need to justify that outlay?

I.e., there is a certain sense in renting software; pay for what
you *use*. But, that assumes you can rent it "a la carte".

My gripe is with tools that I may have to "rent" JUST IN CASE their
might be a future use. Let me pay for 30 days of use 5 years from today,
as the need arises. NOT for the next 5 years where it is sitting idle!

to support. Not just some ancient stuff but something one year old or even newer because your tools stopped working and there is no way to make them work again.

Run it in a VM. Take a snapshot of it on day one. When the license
expires, reset the clock and reload from the original image. Lather,
rinse, repeat.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Don Y <blockedofcourse@foo.invalid> wrote:

On 6/18/2025 2:22 PM, Sergey Kubushyn wrote:

The problem is not the codebase maintenance. You can happily live without
forever as long as you still have hardware it runs on.

Then you just pick an OLDER "perpetual" version and live without the
"new features" that were added when the license became time-sensitive.

Ask me why I run W7 on my machines...

OK, can you tell how can one pick an OLDER "perpetual" Altium Designer
version (actually license) if he didn't have it before? I do have a
perpetual license and grandfathered subscription valid until the end of
March 2027 so my last version will be whatever is out before that
subscription expired. However, if you didn't have a perpetual license before where are you going to get one?

And I'm not only running Windoze 7 for some older stuff, I even have XP
64-bit with full development environment for Windoze Embedded CE 6.1 (?).
That does NOT have a time-limited license so it will run forever in a VM
that I have for it. I even have NT somewhere :)

The FUNDAMENTAL difference is that unlike [almost] ALL modern commercial
software the older software had PERPETUAL licenses and OSS doesn't have any >> at all. Those will NOT stop working in a year because your license expired >> and it is too expensive or impossible at all to renew it for another year.

But, they may contain flaws that make them unusable until someone "volunteers" to fix them.

If you were using it before you can still use it forever. If those flaws was making them unusable you wouldn't have been able using it before. All your
old designs that were made with that tool clearly tell it was not unusable because of those flaws and you need that tool to either reuse whatever you
did (less likely) or to support it by fixing bugs and whatever else (more likely). You will be not able to do either when you tool ceased to work.

I own two copies of Brief. But, can't use either of them because
the code is "too fast" for modern hardware. So, the money spent
on them is no longer yielding benefits for me.

Ditto for the DASH suite. Wonderful if I have a need to return to
one of those old designs for some (paid!) maintenance. But, otherwise,
an unperforming investment.

If this happens you'll end up holding a bag -- all your work done with those >> tools is useless now and you can't do anything to something you still have

You still have the IP. You just don't have it in a compatible tool
that you WANT to pay for.

Are you planning on making any money "supporting" the design for
which you used these tools? If so, then the tool still has financial
value to you -- even if you gripe about not wanting to have to keep
"renting" it.

Would you prefer they charge you $100K for a perpetual license?
Are you sure you will have future need to justify that outlay?

I paid for my perpetual Altium Designer license the same price they charge
now for one year time-limited one. I was also been paying for "subscription" which is 1/4..1/3 of that price per year that gives me access to all new versions but I won't lose any functionality when it expires and I don't want
or can't renew it. It will just leave me with the latest version available before it expired. And that last version will be still working forever.

I.e., there is a certain sense in renting software; pay for what
you *use*. But, that assumes you can rent it "a la carte".

This is an option, yes. And good guys do offer that. E.g. Toyota allows you
to purchase a Techstream (absolute necessity if you want to do anything more complex than tire replacement on a Toyota/Lexus vehicle) licence in WEEK increments for a very reasonable price. You don't need it everyday unless
you are in auto repair business so that is reasonable. They don't insist on purchasing a full year "subscription" for a ridiculous price. However, there
is no guarantee they will behave like this forever so I also have an old
laptop with Techstream (the older one but my Lexus LS460L is not the newest
and fully supported by that version; however you still need their FULL
version with Internet connection to the mothership if you need to re-program some ECU). Not a COMPLETE solution but better than nothing and it allows to
do almost everything except a few things that require tethering to the mothership.

