Founder of FreeDOS recounts the story so far, and the future The last mainstream DOS-based OS was Windows ME, which went out of support 20 years ago. And yet, thanks to free software, DOS lives on. We spoke to FreeDOS founder Jim Hall about how the project started and how it's progressing. Version 1.3 of FreeDOS came out in February, and an interim update is on its way. It has two …
An interesting interview. Great to hear more about projects such as these.
I am an avid FreeDOS user and can't fault it. With the updated OpenWatcom, DESQview and patched Vim (to fix lost keys on Qemu), it is probably my favorite environment.
I would love to see more PCs distributed with it but running on top of a very thin hypervisor to "smooth" out the rough edges when running on current hardware. I personally feel this could even make for a decent product.
[Author -- hang on, interviewer here]
Yes it supports FAT32.
No, DOS never supported file sharing on its own. It was always an optional extra. That's why Novell invented Netware Lite, then Personal Netware, then bundled it with Novell DOS 7. It's why SAGE sold MainLAN, and why in the USA, LANtastic and the $25 Network made livings.
Yes, one MS client could share too, but it took a lot of conventional memory so it wasn't much use, and that's the main reason it was removed and moved into Windows for Workgroups.
But it was never a standard function, and it was not connected with FAT16, which appeared with MS-DOS 3.0 for the PC-XT and PC-AT which had hard disks.
https://en.wikipedia.org/wiki/File_Allocation_Table#FAT16
You may be confusing the DOS redirector, which appeared in MS-DOS 3.1 and was mainly used by MSCDEX for CD drive support.
https://en.wikipedia.org/wiki/MSCDEX
FreeDOS can run the MS client, and the IBM client, and both Novell P2P clients, AFAIK. However none of them natively talk TCP/IP so they are not much use today.
You can add TCP/IP to them but the result is a lot of memory usage -- I have tried it. Performance varies widely, too, as the OS/2 Museum measured:
https://www.os2museum.com/wp/dos-smb-client-performance/
"Share" was a DOS command that was part of MS-DOS from v 3.0 or 3.1. https://en.wikipedia.org/wiki/Share_(command)
It allocated memory to allow you to share files between different processes. Since DOS was mostly a single-process operating system, the reason for allowing files to be shared between different processes was to allow network sharing.
Networking was not a native part of DOS 3, but SHARE was.
The resident parts of DOS that SHARE configured were moved into Windows. This had the effect of preventing the use of network sharing without Windows.
Novell introduced peer-to-peer networking in 1981. In response, MS built SHARE into DOS 3, and introduced MS-NET.
Yes, you needed to add support for whatever network protocol you wanted to support. The MS TCP/IP client for DOS was free after WinNT 3.5. From DOS7, the MS TCP/IP client did not support peer-to-peer DOS networks, because the necessary SHARE support was moved into Windows.
HAVE ONE ON ME ===============>
As one of the many poor unfortunates out there who's job it is to recode legacy processes and apps for a living right now these projects are a god send. They allow us to put certain items on life support within containers that can then be run on bang up to date modern O/S systems. Thereby keeping them running long enough to complete their jobs while we recode and/or dismantle the surrounding kit, without all the panic that something horrendous is going to happen at any moment.
Yeah the cloud is alive and well but with PC tech now over 40 years old there's an awful lot of legacy junk out there that is ultra critical but hidden away, these amazing volunteers help the world keep running and they have no idea how valuable they are in playing their part in almost stopping parts of civilization collapsing!
DR-DOS is still around. I have fixed the non-working boot disk images from Archive OS:
https://archiveos.org/drdos/
You can download working boot floppy disk images from my blog:
https://liam-on-linux.livejournal.com/58013.html
I also have some ready-installed Virtualbox virtual HDD images if installing your own is too daunting:
https://liam-on-linux.livejournal.com/79015.html
Noooo, dear Vultures, please don't fall into the quagmire that too much of the interweb has sunk into of late, of using hundreds of frames of video (or hundreds of seconds of audio) when just hundreds of plain old written words would honestly do perfectly well, if not better.
Video or audio interviews invariably end up with too much waffle as the interviewee tries to think how to answer the question while also starting to say something at the same time ("Yeah", "You know", "I'll just kind of give you a little of how that came about", etc, and that's just in the first answer; no personal criticism of the interviewee intended, it's just the nature of unedited live/realtime interviews), not to mention taking ten times longer to listen to than just reading the (appropriately tidied and edited) text of the salient points of the conversation.
These automated transcriptions are just as bad, because they still retain all of that waffle (and the grammatical quirks that we might use in a flowing babble of speech which don't read very comprehensibly when actually read back as text, along with the lack of phrase and paragraph breaks, etc).
I would have been vaguely interested enough to have read a properly written/edited article about this particular topic, but the transcript is rather unreadable, I'm afraid, and (and I'm sure I'm not alone here) I have better things to do with my time than spend it listening to an audio/video interview when there isn't really a need for one. :-(
Before you know it, you'll end up pleading "Hit those Like and Subscribe buttons!" with a desperate rictus grin frozen on your faces, and then there really will be no option but to search for the current location of the old carpet offcuts and the quicklime…
The only reason for "fluff" videos (I was going to forgo the adjective, but any video purporting to enhance what I expect to be a well-written article is nothing but "fluff") because it reduces the work-load of the author...and immensely, and absolutely increases the work-load of the reader.
In a communication a while back, a very-well-respected, highly talented and hard working technical (and -otherwise) author has summed up the sorry state of technical "articles" on the internet with this extremely cogent synopsis (paraphrase): "...it is impossible for anyone who knows their subject well to make a living writing well-researched and well-written articles. The internet has become nothing but a cesspool of (and THIS is a direct quote) crap videos..."
After it became obvious that a once-very-good source of written information and instruction (title perhaps has the word "day" in it) was turning into nothing more than a 250-word source of fluff obviously lifted from somewhere else, but ENHANCED with a "crap video", led into with the words, "VIDEO AFTER THE BREAK", I posted a comment to the effect that I didn't get my hard-earned degrees by reading comic books, and was certain I wasn't going to learn anything of substance from all their "...videos after the break...".
The comment was removed after two hours.
...same exact reason I refuse to "read" or frequent a "technical" venue whose title includes the initialism "...vblog..." ("video blog". perhaps? Of course.) One doesn't "read" this source, at all, for the simple reason that the site's "author" doesn't write anything of any substance; instead one is subjected to having to suffer through watching nothing but the videos he makes.
Kudos to the author for cashing in on the "new normal", the 'Facebook' mentality of "Don't Make Me Work for ANYTHING".
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
