Ask HN: Who's still keeping a DOS machine up because the business depends on it?
1 day ago
Do you currently work with or know anyone who is still using:
* dBase/Clipper/CLARION/Paradox/other DOS RAD environments on period hardware to run business processes?
* CNC mills/spectrometers/microscopes/other industrial instruments controlled by ISA cards (either bespoke or standards like GPIB)?
* Anything with a parallel port dongle?
If so, I'd be very interested in hearing your experience here, or feel free to send me an email at the address in my profile. I'm not trying to sell anything, just doing some research for an idea around keeping these going on modern hardware.
A certain nuclear power plant had a Windows NT 4.0 machine running as late as 2007. The reason is interesting.
The machine's purpose was to report status of the control rods that mitigate nuclear reactions. Basically, "are the rods inserted, and if so, how many / how far?". I want to emphasize that this was reporting only, NOT control.
The original software was written back in the 80's, when the plant was originally commissioned, for AmigaOS. Of course, it's hard to buy Amigas anymore, and the original one died long ago (nobody remembers when).
So in the mid '90s, the utility purchased an AmigaOS emulator that ran on Windows NT 4.0, which was current at the time. The emulator (IIRC) was developed by a firm in the UK. The firm went out of business sometime in the late '90s. The control rod monitoring software ran under this emulator on top of NT4.
Windows NT 4.0 was the last OS to allow the emulation software direct access to the physical hardware that produced the status signal. Later versions of Windows abstracted the hardware access away, and the monitoring software broke. Because the emulation company had gone belly up, there was no way to fix the incompatibility.
So the utility had a choice: get new hardware/software certified (by NRC?), or keep doing what they were doing with the software (and hardware) that they had. They chose the latter.
So this is how, in 2007, during a tour of the facility, I stumbled across a Pentium 1 system running an AmigaOS emulator on Windows NT 4.0 that was responsible for displaying the status of the control rods of a nuclear power plant.
Spare hardware for this setup was purchased off of eBay and stocked on an adjacent shelf.
To me the funny part of this story is that they used to show you this warning as part of the EULA when installing Windows NT 4, which I remember joking about:
NOTE ON JAVA SUPPORT. THE PRODUCT MAY CONTAIN SUPPORT FOR PROGRAMS WRITTEN IN JAVA. JAVA TECHNOLOGY IS NOT FAULT TOLERANT AND IS NOT DESIGNED, MANUFACTURED, OR INTENDED FOR USE OR RESALE AS ONLINE CONTROL EQUIPMENT IN HAZARDOUS ENVIRONMENTS REQUIRING FAIL-SAFE PERFORMANCE, SUCH AS IN THE OPERATION OF NUCLEAR FACILITIES, AIRCRAFT NAVIGATION OR COMMUNICATION SYSTEMS, AIR TRAFFIC CONTROL, DIRECT LIFE SUPPORT MACHINES, OR WEAPONS SYSTEMS, IN WHICH THE FAILURE OF JAVA TECHNOLOGY COULD LEAD DIRECTLY TO DEATH, PERSONAL INJURY, OR SEVERE PHYSICAL OR ENVIRONMENTAL DAMAGE. Sun Microsystems, Inc. has contractually obligated Microsoft to make this disclaimer.
Also:
The machine's purpose was to report status of the control rods that mitigate nuclear reactions. Basically, "are the rods inserted, and if so, how many / how far?". I want to emphasize that this was reporting only, NOT control.
It would take a whole lot more context to make this somehow comforting. :D
> It would take a whole lot more context to make this somehow comforting. :D
This is this fascinating 6 part documentary about the Chernobyl incident explaining how it was caused by bad control rods. But the main point is that control rods prevent the facility from going boom, so be glad it’s not the AmigaOS emulator on a NT machine handling it.
This is a great post. How does someone write software that needs to run for ~50 years where the hardware will need to be replaced with non-equivalent, newer hardware? If I were facing this issue today, I might start with an OS that has excellent emulation. Example: Can I run 32-bit MS Windows 95 via emulation on a variety of current 64-bit OSes, like MS Windows, Linux, AIX, HP-UX, etc. If yes, then we can assume(?) this emulation will remain relatively stable even if we upgrade our hardware later. Maybe I am overthinking the whole problem: Can VMs do exactly what I want today? Will VMs running ancient OSes, such as 32-bit MS Windows 95, continue to be stable/viable in the future? I am unsure.
Make sure that you have as thorough a specification of what the system is supposed to do as you can.
Then define the version control system and the build process, specifying the dependencies, and so on.
Think about any opaque blobs in the system and try to eliminate them so that you have plain text source code so that no tools are needed to read the code.
Make sure that the build process runs entirely locally and never fetches anything from outside.
Simplify everything, use only tools and languages that are well understood and supported.
The real problems are not strictly technical but social: how do you prevent loss of the code, the tools, the specification, how do you maintain the expertise needed to maintain it. How do you ensure that all those things that are obvious to you now are written down in all their gory detail so that your great grandchildren will not apply their new and different preconceived ideas to the system?
Document all this on paper as well as electronic storage, make sure that version ids are recorded on every page as well as being available to the user of the machine or program.
In the industry in which I worked for the last thirty years of my career it was not uncommon to have things come back for repair after fifty years use and to be able to consult the original drawings and bill of materials so that exact replacement parts could be made.
1 reply →
Wasn't there some idea that went like "if something has been around for x years, it will likely be around for x more years"?
Maybe follow that. It's 2026 right now. What way of writing code would have worked 50 years ago (1976) that still works today?
Well, C, SQL and Lisp were all around 50 years ago and still exist today.
In terms of hardware, 50 years ago there was intel 8080. apparently code for that can still work via emulation today.
So if you wrote C for low level, Lisp for high level, and compiled the software to run in an Intel 8080 emulator, that combo would likely still work in 50 years.
1 reply →
> How does someone write software that needs to run for ~50 years where the hardware will need to be replaced with non-equivalent, newer hardware?
Some ideas:
Write it in a popular language/ecosystem, stick rigidly to well-defined APIs, use commodity hardware, flatten out any malignant cleverness, maintain documentation on why every part does the thing it does, and make the source code readily available.
This is based on working with some very old systems, and each point above is the opposite of something that made life harder.
6 replies →
run for ~50 years where the hardware will need to be replaced
If it needs to run for fifty years, specify hardware that has a long support cycle such as Mil-Spec.