My gripe is with tools that I may have to "rent" JUST IN CASE their
might be a future use. Let me pay for 30 days of use 5 years from today,
as the need arises. NOT for the next 5 years where it is sitting idle!

The problem is you might be not able to get it in 5 years AT ALL, no matter what's the price you are ready to pay. And paying "rent" for all those 5
years does not guarantee anything -- when (not "if") the company that made
that tool goes bust you are fried. It is not even like buying vs renting
your house -- you can find a different one any time unlike those tools that
are all unique and once they gone there is no replacement.

to support. Not just some ancient stuff but something one year old or even >> newer because your tools stopped working and there is no way to make them
work again.

Run it in a VM. Take a snapshot of it on day one. When the license
expires, reset the clock and reload from the original image. Lather,
rinse, repeat.

That was only good in early naive days where everything was extremely simple and there was no Internet. Many tools these days won't even start if they
can't connect to the mother ship to check your license. Which is also a
great simplification, it is more complex than that and gets even more sophisticated by the day...

---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. * ******************************************************************

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/18/2025 4:47 PM, Sergey Kubushyn wrote:

Don Y <blockedofcourse@foo.invalid> wrote:

On 6/18/2025 2:22 PM, Sergey Kubushyn wrote:

The problem is not the codebase maintenance. You can happily live without >>> forever as long as you still have hardware it runs on.

Then you just pick an OLDER "perpetual" version and live without the
"new features" that were added when the license became time-sensitive.

Ask me why I run W7 on my machines...

OK, can you tell how can one pick an OLDER "perpetual" Altium Designer version (actually license) if he didn't have it before? I do have a

You can't. What you CAN do is not upgrade when you see the license
terms changing in a way that you don't like.

I stopped upgrading my Adobe tools when they went to the "CC" series.
I knowingly forfeit any improvements in the tools in order to avoid that "online" issue.

perpetual license and grandfathered subscription valid until the end of
March 2027 so my last version will be whatever is out before that subscription expired. However, if you didn't have a perpetual license before where are you going to get one?

And I'm not only running Windoze 7 for some older stuff, I even have XP 64-bit with full development environment for Windoze Embedded CE 6.1 (?). That does NOT have a time-limited license so it will run forever in a VM
that I have for it. I even have NT somewhere :)

I have VMs for tools running back more than 30 years. Previously, I had
set aside a physical computer to support development environments that
I had used for early projects. Originally, dumping the disk image onto
9T tape so I could reuse a single physical machine for multiple project environments. Later, swapping physical disks to provide the environment
that was in use.

[Ultimately, replacing physical machines with virtual machines to
facilitate discarding the old hardware -- except for those environments
that required specific hardware]

The FUNDAMENTAL difference is that unlike [almost] ALL modern commercial >>> software the older software had PERPETUAL licenses and OSS doesn't have any >>> at all. Those will NOT stop working in a year because your license expired >>> and it is too expensive or impossible at all to renew it for another year. >>

But, they may contain flaws that make them unusable until someone
"volunteers" to fix them.

If you were using it before you can still use it forever. If those flaws was making them unusable you wouldn't have been able using it before. All your

That's specious reasoning. You may never have tried to do a copper pour.
Or, an 8 layer pad stack. Or, a 500 pin component symbol. etc.

I stopped using MS's compilers when I tried to take a pointer to a member function and the compiler gagged. Previously, it had generated correct
code.

old designs that were made with that tool clearly tell it was not unusable because of those flaws and you need that tool to either reuse whatever you did (less likely) or to support it by fixing bugs and whatever else (more likely). You will be not able to do either when you tool ceased to work.

You don't own the software. ANY software (save any that you create
yourself). So, you are at the mercy of the actual owner of that software.
If you don't like his terms (given the benefits of the tool), then you
are free to find another.

I own two copies of Brief. But, can't use either of them because
the code is "too fast" for modern hardware. So, the money spent
on them is no longer yielding benefits for me.

Ditto for the DASH suite. Wonderful if I have a need to return to
one of those old designs for some (paid!) maintenance. But, otherwise,
an unperforming investment.

If this happens you'll end up holding a bag -- all your work done with those
tools is useless now and you can't do anything to something you still have >>

You still have the IP. You just don't have it in a compatible tool
that you WANT to pay for.

Are you planning on making any money "supporting" the design for
which you used these tools? If so, then the tool still has financial
value to you -- even if you gripe about not wanting to have to keep
"renting" it.

Would you prefer they charge you $100K for a perpetual license?
Are you sure you will have future need to justify that outlay?

I paid for my perpetual Altium Designer license the same price they charge now for one year time-limited one.

Then you should be grateful! Others no longer have that option.
SWMBO uses an (ancient -- 2K?) MSOffice suite to maintain her databases.
She made a deliberate decision NOT to upgrade to features (and
constraints!) that she wasn't comfortable with. Has she missed out on some features that might have benefitted her? I know it's always tedious
to install it on OSs that are considerably newer; eventually, she will be "stuck" with whichever the most recent OS simply because something will
be incompatible with OS version N+1.

I was also been paying for "subscription"
which is 1/4..1/3 of that price per year that gives me access to all new versions but I won't lose any functionality when it expires and I don't want or can't renew it. It will just leave me with the latest version available before it expired. And that last version will be still working forever.

I.e., there is a certain sense in renting software; pay for what
you *use*. But, that assumes you can rent it "a la carte".

This is an option, yes. And good guys do offer that. E.g. Toyota allows you to purchase a Techstream (absolute necessity if you want to do anything more complex than tire replacement on a Toyota/Lexus vehicle) licence in WEEK increments for a very reasonable price. You don't need it everyday unless
you are in auto repair business so that is reasonable. They don't insist on purchasing a full year "subscription" for a ridiculous price. However, there is no guarantee they will behave like this forever so I also have an old laptop with Techstream (the older one but my Lexus LS460L is not the newest and fully supported by that version; however you still need their FULL version with Internet connection to the mothership if you need to re-program some ECU). Not a COMPLETE solution but better than nothing and it allows to do almost everything except a few things that require tethering to the mothership.

As they own the software (tool), they can decide what and how to charge
for it. If they assume most of their customers are dealerships, they
don't need to be "reasonable" supporting individual users.

My gripe is with tools that I may have to "rent" JUST IN CASE their
might be a future use. Let me pay for 30 days of use 5 years from today,
as the need arises. NOT for the next 5 years where it is sitting idle!

The problem is you might be not able to get it in 5 years AT ALL, no matter what's the price you are ready to pay. And paying "rent" for all those 5 years does not guarantee anything -- when (not "if") the company that made that tool goes bust you are fried. It is not even like buying vs renting
your house -- you can find a different one any time unlike those tools that are all unique and once they gone there is no replacement.

There are replacements. It's just that the cost of making the switch is
high. Just like moving into a different house!

to support. Not just some ancient stuff but something one year old or even >>> newer because your tools stopped working and there is no way to make them >>> work again.

Run it in a VM. Take a snapshot of it on day one. When the license
expires, reset the clock and reload from the original image. Lather,
rinse, repeat.

That was only good in early naive days where everything was extremely simple and there was no Internet. Many tools these days won't even start if they can't connect to the mother ship to check your license. Which is also a
great simplification, it is more complex than that and gets even more sophisticated by the day...

But, that's their perogative. You are free to NOT use their tools.
They aren't preventing you from performing those activities; they
are just setting the terms for your performance WITH their tools.

I maintain a *working* development environments for every project I've
created. There is some cost to doing so. But, I provide "lifetime,
free bug-fixes" to clients. So, it's a gamble that I make regarding
the effort that *I* will have to expend to recreate a particular
environment *if* a bug is uncovered.