That’s why the Z80 was around for so long and why so many companies in the 80’s and 90’s still deigned around it. (1)
If you let programmers drive system design, winding up with an “Amiga” in your certified design is a more probable outcome…
(1) there’s a moderate chance that new hardware that can run Z80 code will be around fifty years from now.
> how does someone write software that needs to run for ~50 years where the hardware will need to be replaced with non-equivalent, newer hardware?
Make it open-source, or don't buy without source.
3 replies →
This was one of the promises of Java. You write your application and then you get a virtual machine (JVM). As time goes on you only maintain the virtual machine while the application remains the same.
> How does someone write software that needs to run for ~50 years
You in a lot of ways have to respect the ability to have done this in the first place, compared to todays slopware which manages to break within 1-2 mandatory OS update cycles.
Write it in C or Fortran.
I'd recommend not having it so dependent on the hardware and choose an OS that is relatively hardware agnostic (e.g. Linux). Also, ensure that the software is open source and if compiled, has an open source toolchain to do so.
Though people often decry it (due to the many, many footguns), writing something in Shell/BASH will make it trivial to move to newer machines.
A friend works for an airline as a flight simulator tech. Their entire software stack, including the compiler and OS, is FAA-certified.
Then their ancient Honeywell(?) mainframes reached end-of-life they scouted for compatible hardware, of which there was none. The cost of certifying new software, plus the time involved, was astronomical. So, after consulting with the FAA, they paid a hardware company to clone the ancient mainframes in modern silicon. The FAA signed off on it, and they had all-new computers - much smaller than the originals - running the old stack.
It's _very_ hard to believe some company to simply "clone the mainframe" into chip. Mind sharing any link to this effort?
13 replies →
That's interesting. That means there are fewer checks for hardware than software?
3 replies →
>Then their ancient Honeywell(?) mainframes...
https://en.wikipedia.org/wiki/GE-600_series from the 36-bit era which turned out to be a sweet-spot for Lisp; also "famous" for running Multics (old joke flipped: "which was many of whatever unix is one of")
GE sold it's division to Honeywell, and from there to Group Bull (French) and then to NEC (Japan)
Having worked on nuclear plants (as a reactor operator) around that time this doesn't surprise me in the least.
Thats far more advanced than the systems I worked with, one of which reported rod position via resistance measurement on a brushed cylinder (one for angular and one for depth).
Cleaning and calibrating those was a constant maintenance item every time the reactor was shut down.
I grew up in a little post soviet republic, in a small town with a big nuclear power plant.
Coincidently, that's where my father was working as a programmer until the shutdown of the plant.
Anyways, he was maintaining a bunch of systems related to collecting data from environment monitoring sensors around the plant. In mid 2000s or so he showed me some data collecting machines using what looked like DOS text UI that mentioned being written in Leningrad, 1989 or 1988.
The software was running in an emulator on Linux, which my father used to replace original x86-compatible machines runninh dos.
Pretty sure thr whole thing was running for quite a while since my family moved elsewhere.
Did they not need NRC recertification when they moved from a physical machine running AmigaOS to an AmigaOS emulator?
That is an advanced question best asked of the folks in charge of the plant. At the time I was just trying to figure out how to explain to the IT auditors why there was no antivirus software on this piece of crap.
8 replies →
It is wild to me that you would use any GPOS for something like this instead of dedicated firmware for the task. This is evoking similar reactions to when I read articles about infrastructure being hacked remotely: Something must have taken a wrong turn with the architecture design for this to happen!
This reminds me of the apocryphal story of the IBM System/360 running a 1410 emulator that ran an IBM 705 simulator that ran a business-critical application.
Not wholly apocryphal. The U.K.'s air traffic control ran, and possibly still is running, on an IBM 4381 substituting for an IBM 9020.
* https://news.ycombinator.com/item?id=49763972
They've reportedly already tried, and failed, to replace it with a more modern system.
Which is why we should target VMs directly. Then that VM can be brought forward to new platforms rather than emulating the entire machine, not that emulating the entire machine is bad. One could argue that the emulation stack you outline was made possible by putting in the work to make simple, fully specified and documented hardware.
I find it deeply upsetting that anything important for a nuclear power plant is using an OS without protected memory.
You don't need protected memory if your program is the only thing running. Plenty of microcontrollers have no protected memory and get used in safety critical applications.
Protected against what?
3 replies →
NT4.0 in 2007? That wasn't even very late. I was replacing NT4 servers into the early 10s. Car dealerships were terrible for keeping that crap around.
"Windows NT 4.0 machine running as late as 2007"
lol. I know a ... factory, that had a BBC model B (with a rather complicated wiring loom) still doing a job around that time.
The IT supplier at the same factory went to a museum in Cambs. around late '90s, early '00s to ask if they could buy an exhibit because something had failed locally. The museum gave them the part.
That was just aerospace and nothing fancy like your nuke plant!
> A certain nuclear power plant had a Windows NT 4.0 machine running as late as 2007.
To put this in perspective, NT 4 was released in 1996, 11 years before 2007. That’s roughly the same as someone running Windows 10 today, which a lot of people still do. The Pentium 1 was only 14 years old at most. These parts of the system shouldn’t be surprising.
The AmigaOS and software from the 80s are of course the older more interesting bits, but for critical infrastructure and safety, shouldn’t we expect and design for our hardware and software to last a long time, and not try to keep up with tech fads every decade? I feel like the main problem isn’t hardware or software that’s old, but that there wasn’t a longevity plan.
It’s always been an interesting question how to write software that will last a hundred years. Maybe with AI this is the first time in history that planning to port & upgrade to new hardware every 5-10 years seems totally reasonable.
> That’s roughly the same as someone running Windows 10 today
While true the situation is different. Hardware and software moved a lot faster in the 90's and early 2000's. Everything has matured, is far more standardized, and moves more slowly now. Hardware from 11 years ago when Win10 was released isn't that different from today's hardware. Same goes for Win10 and Win11.
3 replies →
Nothing wrong with this.
If it ain't broke, don't 'fix' it.
Just make sure that its well isolated from the Internet!!
Was this status indicator simple vertical bars that were black and white and displayed on an old CRT Monitor? Almost looks like black and white terminal vertical bars (like Alsamixer)?
I'm curious, how did the hardware present itself? It couldn't have been an Amiga Zorro card since it would've been impossible to get that into a PC. Did it connect over a serial or parallel port?
I recall it being some custom ISA card with connectors on the back. The emulation software had drivers that interfaced directly with the card.