However, I don't make any guarantees to add features to projects that
I've completed. (I'm typically off on some other project with little
concern for embelishing some PAST design)

Several times, clients have contacted me because they failed to
maintain such a development environment and now find themselves
wanting to make changes to my design (inhouse or with some other
agency) -- only to discover that they "can't get there from here";
they can't edit a schematic, alter a layout, patch a bit of code,
etc. Because they can't get the tools that they purchased to run
on a "modern" OS!

Should I charge usurious rates to take on those projects?
Projects that I *don't* want? Or, just let them realize the
consequences of NOT having proper engineering and CM practices
in place? (I've had clients ask me to recover the source
code for their products -- not of my design -- because THEY
failed to keep it! You only make this mistake ONCE)

I've worked with firms that relied on licensed third party
products to incorporate *in* their own products. And, the
third party ELECTING not to sell any more licenses for the
"old" version, thereby mandating a reengineering of the
product to accommodate the new licensed component.

When You don't control EVERY aspect of your development environment,
you're at SOMEONE's mercy!

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/20/2025 4:52 PM, Lawrence D'Oliveiro wrote:

Would you entrust mission-critical business operations to obsolete, unsupported software?

People do this every day. How do you define "support"? I define it as
"being able to get a problem that I am having FIXED to my satisfaction".

Pick a product (hardware/software). What chance of having YOUR particular "issue" addressed if:
- a commercial product from a company still in business and supporting it
- a commercial product from a company no longer actively supporting it
- a commercial product from a company no longer in business (possibly sold off) - an FOSS product with some structured backing/financing
- an FOSS product without structured backing
- an FOSS product that folks have lost interest in supporting

Depending on the nature of "your" issue, all of the above can leave you flailing about, looking for an acceptable work-around (a tacit acknowledgement that the issue will remain unsolved for the foreseeable future -- at least beyond your anticipated release date!)

I use PostgreSQL in my current project. Very well supported. Some money behind it, too (even if only the Enterprise version, that support eventually spills over into the "free" version). And, likely to remain a viable product/project for the foreseeable future!

I'd like to be able to create "databases" (tables) on R/O media. While not likely a high demand issue, I could see folks wanting to leverage such an ability even if the media wasn't truly R/O (i.e., to be able to define a tablespace that is immutable and, thus, protected from ACL f*ckups, hacks,
etc. -- "No, you physically can't WRITE to those tables!")

But, expecting folks to spend their limited development time supporting MY needs is wishful thinking. So, I have to embrace the codebase and plan on adding those features, myself.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/21/2025 12:50 AM, piglet wrote:

Would you entrust mission-critical business operations to obsolete,
unsupported software?

Far more than I’d trust the ever changing buggy offerings of Adobe/Microsoft/etc .

There's nothing wrong with "buggy" -- if you are aware of the bugs,
how they potentially impact your usage AND have viable work-arounds.

The "ever-changing" is the pisser; it complicates KNOWING what you
have in your hands.

Banking finds COBOL works just fine. I know some
major pharmaceutical plants still run critical processes on HP1000-21MX minicomputers from the 1970s!

These are industries where there is little change. So, over
time, all of the quirks will surface, be recognized and addressed.

[Pharma, in particular, makes change costly by necessitating re-validation]

Where you run a risk is using some "legacy" (polite term for "unsupported") tool in a NEW way; one for which you have no actual knowledge of its performance. Worse, EXPECTING it to perform comparable to the other
uses to which it has been successfully applied.

E.g., a tool that can handle thru-hole PCB layout might fail miserably when tasked with an SMT application.

My first application of AutoCAD's AME (~1986?) worked fine building 3D
models. *UNTIL* I tried to model a perforated enclosure! The number and proximity of the holes "subtracted" from the enclosure skin solid tickled
a bug in the floating point implementation -- in a very obvious (incorrect) way! Had I not attempted that task, the consequences of the FP problem
might not have been as (visibly) apparent. Perhaps two parts might not
have fit together properly, etc.