2 replies →
How on earth are these massive pieces of infrastructure relying on software they don't have the code to? i'm just utterly confused how these things happen.
If it's still running I wonder what they are going to do about the year 2038.
Set the date back to 1980 and sticky tape over the corner of the screen where the date is displayed.
I wonder if there is push in the public sector for open source software/hardware for cases like this.
I completely get the decisions over time here, but it's unsettling having a relevant piece of software (reports are important too!) working on with parts from ebay, in 50 years time they might be gone.
I've dreamt of governments hiring devs with promises of full time OSS work. It _might_ get some people over the lower salaries of government work. But I overheard a state CIO once say "We configure, we don't create".
That's to say nothing about opinions of open source that rival early 2000s Microsoft.
For old ass critical software and hardware, the org can always have a backlog of hardware and contingency plans (paying someone to fix the hardware, paying someone to create a fully certified and modern solution..) for when they start running out of parts.
This is crazy. Why on earth wouldn’t the respective government spend money to modernize a critical infrastructure like a nuclear power plant?
This is how we would end up with nuclear disasters, not because the technology is bad, but purely because of mismanagement and human negligence.
This is crazy. Why on earth wouldn’t the respective government spend money to modernize a critical infrastructure like a nuclear power plant?
Safety and reliability come from understanding the system. Something that's old but that you know everything about is far safer than something new. This risk is that the people who know move on, or the sources of replacement parts stop working, or that other things around the system change. Then the risk curve inverts and you find the old system more of a liability than a asset.
Almost always people choose to update things either too early or too late. Knowing when to do something is hard.
5 replies →
> This is how we would end up with nuclear disasters, not because the technology is bad, but purely because of mismanagement and human negligence.
I'm pretty sure constant upgrading would be more likely to cause disasters.
1 reply →
If it's not broken, don't fix it.
Same reason why MTA in NYC still operates subway using 100 y/o signaling hardware on most of the lines.
In a way, your comment and the responses you get is a perfect allegory to inexperienced vs experienced engineering.
6 replies →
> Why on earth wouldn’t the respective government spend money to modernize a critical infrastructure like a nuclear power plant?
Same reason they don't modernize health care infrastructure. It's too expensive, people die if they get it wrong and they get voted out of office for their troubles.
Lindy's law:
For non-perishable things - i.e technologies, ideas, books, institutions - the expected remaining lifespan is roughly proportional to how long they have already survived
Time acts as a filter: what has already withstood a long stretch of disorder is more robust, so the longer it lasts, the longer it is expected to last.
The old nuclear software running on mainframes have passed the test of time. The new software, has yet to.
Nassim Taleb talks about this at length in his book, Antifragile.
Now you know why "nuclear power may have been dangerous in the past, but with modern technology it's completely safe and Chernobyl will never happen again" is not a good argument.
6 replies →
There is no reason a more modern system would be better in any aspect.
>So this is how, in 2007, during a tour of the facility, I stumbled across a Pentium 1 system running an AmigaOS emulator on Windows NT 4.0 that was responsible for displaying the status of the control rods of a nuclear power plant.
Vernor Vinge's A Deepness in the Sky depicts a human society thousands of years in the future, in which pretty much all software has already been written; it's just a matter of finding it. So programmer-archaeologists search archives and run code on emulators in emulators in emulators as far back as needed. <https://web.archive.org/web/20231114211656/http://www.gareth...>
(Heck, recently I migrated a VM to its third hypervisor. It began as a physical machine a quarter century ago.)
That must have been a crazy moment for you when you first saw it and a big sigh moment for whomever had to explain it to you the first time. Thanks for sharing this; it made me smile!
If "a few years ago" counts, a relative of mine abroad used an MS-DOS application attached to an Epson dot-matrix printer that supported a PITA foreign script to print invoices. It wasn't too painful, since the system supported batch jobs fed by a format that Excel could export.
[Source: Was NT developer - have code in NT 4.0...]
The big problem will almost always be some hardware dependency that breaks, almost always in I/O. I've seen video software used by TV producers that depended on an ISA (not EISA nor PCI, ISA) card to read NTSC input and impose text over it. And it broke as they were getting ready to add commentary to a live broadcast of a race (which I was driving in.)
The world is full of machine tools and metrology devices that are stuck on some ancient computer/software because they made a weird proprietary protocol over the top of a centronics connector, or some other "WHAT were they thinking??" construction...
All of these could be overcome with a new I/O device, and/or writing a device driver. So using say a raspberry PI that talks ethernet to your "monitor" PC is probably relatively change proof.
If there is any documentation, if anybody can find said docs, if there's budget for a new I/O card and the device driver to go with it. Often the parties who have those resources would rather sell you a new machine tool, metrology machine, or I suppose, nuclear reactor (:-)
Of course, certification is a big deal... A particular certification structure led to the 737 MAX tragedies. And since certifying authorities don't always have detailed tech knowledge of what things are, they have a hard time saying "no, you can't slide that by, you have to build and certify a new thing".
Not MSDOS, but I feel obliged to shared the story of an auto shop in Poland using a Commodore 64 to run the calibration and wheel balancing some years back. It was viral news at the time but worth sharing again.
https://www.youtube.com/watch?v=tXdLtnt-nvE
Original Polish article - https://www.trojmiasto.pl/wiadomosci/Warszatat-samochodowy-z...
I wouldn't call them business-critical computers, but I do have about 3-4 computers under one of my workbenches dedicated to certain work I did for a client. One of them has an EPROM programmer set-up, which includes a custom ISA-card, and a ribbon cable to an external box with a ZIF-socket. The software for that requires MS-DOS. To network that box, which is a 486SX, I run WFWG 3.11. Since these never go on the enternet or anything outside 'the lab', I don't worry about the old MS networking.
A second computer has an MS-DOS program that captures data input on a pair of RS-422 ports (two more special ISA cards) That let me look at packet data at a low level and watch which serial port received what data, and track character-level handshaking of a proprietary protocol.
A third computer under Windows 2000 ran a 68HC11 cross-compiler that didn't require a dongle. I also used this to run a remote debugger that allowed me to single-step through the 68HC11 code running on the target but moved to the target's RAM. To do this 'properly' I was constantly generating hard-copies from the compiler which would show me the C-code, and the resulting assembler code that I'd track through the debugger. I used a lot of paper on a project sometimes.
A fourth computer with MS-DOS, with WFWG3.11 also installed on it, had all my old CAD programs on it, from before they required dongles, so that I could pull-up drawings of some of the hardware I designed. Some of those were finicky about which mouse I use. I still have an old Logitech mouse C7 that two of my CAD programs required.
I was working on a replacement platform that put this into an Atmel-based controller, shipped a couple of samples, then went blind in one eye, while my client had a stroke and was out of action for a year. Between both of our health problems, that project kind of died.
I worked in a state gov't agency as a temp, twice, in the same department. Several of us had shared access to a PC that let us log-into a special terminal program and read records on some mainframe someplace. That PC had a 233-MHz cpu, I'll assume single-core. At that time, it ran Windows NT 4.0, and it had been patched to use a USB mouse. But we didn't rate an optical mouse, so someone would come in twice a day with a tote of cleaning supplies and clean the mouse ball.
After September 11, 2001, somebody came in from the federal government and ordered Windows XP to be the minimum OS for some security issue. The person who had to update that system said a 233 MHz machine was too-slow for XP, and she didn't have a replacement computer, she was already short on computers.
The next time I saw it, it was running XP.
dBase, MS-DOS 3.x, not using industrial anything. And as its a front of house tally machine, there's literally zero incentive to ever upgrade it. Downtime is measured in the yearly reboot, but there are non-operational business hours, so everything is just scheduled around that.
We've already got it running on modern hardware. It's running under qemu. And the dBase stuff gets ripped out and sent to a REST server for broad monitoring and so on.
We... Have one small oddity? There's a tape backup system, that throws everything through the soundcard. (Sound Blaster only.)
Would be nice if onboarders didn't see dBase and just throw everything at AI instead of actually learning the skills they'll need when data migrations happen. But that's a people problem that can't be solved with tech.
Half of our system is running on 2002 versions software. To my desperation, my calls for updating has been persistently denied. There're no technical reasons to keep it as is, it's more about wetware issues.
We still use dBase files in some point as transient format, not using dBase as such though. Oh, and plenty of positional text files, but that's more a question of interop with partners.
Not any longer. Back in the mid-2000s the company I worked for sold point-of-sale systems based upon a multi-user 4GL application (TAS Pro). Similar to dBase/Clipper etc., using the Novell Btrieve ISAM database engine.
With the advent of USB and SATA we ceased to be able to source industrial embedded x86 boards which could run the application. At the time this was with Novell DR-DOS, which was relatively modern in terms of hardware support but still behind the times and we were struggling to manufacture new systems. The company was very small, so a full rewrite was investigated--I did a prototype using PostgreSQL and Gtkmm--but it wasn't realistic. Today, an AI could probably do a full rewrite in a day or two, including migration tooling. Back then, it would have been a multi-year effort for one person given the application's size and complexity.
My solution was to run Linux since it had full support for the hardware. Debian Sarge at the time. This used a Perl frontend and dialog(1) to present a simple menu system at startup. This did backups, software updates (over dialup!), remote access for support (again over dialup), ran backups and ran the main application. It would start DOSEMU which provided the application with VGA display, COM ports and parallel ports for the application to drive "directly" (from its perspective). It also used Samba and CUPS to provide the DOS environment with shared network drives with exclusive byte-range file locking needed for the multi-user network database to work with concurrent users without data corruption, and also multi-user report printing and receipt printing.
I left the company a year after this was put into full production, but the last I heard it kept the company viable with a supportable product for many years after until its owners retired. This kept software from the early 1990s running well into the 2010s, and there are likely still sites running it to this day.
Not DOS, but I sold a '99 HP computer running win98 last month, to replace an identical failed system. It was used to control a 50 meters long custom industrial paint booth line for special equipments (like 57" rims). Luckily, their HDD was OK. Urged them to make a few sector-to-sector copies of it on a few spare HDD. This $100 sell surely has prevented this business from massive bills or even going bankrupt in the meantime from production loss.
Insane to me that someone would use a windows machine to run a piece of equipment when PLCs exist and are supported for decades by companies like Rockwell, Siemens, Allen Bradley
There are lot of windows based solutions since more than 20 years ago for two main reasons : recipe management system, and logging. A basic SCADA of sort. PLCs systems of that era could either not manage it (TSX, S5,S7-100/200), or it would be additionnal very expensive, complex and typically OPC-based solutions (I'm shivering).
Some of theses solutions are built upon common PLCs but some are built as standalone solutions when I/O are simple enough (modbus mainly, RS232/422/485, ethernet/IP, controlnet, ...). Seen lot of them in breeding nutrition management systems, in-house developped processes, ...
Think that’d be worth more than 100 :)
I had it listed at this price for a few months, simeone from their company purchased it and rode 2 hours and half to come get it. Couldn't and had no reason to up the price at this stage. Indeed, this was a very good deal for them.
I know a shop that still runs an Apple ][: they have a Numerical Control machine. For those who don't know, NC is the predecessor to Computer Numerical Control, i.e. CNC machines.
They run on punch tape.
So this Apple ][ is connected to simulate the punch tape reader, and send the (I think) G-code commands to the machine. So it's now a CNC, managed by an Apple ][.
I had a functional STA Compact [1] coagulation analyzer until 2022. It ran a customized DOS version on a ITX computer board that was integrated into the analyzer. UI was in text mode and used the keyboard only. It had no mouse. The video in the link shows the UI for a few seconds at the end. We kept it functional as a backup and even used it a few days when the replacement broke down. It was one of the best analyzers I ever used.
The newer version of that anaylzer, STA Compact Max [2] has mostly the same analyzer hardware, but newer ITX board inside. It runs Windows (XP, 7, 10...). Bugs galore.
I know labs that used a Beckman-Coulter HmX hematology analyzer [3] until 2015 or so. That machine is attached to an external MS-DOS computer via a very thick cable and an ISA board. The cable connector looked a bit like a 68-pin SCSI-3. The software used VGA graphics mode.
2014-2017 I was supporting a clipper setup for issuing work order and many other tasks for an old civil aviation repair plant. I’ve ended up virtualising and moving it to a hyperv cluster. Definitely the case of “if it works, don’t touch it”. Also set up a closed network for CNC machines to let engineers in the shop send jobs over the net instead of running around with floppies or manually entering gcode through the console. Fun times.
We have a client (a factory) I wrote a massive (for one person without AI) Clipper system for that runs the factory; this is decades ago but it just works and when they tried to replace it (not by us, but promises by others), it was worse and they did not switch. Cheating, but another client is a bank; they found its is really inefficient to not use DOS. Their machines use Windows but fullscreen DOS box you cannot close or minimize with the app in it.