[It was REALLY easy to provide a test case to the folks at AutoDesk so
there was no fussing about whether or not this was a genuine bug or "OE".]

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On Fri, 20 Jun 2025 17:46:00 -0700, Don Y
<blockedofcourse@foo.invalid> wrote:

On 6/20/2025 4:52 PM, Lawrence D'Oliveiro wrote:

Would you entrust mission-critical business operations to obsolete,
unsupported software?

People do this every day. How do you define "support"? I define it as >"being able to get a problem that I am having FIXED to my satisfaction".

We are not having problems with PADS.

Well, except that it runs under Windows.

I can open old schematics and PCBs that were created under DOS,
literally decades ago, and print, edit, ECO, revise, save. That is
impressive.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/20/2025 1:59 AM, The Natural Philosopher wrote:

Being old and cantankerous that is exactly how I do my simple PCBs to this day,
except that I use Corel draw and layers.

I spent a week trying to get a PCB cad program to work, gave up and did the PCB
in a day.

The amount of effort you put into setting up and learning a tool depends on
the amount of work you, later, expect the tool to do for you. (obviously,
you spent some time learning Corel... likely NOT for PCB layout!)

If I need to dovetail two pieces of wood together, I might opt to
just drag out a coping saw and hand chisel/mallet; it's quick and easy.

OTOH, if I am going to assemble a set of drawers for a cabinet, I'll drag
out the dado jig and a router and plan on investing the time to get *them*
set up, properly.

Schematic capture, PCB layout, mechanical CAD, thermal analysis, etc.
all have high initial overheads that you have to overcome. But, can
reap great rewards depending on the frequency of their use and the sophistication of the problems addressed.

Sadly, this also acts to tether folks to particular solutions -- perhaps
longer than common sense might dictate. E.g., I quickly left DASH-PCB
behind -- and, with it, the rest of the design suite :< -- when it
just wasn't the "right" tool for SMT work (~1990) and SMT was the obvious "future" for my designs!

But, it had more than paid for itself with the time saved and increased billings it allowed in the few years I used it! (and, the hardware had
still more uses going forward)

There is an online website that will convert Corel drawings to the correct format for PCBs.

You might want to chase down an ancient (40 year old) copy of Wintek's smARTWORK. It was a toy layout package that was a breeze for small jobs
(I think it ran on a single floppy!) Definitely more "value" than Corel
when it comes to PCBs.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/22/2025 8:20 PM, Lawrence D'Oliveiro wrote:

On Sun, 22 Jun 2025 14:10:07 -0000 (UTC), candycanearter07 wrote:

Lawrence D'Oliveiro <ldo@nz.invalid> wrote at 23:52 this Friday (GMT):

Would you entrust mission-critical business operations to obsolete,
unsupported software?

A lot of companies do, for one reason or another.

Sure, it works fine ... until the day it doesn’t.

That;s true of everything. INCLUDING FOSS products.

And then, in the words of the song, who’re ya gonna call?

Are the FOSS supporters MORE capable and attentive than
those that have taken money from you and rely on your
high opinion to promote their product to others?

dhcpd(8) doesn't support ethers(5), bootptab(5) nor networks(5).
Instead, the administrator is supposed to HARD CODE these things
in the configuration file.

How can /that/ be a good idea? Ignorance?

How likely do you think they are going to patch the codebase
to support these things?

For a *paid* product, there is a chance that the author can
see the value in yet another feature that sets their product
apart from others, esp if it doesn't BREAK anything.

For a FOSS product, you're dealing with "personalities".
If the Maintainer(s) have a different attitude about how *you*
should use their tool, they are likely to shrug it off as
they have ALREADY ESTABLISHED PRIORITIES for their development
efforts given their limited resources available. ("It already
works without them")

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/22/2025 5:10 PM, Don Y wrote:

You might want to chase down an ancient (40 year old) copy of Wintek's smARTWORK.