Not DOS, but two similar era cases.
A FTSE 100 company I used to work for, still maintains a SCO UNIX machine that was first installed in 1993. It hasn't had any upgrades since, has specialist hardware in it, and still has an IP stack that is classful (ie. no subnet masks, 10.x.x.x is always a /8). It runs in a 100% airgapped network, and it isn't essential to keeping the core companies operations, but it is a ticking time bomb nobody seems to want to address that will cause a lot of distruption when it finally dies.
Also I was recently (very lightly) involved in a consultancy project to remove an OS/2 based point of sale system for a small local retailer with 6 stores.
If you don’t mind me asking, how can a thing that causes disruption when it dies still not be considered essential?
Blind Control for the C-Suite?
This thread reminds me that there is a huge benefit to using a super common OS+hardware.
In 30 years time it will all be outdated... But emulators will exist and a business process in 2050 that still depended on something I built today could still run with minimal effort or risk.
If you have the source code it's going to be a lot easier to continue.
It depends on the gcc/llvm version.
AND more to the question - the contractor who remodeled my house in 2014 was still using some ancient DOS app, possibly running on a DOS machine or at least an emulator of one, to generate billing. Which was handed to me on paper....
(He was a good contractor, so clearly having the latest billing software is not the deciding factor in that.)
An old story. I used to work for Fox Home Entertainment (CDs etc)this is 20+ years ago. And we had proper DOS running on old machines. We used it to do all our finances (including all of Japan, Asia sales). Most of the money we made on movies like Titanic were tracked on DOS. Until about 2003 if memory serves me where we moved to SAP. Why this was worrisome was how easily fraud occurred. In my tenure there $5-10M was stolen. At the time sent to companies in Brazil. I'm still surprised by larger firms where they keep these things alive to retain the data. CocaCola did it. A few large Insurance firms too. Crazy.
A company I used to work at still ships a piece of hardware with an Intel 186 microcontroller in it. The compilers are 16-bit DOS applications and the tool to program the DIP packages with the program was also a DOS application. I’m not sure what the physical interface between the programmer and computer was.
I run 2 old Juki pick and place machines. They run DOS 6.22 on a 486 I think. My biggest pain points are, 1) booting them up without errors, and 2) transferring files to them wirelessly without any floppies. (without installing any DOS software)
Closest I got was with a GOTEK and an ESP32 emulating a USB. But it was flaky so I dropped it for now.
Would a Grease Weasel work for that situation?
The Detusche Bahn still runs DOS and Window 3.11 in their ICE I trains! They searched for admins 2 years ago. https://www.tomshardware.com/software/windows/ms-dos-and-win...
i don’t want to say this explains things, but i do want to imply it
You should check out the NYC subway, which has electromechanical relays, cloth covered wiring, and pre-WW2 analogue panels.
They do not even have an OS!
Let me assure you that Deutsche Bahn also still operates about 500 mechanical signal boxes, about 100 are from pre-WW1. Not these fancy, modern electromechanical relays (which Deutsche Bahn of course also has).
That's far superior than running an outdated general computing platform. Electromechanical devices are simple, easy to understand, and replaceable. It may be harder to diagnose problems but with maintenance it will work forever.
1 reply →
Not DOS but we recently had to virtualise a Windows XP box as it was running crucial software for a visitor attraction. The software was written by a member of staff who sadly passed away and the PC it was running on was approximately 20 years old.
We managed to virtualise it and then run it in a VM, and it needed a couple of USB-to-serial adapters passing through as well.
We're also finding some kiosks are running Shockwave Flash apps provided by companies that are no longer trading. We can decompile the files to extract the assets but at some point it's easier to just replace completely.
I've not virtualized things in that way for ages. I think I was using VirtualBox years ago and com0com for serial communication.
What do you use for VMs these days?
Me!
I have a client with some business software that runs on DOS. They have a dozen seats. A single Linux server gives serves a dozen VNC sessions with the software running in DOSEMU. Printing is done with the Linux print stack and three Ethernet print servers.
It has been working without a hitch since 2008. Oh, so has the server, which is a VirtualBox VM.
The Gear cutting job shop I worked for had a CNC lathe running with a GE Computer from the late 1970s that stored data on punched Mylar tape, until it was bought out in 2019.
The bevel gear generator setup was some formula calculations running in version 1 of TK!solver, which would have been a $5000 upgrade to get the latest version, so they ran them under MS-DOS in DOSbox, and saved the grief.
Without that software, they would have to pay Gleason $500+ to do the calculations for each new gear setup.
There are strong and reasonable reasons to keep very old systems alive in offline production environments.
I had to keep an old clipper/dbase application running until recently. It worked perfectly seamless with vdos (vdos.info). It's commercial software, but very reasonably priced. Dosbox is another alternative, but didn't work as well with printing and I think also wasn't able to do row locking of dBASE databases over the network.
The last company I worked for still had customers they were dragging forward to get upgrades from coax connexted green screen terminals to modern network thin clients. It wasnt on DOS, but very old unix OS, tape drives for backups, parallel for printers, serial for scanners. No USB. Lovely troubleshooting when you were dealing with 30+ year old hardware setups. There was only one company that would even servicenthe green screens if they failed, and you couldn't buy them anymore.
One of our vendors is a contract electronic manufacturer (circuit board stuffer) that has a wave soldering machine running MS-DOS on a PC that's built into the control cabinet. It uses the parallel port to communicate with Opto22 industrial i/o modules, which then is wired to all the temp sensors, relays, etc. It just keeps working.
My dad ran a commercial 2-way radio business. There were multiple radios with programming software that would not work on anything above a 386 and DOS. He had a couple computers from the late-80s running for nearly 25 years, though I think he had to replace one of them with another of similar vintage. Also, had a 486 that he used for customer notes in WordPerfect up to about 2015.
I suspect he could have found a way to use a newer computer, but if it ain’t broke…
Intel was seling 386 CPU up until 2007. I personally was using brand new special equipment with 386 soldered to custom MBs with no VGA, serial only access. I suppose you can purchase such boards from ebay/recyclers today.
I still have a circa-1996 486 running Windows 95. It has an EPROM burner card and software that I use to reprogram the ECM in my car.
The cost of upgrading to modern software and hardware is substantial, plus the learning curve for something I only do rarely nowadays.
Siemens Field PG. It is a rugged Notebook for PLC programmers. At least in the last version until last year, the "Field PG M6" was possible to buy the Notebook with a special serial port for old S5 PLCs ant the Step5 software came in a special DosBox version with it.