HiWire (schematics) and smARTWORK (PCBs) were as awesome as they were
because they were written by hardware engineers for internal use
(Wintek's main products were 8051 dev boards) and later made into
products in their own right. As such, I'll bet the code was pretty
awful :-) but the programs were a dream to use. with HiWire you just
basically thought your schematic onto the screen.

I used HiWire to lay out the ceiling tiles for a kitchen remodel.

--
I'm part of the vast libertarian conspiracy to take over the world and
leave everyone alone.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

On 6/23/2025 10:26 AM, Crash Gordon wrote:

On 6/22/2025 5:10 PM, Don Y wrote:

You might want to chase down an ancient (40 year old) copy of Wintek's
smARTWORK.

HiWire (schematics) and smARTWORK (PCBs) were as awesome as they were because they were written by hardware engineers for internal use (Wintek's main products were 8051 dev boards) and later made into products in their own right.  As such, I'll bet the code was pretty awful :-) but the programs were a
dream to use.  with HiWire you just basically thought your schematic onto the
screen.

smARTWORK did very little *for* you -- beyond the graphics of pads, silkscreens and traces. And, had a lot of "geometry" limitations.

But, it was REALLY easy to use, if you assumed responsibility for placing every component and pad "unaided"/"unadvised".

I suspect the code was actually pretty trivial! It was the ultimate in terms of gridded tool; each "cell" had only a few possible ways of being drawn
(for a large value of "few"): empty, top layer wire traveling horizontally, top layer wire traveling vertically, bottom layer wire traveling
horizontally, bottom layer wire traveling vertically, top horizontal with bottom vertical, top vertical with bottom horizontal, top doglegging north,
top doglegging south, etc.

So, the software just saw each cell as being in a specific state (i.e.,
which of the N graphic depictions POSSIBLE is used, here) and the
user's actions (when the cursor had the cell selected) altered the
possible NEW state of that cell.

In theory, one could have pre-drawn each possible cell representation
and then just pasted them from memory onto the screen (or, algorithmically
do that at runtime -- for very little effort).

But, it was inexpensive (I had a buddy who relied on it for relatively
low-tech designs) and easy to make changes.

It had NO DRC capabilities (that was handled by the implicit design of
each "cell possibility"). And, ripping things up was a purely manual
effort -- you really had to be sure of what you wanted and WHERE you
wanted it!

I often got a call in the wee hours of the morning to help him "check"
his layout against his schematic (something dreadfully boring and
error-prone if done by a single person).

[I recall him moving to using AutoCAD 2 (long before MEP) but that
may have only been for schematics?]

I used HiWire to lay out the ceiling tiles for a kitchen remodel.

I did similarly (with AutoCAD) to layout the floor tiles, here,
taking into account the thickness of the grout lines between them.
In addition to getting a good "count", it also lets you decide
where the "key" tiles should be located for maximum visual effect.

E.g., if you just "start anywhere", then you may discover that
the tiles travel down a hallway "off-center" -- much more noticeable
than if you had centered the tiles in the hallway FIRST and let the
tiles in the larger "other" rooms land where they may.

I also used the same "house floorplan" to document the locations
(and branch circuits) of each electrical fixture and all of the
network drops I've installed. It looks much more professional
than a "sketch" of dubious proportions. And, lets me group
items of related types so I can choose which subsets to show
in a given rendering (e.g., "show me all of the camera drops")

I used Illustrator to design the models for my gesture recognizer.
I could just "draw" a gesture (e.g., a "bowtie", a "box", a "spiral",
the digit '3', etc.) and then *extract* the few lines of PostScript
from the file that actually define that "path" -- instead of
trying to create those textual representations from scratch.

Knowing how your tools work gives you insights into how they
can be (ab)used to your advantage!

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)