The Field PG has also another special port called "MPI/Profibus" for S7 PLCs, 2 Ethernet ports and a DVD drive.
https://support.industry.siemens.com/cs/document/109766662/d...
Not DOS but another dad in my town works for one of the opera houses in NYC.
They have what are called "supertitles" [0] so that people can follow along with what the opera singers are actually saying. There is one on the back of each seat in the theater.
As near as I can figure, it's basically hundreds of LCD displays on some kind of bus network and connected to a computer via serial connection aka "D pin connector".
The software only runs on Windows XP and he has to keep finding old Windows XP machines on eBay in order to keep it working. I think he also has a couple hard drive images of the working OS + software.
I had suggested using a USB to D pin adapter and then use an LLM to sniff the protocol and then reverse engineer it.
His response:
"It HAS to work correctly. In 30 years we've never had an outage with the old system."
0 - https://en.wikipedia.org/wiki/Surtitles
Ah, that sounds like it's The Met, who have a bespoke electronic libretto system.
You're right, it is a bunch of LCDs on a bus (iirc rs485).
Couldn’t you install a cracked windows xp on any kind of machine essentially? Or what is special about “windows xp machines”?
Windows XP drivers don't exist for modern computers. I do wonder if 32-bit Windows 10 would be an acceptable alternative, though.
Does the idea make use of https://86box.net? Just heard of this recently on hn and it seems promising for anything that requires low level/cycle accurate emulation vs. your typical high level emulator.
It should work with Siemens MPI/PPI interfaces, but I predict issues with ex-Telemecanique (now Schneider Electric) unitelway realtime DOS driver which is extremely picky.
My idea is a DOS emulator with a really thorough log of every DOS interrupt called and I/O port interacted with. This could then be used to automatically, say, insert rows into a sqlite table when the DOS app updates a .DBF file, or forward GPIB data to a modern USB adapter. I wanted to avoid low level/cycle-accurate emulation as much as possible, only adding quirks as needed to support the most popular business application suites.
I work at a multi billion dollar facility for a Fortune 100 company and we still depend on DOS. Keeping the mandatory DOS machines going is an exercise. Especially integrating them into modern systems.
Edit: There is no required hardware, however maintaining a stable serial connection through a VM and timing issues due to faster hardware has been fun.
The serial port stuff is very close to what I'm working on - what's on the other end of the connection? What does "integrating them into modern systems" look like for you: getting data out/pushing data in/both?
We typically integrate them by encapsulation. Building multiple layers of more modern systems for them to communicate through. A combination of Windows and Linux boxes. The other end of the connection is very old industrial equipment. This is not to run the equipment per se but to configure and program it.
Actual MS-DOS or FreeDOS?
MS-DOS or Windows 2003 running DOS vdm.
Just out of curiosity: is it possible to somehow attach an ISA Adapter via USB to a VM? ISA does depend a lot on DMA and timing, but the clock is only 8 mhz and gives a software layer on a modern cpu a lot of time to do stuff between ISA bus cycles.
Keeping old hardware running is very fragile. Most spare parts are decades old now too, and nobody knows if they still work properly.
To a VM? Possibly.
The DMA might actually be the easy part.
The main concern is how many VM exits per second a modern CPU can handle. Even though the interface is "slow", you can probably do ~500k to 1 million IN/OUTs or MOVs per second to ISA. You'd have to trap (VM exit) every access to a IO port or memory within the ISA window.
During the pandemic I cobbled together a KVM-accelerated emulator for an old 386EX based board I had and trapping every memory access to IO was slower than the real hardware.
USB and VM latencies are both questionable here, I would think.
ISA motherboards still work and have been mass produced in the hundreds of millions. (I have about a dozen in my retro closet.)
The esoteric ISA expansion cards would be the bottleneck, not the motherboards.
All those layers add some latency, but 8 mhz is really slow. USB 4 can do extremely low latency, otherwise PCIe over USB wouldn't work.
3 replies →
I bet most AI is capable enough now to migrate you away to some emulator for a few dollars.
And I would even bet that in most cases you could run a solution on a sub $10 board from aliexpress.
or a $25 android phone
It's not totally for business, but I run some antennae and radio telescopes that feed through a Gateway 386/33/16MB RAM. As well as a Getac, no idea what model, but it's from the early 2000s. They both feed through to a A/B switched Paradise CRT (so cute). dBase is on the list.
It's been a few years now, but I know a major manufacturer had some windows XP boxes around because they were the newest hardware/software they could get with native parallel port support to interface with their machines.
To give you an idea of the complexity of the manufacturing lines, shutting the machines down and starting them back up again was measured in days. Upgrading them would be obscenely expensive, let alone replacing them, just in the opportunity cost of not making anything alone.
For anyone that needs true native serial & parallel ports, a Dell Latitude E series laptop (or Precision 7xx0) with docking station is an easily-available option.
The last machines with dock port have 6th gen Core CPUs (Latitude E5x70), or 7th gen for Precision 7x20.
I don't really know the full details, as it was my brother who worked there and we just chatted over beers, but I believe the software itself that communicated over the parallel ports was 16 bit, and that (along with the native ports) was the reason they kept the old windows XP machines around.
It is possible, perhaps almost a certainly, that the company did not own the source code for the application and could or would not attempt to rewrite it to work with a modern operating system.
This is a top brand name company, so it isn't like they couldn't throw money at the problem, but with the number of lines in the number of plants they had, there wasn't really any reason to not keep the old hardware running (sans Internet connection).
Not as pure, but for what it's worth I've had success with a Syba SD-PEX50023 in applications where USB-based parallel port emulation failed.
Wouldn't an example of the given scenario be writing code 50 or more years ago in mainframe COBOL, and then still expecting to run it to this day?
I've an AMD K5 processor machine running at 100 mhz. for testing Transputer ISA cards when someone buys one, I also use the same machine to format and copy my Pascal compiler into a pair of 5 1/4" floppy disks. All using DOS.
> Transputer ISA cards when someone buys one
Your use of “when” is interesting - suggests it’s not that occasional. Wonder what people are doing with these nowadays.
One interesting thing of living in the AI era is that every single one of these old apps (that people still use and never paid big money to upgrade because it it ain't broke don't fix it) can now be rebuilt for $5 in ten minutes if you really know what to do (YMMV)
If you don't know, you'd have to pay an expert big money to do that and the migration abandoned just like before, so another decade will go buy until AI does that without permission just because it found a possibility for optimization and it is instructed to do so without human approval, because governments
Absolutely. And run faster, cheaper and safer.
Just as long as you still have all the documentation and the spec was done right and updated as things changed over the whole lifetime of the system up to now. And a good test harness. And industry-specific knowledge.
You would think so, but most of the low hanging fruit has already been rewritten and replaced. The biggest hurdle is hardware lock in, ISA boards, hardware keys, or other hardware interfaces that can’t be easily connected to new systems. Never mind the problem of verification. That app that takes days to rewrite can annoyingly take months to verify.
> if you really know what to do
The so called "load-bearing gap" you pay experts for.
Do scanners count?
I'm pretty sure they moved on from them, but a while ago I did seasonal work for a company that rented out scanners for year end inventory (tax related, IIRC). The scanners had to be "cooked" (flashed) on racks, boxed, and sent. The flashing usually worked, but on occasion, the scanner would fail the checksum or something and drop you into a DOS prompt.
i know still some accountant who use a DOS program to do his work because he hates the fact modern UI has load times. slows him down too much o.O (his program is all keybinds).
i do not have exp myself but i found it a delightful reasoning. i told him modern PC could also do the same on modern OS (ncurses or ratatui or so maybe) but that if he ask a programmer to make an upgrade he'd need to find the right kind of guy. He had someone do it few times but they came back with web interfaces.
A Customer of mine has an 386 machine with an ISA interface board connected to an optical comparator instrument. I believe they also still have some MS-DOS-based computers connected to CNC machines, too.
I have two Customers who use DOS applications under DOSBox (a Clipper-based accounting package at one, and a Symantec Q&A-based application at another).
I don’t know the specifics of their systems but both of my local lumberyards here in SoCal still use DOS systems for their point of sale and inventory management, with slow dotmatrix printers and everything.
They simply have no reason to modernize. The one big benefit they’d have is an online inventory system but they make most of their money from negotiated bulk orders so there’s zero incentive.
I don't, but the abbey at the city I live has an ancient PC running a program from the 80s that controls the bell ringing.
I recall seeing it connected via a parallel port to a large board that then operates the switches/relays of the hammers.
I had to keep a disturbingly old machine with GPIB running until a few years back. They are still in operation in Australia (who we sold them to).
It was a test/repair bench for military aircraft avionics, with the only flying platform left using it being the legacy F/A-18 (A/B/C/D). Harris H-100 minicomputer hooked to an HP terminal and several full racks of GPIB test equipment with half of a rack of solenoids for switching between stimulus and response and which pin.
Not what you asked, but the question brought back memories.
I know of a DOS machine at a wool testing place that uses an ISA card based imager for measuring the micron and other things. They also have a supporting DOS machine with a dBase looking database on it. When I spoke to the owner he didn’t really understand the risks involved in running ancient hardware and in fact had a storage room with period PCs in it if the machines failed. He didn’t have much of a backup plan for the imaging hardware though.
My opinion is that if these guys are still running this gear, they already have a strategy in place (no matter how flawed) and would likely retire before throwing money at something they consider a non problem.
Good luck with the idea. It likely is a tiny niche market with customisation for every customer and given they are normally hardware constrained AI may not be much of a help.
All of them have no money as well.
My idea is basically a DOS runtime with hooks on interrupts and I/O that would allow for either migrating databases to modern formats, or converting I/O accesses for modern USB peripherals. Customization for every customer and AI not being much of a help are pluses for me... lack of money not so much.
I have been porting sections of the business software I mentioned earlier from QuickBASIC on DOS (nice to have the source code!) to Tcl on Linux using ChatGPT to automate the conversion. It gets it right the first try, almost every time.
The BASIC code is very simple by modern standards, and Tcl overlaps BASIC neatly, making conversion easy.
Unfortunately it's just a short-term stopgap to add some features management suddenly can't live without, that weren't practical to implement in BASIC/DOS. They'll replace the whole stack with a commercial product and support contract when they find something they like.
You can still get ISA slots on systems as new as Broadwell with an adapter.
We were talking about Visual FoxPro only a few days ago. [1] And one of those same companies in my comment [2] uses an old ordering system that is still in DOS and they boot up a VM to use it. They still have a printing machine ( not a normal printer ) for those order printing connected via a Parallel Port. I don't know the details of how that works though.
And by the end of all that, the order will be Faxed. Manually.
Did they tried to email it or automate it. Hell yes. Did it work? No.
[1] https://news.ycombinator.com/item?id=49811716
I do, for two clipper programs. But no extra machine, just a VM. Until not long ago, it was running without VM on 32 bit Windows 10.
We keep a FreeDOS machine at work for dialing in to configure and service paging terminals and using old two-way radio programming software.
Most of the stuff I've moved over to run in DOSbox but it's still in use occasionally for the software that's difficult to port over.
I did, for a while. Then, I used VMs. I haven't had a Windows VM for years.
If I were a serious gamer (I'm not), then I'd get a hardware rig, but otherwise, I've not needed it.
I had a doctor that kept an original NT machine, for many years, because his bespoke app worked on it.
I've seen some older retail stores use DOS machines, ones where the owners are 70+. They just don't have a reason to upgrade what they're used to.
A few years ago, I helped repair a broken USB port, on a USB-floppy emulator...
...which was so the operators could sneakernet their NC tape files into an IBM PC 5150 that thought it was talking to a floppy drive, of course...
...which hosted an ISA card that emulated the punch-tape reader that the NC lathe had originally been designed with.
The NC lathe hadn't missed a beat in 50+ years, why mess with what works?
Not recently but when I was a teenager (2001-ish) my supermarket was using a DOS computer to handle stock, print new labels when prices updated each day, etc.
Perhaps look into environmental services labs that do contract EPA work. I would not be surprised to see entire HP-Agilent(ISA-carded) - (think liquid chromotography analyzers) fleets that would match what you're looking for.
Still have a liquid scintillation counter running on DOS with raw data printed on a dot matrix
You're trying to create and sell a version of dos box? These companies have not replaced their computers because to the user there is no separation between the computer and the machine it controls. So the computer will be updated along with the equipment, which is not connected to the internet or a network anyways so it's not a cybersecurity risk. For the user, if it runs fine, there's no need to update just the computer, and the computer is updated along with the machine
In other words, these are embedded systems.
Not exactly, freely available solutions are already good enough for people who just want to run their apps on modern hardware, and have been for a long time. What I was thinking about writing is a tool that eases migration by using interrupt and I/O hooks to detect common patterns and, say, builds a modern SQLite database as the business uses their dBase program. Whether anyone actually needs this, I don't know.
There's a few car dealerships that still use databases from the 80s. As for why they haven't migrated, I'd guess must be reasons I'd assume at this point it's a choice they have made
It's not DOS but my private school still uses an IBM 5100 to do payroll and record attendance.
What's the reason for not using a more modern MIS?
(I can certainly understand why organisations with niche requirements might hang on to old systems for as long as they continue to work, but I'd have thought that school payroll and attendance were much too commonplace to fall into that category)
I'm kidding. BRO, no school would be running the first (imo) commercial desktop still in the year 2026. The machine I cited was released in 1976!
This is probably a wrong audience. Most people here are very technical - the kind that would not put up with tiptoeing around a sacred computer from decades ago running some magic software from the DOS era. This crowd will aggressively move it to modernized architecture or at least emulate it using a more modern platform.
The company I worked at in the late nineties, used Managing Your Money (MYM) for bookkeeping - initially on a dedicated DOS machine. My colleague set up a DOSEMU-based emulator and moved the software to it (bye-bye, dedicated DOS machine!). Later, he re-wrote it in STk (Scheme w/ Tk bindings, running natively in Linux) as ageneral-purpose transaction storage. ¯\_(ツ)_/¯
This is not always possible when the systems are validated and regulated.
Yeah. Sometimes old hardware forces its hand and causes a big PITA.
I had a job working on VAX/VMS around 2009, but it was actually an emulator running on a PC because they just couldn't find any hardware to fix the VAX. At some point before I joined they had to get a ton of government approvals to use the emulator.
Fortunately they had a project to modernize the whole stack. And it was going to Solaris :-/
Until recently several customers have an industrial machine that boots DOS. I think the oldest effort we deal with is Win XP now. I do have a C>64 at home ...
In a world that requires security audits etc here's some notes that might help:
Scenarios: You have a device that only talks NetBIOS or even NetBEUI for file sharing. You need to get files to and from it and your CAD workstations run Windows 11 and anything less than SMB 3, with signing and sealing and all that jazz is laughable.
... Devices requiring ftp, telnet, serial over carrier pidgeon
Solution options:
Networking: The first job is to segregate your industrial gear from your general network and you might want to create two or more industrial gear and internet of things type networks with differing access policies. Start with one for now, you can improve it later.
You use 801.1Q VLANS logically or physically (why the hell did I want to write literally?) separate via separate switches.
Please do not skimp on this step. Buy a network bod to sort it out or put your IT bod on a course or hopefully they will have labbed this up at home already!
Comms: Samba is bloody wonderful and speaks everything that MS has ever done and deprecated. Even Samba does need to be asked to start speaking some of those dialects again but it will.
Put a Linux (other Unices are available) box on the same VLAN as your machines. You will need to set quite a few options relating to the minimum version for its server version. When talking to the machines, Samba is the server and when talking to your file server, Samba is the client.
Create a data share on a Windows box on the corp LAN. For example it might be on a file server for multiple access or just a single CAD PC.
Get the Samba box to mount that share to itself at say /srv/data_share_mc1
Now get the Samba box to create a share from /srv/data_share_mc1 which it calls DAT_MC1 (8.3 chars for NetBIOS)
With luck, you should be able to map DAT_MC1 on your machine.
Few years ago I was working for a F500 company and during an assessment we discovered couple of OS/2 Warp machines that were controlling CNC equipment
I've stumbled upon DOS on the LCD screen of the otherwise seemingly modern gas station.
I know of some laboratories where they have a host of old machines for software to support some critical instruments that only runs off a dongle or an ISA card and it still does accurate work.
Not DOS, but my company has a Windows XP machine with a dialup connection that I have to keep running for obtuse legal reasons.
If it goes down, the company gets fined. It happened in the gap between my predecessor leaving, and me being hired.
Does it have to be a XP box or is just keeping a serial dialup connection working the key?
Is it a physical machine and not a VM? I already had issues with getting XP compatible spare parts years ago. Had to buy used parts that were already pretty old (so high risk of failing again).
It's a physical machine.
Fortunately, I don't have to deal with swapping parts. There's an outside contractor on call 24/7 for that.
The last time I saw it was about two years ago when I taped yet another piece of paper to the filing cabinet it sits on reading "DO NOT TURN OFF UNDER PENALTY OF LAW!"
Someone keeps taking the paper down.
1 reply →
I virtualized a bunch and they seem to work well. At one point, I even had Ultima running on one. =D
Not now, but few years ago I worked on a toll that runs DOS because… yes.
Also worked ~2014 DOS slot machines - in next year we ported to Linux
It’s gone now, but we used to have a random OS/2 box well past the point where that was extremely weird.
What's weird about Warp?
Like most people who bought into the OS/2 hype, I bailed after 1.3.
The weird part was still using it into the 2010s.
I used to be doing that, but this was a bit over a decade ago at this point. There was an SBC with DOS on a chip that connected to Novell Netware. I set a few variables in autoexec.bat, let it connect to the network drive, then had it run a script from there to admin them / do backups (by which I mean xcopy the C: drive).
These were for some industrial CNC machines with a poorly version-controlled app (send us the single-file .c for the program, we'll make changes & recompile for you, then send back a new .c and .exe file) that would just malloc() all the memory on the machine for a million element array instead of tracking the 1k or so unit IDs a more efficient way. They were six digits long, it worked... mostly.
I kept a 3.5 floppy with the contents of sys a: that I used to run sys c: after restoring a machine from backup in my office.
EDIT: oh, and of course by "restoring from backup" I mean reversing the xcopy to copy everything from the network drive's copy back to C:
EDIT2: My memory is going, they didn't use malloc() at all, anywhere in the programs, they just had a static million element array.
Not current and not DOS... but years ago, around 2011, I helped set up a series of QEMU emulators running DEC Ultrix and MIPS Unix software that called out and retrieved data from weather stations only reachable by modem. The software had been written years prior in C and nobody had the source.
The box running the QEMU emulators had a bunch of old US Robotics modems hanging off it via USB serial dongles.
>Anything with a parallel port dongle?
https://news.ycombinator.com/item?id=46849567
[dead]
[dead]
[flagged]
[dead]
[flagged]
In the AI era, can’t we just release a new version of DOS?
FreeDOS exists, no need for AI:
https://freedos.org/