News Categories |
Main /
History of DRDOS
On this page… (hide) 1. History of Dr-DOS
2. Former owners of Dr-DOS
3. DOS History (Timeline)Jim Hall - founder of FreeDOS - has collected all history infos about DOS! Timeline of DOS (updated 10/2005) doshist.txt? 4. DR DOSDR DOS began with version 3.3 to compete with MS/PC DOS 3.3. Versions were created as 3.3x and 3.4x. Digital Research sold DR DOS as retail, unlike Microsoft/IBM which only sold DOS bundled. After MS/PC DOS 4 was met with unpopularity, Digital Research began working on ways to free conventional memory, and released their next version as DR DOS 5 in May 1990. New commands allowed the kernel, shell, device drivers, and TSRs to be loaded into upper and high memory, freeing up a great deal of conventional memory (the maximum achievable was said to be 627 KiB). Also, video memory could be mapped as regular memory, allow for more conventional memory. Microsoft began aggressively marketing MS DOS 5, which did not yet exist, to try to curb the market. By the time MS/PC DOS 5 was released in 1991 (a near carbon copy of the DR DOS high/upper memory model), DR DOS 6 was being released. An intentional bug by Microsoft, called the “AARD code”, would cause the installer of Windows 3.1 beta to crash, frightening new users into thinking that DR DOS was damaging their system. As such, DR DOS began to lose ground to Microsoft and IBM. Subsequently, Digital Research sold DR DOS to Novell, who released it as Novell DOS 7. By this time, however, Windows 95 was out, and Novell could not gain much ground. The system was sold to Caldera, who released it as the open-source OpenDOS 7.01. A beta, OpenDOS 7.02, was released after that; by the time it hit the market, it was branded Dr-OpenDOS 7.02. Caldera later closed the source and released it as Caldera DR DOS 7.02. Enhanced DR DOS is based off of the 7.01 source. Caldera transferred it to a branch company, Caldera Thin Clients, who later became Lineo. Still keeping the Caldera DR DOS name, version 7.03 was released, the current stable build. A DR DOS 7.04/7.05 exists; it is only the kernel and COMMAND.COM, and was designed for Seagate SeaTools. It featured FAT32 support, but had a bug that stopped files from being renamed, causing it to receive little use. Lineo was sold to MetroWerks, and some former employees took DR DOS with them, forming a new company, DeviceLogics. DeviceLogics released DR DOS 8 on March 30, 2004, featuring FAT32 support. (It was branded “DeviceLogics DR DOS 8.0″). Later that year, DR DOS was spun off into DR DOS Inc., and this company released DR DOS 8.1 in October of 2005. Within days, however, it was discovered that 8.0 and even more 8.1 included pieces of Enhanced DR DOS and FreeDOS, and violates FreeDOS GPL and several other copyrights and licenses. After complaints from various people, including Udo Kuhnt and Jim Hall, DR DOS 8.x was pulled, and DR DOS Inc. sells Caldera DR DOS 7.03 again - see http://www.freedos.org/freedos/news/press/2005-drdos.html . While they owned it, Caldera sued Microsoft for unfair business practices regarding DR DOS; the lawsuit was settled. DR DOS Inc. also offers to sell DR DOS outright; the price is $25′000. 5. Timeline since 1997
6. Stories6.1 Matthias PaulLineo DrDOS (2000) is the successor of Novell DOS (1993–1996), which derived from Digital Research’s DR DOS (1988–1993), and DR Multiuser DOS, which in turn were based on the DOS Plus (1985–1987), CP/M-Plus, DR Concurrent DOS (since 1986), and DR Concurrent CP/M (CCP/M) (since 1982) families, directly going back to the legendary CP/M (Control Program for Microprocessors) (1973) and the multitasking/multiuser version MP/M (Multiprogramming Monitor for Microcomputers). If you have a look, you can see the copyright messages going back to 1976, even though DRI’s DOS family actually started out of DR Concurrent CP/M-86, which had some DOS emulation, and the multitasking DOS Plus for the Amstrad/Schneider PC1512, which had a Concurrent CP/M-86 emulation layer. However, some of the code or technology might even go back to the original CP/M code base, that is back to the days the whole microcomputer area started with Intel’s 8080 in 1973… Other members of this OS family tree have been for example DR PalmDOS (1993), DR Concurrent PC-DOS, DR Concurrent DOS/386, DR Concurrent DOS/XM, CP/M-80, and CP/M-86, even a CP/M-68K and a graphically system extension named GSX. The list goes further with the networking software DR NET, the graphically multitasking shell GEM (Graphical Environment Manager) for both, Intel and Motorola platforms (e.g. on Atari computers with their GEMDOS), and last but not least the modular protected mode real-time operating system DR FlexOS, including its subsystems X/GEM and FlexNet. Currently shipping products such as Concurrent Control’s CCI Multiuser DOS, Intelligent Micro Software’s IMS REAL/32, and Integrated Systems’ FlexOS are now on independent development paths, but also inherited the DR Multiuser DOS respectively DR FlexOS source code. “Ok, and what about MS-DOS/PC-DOS, then?” Microsoft’s MS-DOS alias IBM’s PC-DOS 1.0 (1981) (optimized for the IBM PC) were originally based on 86-DOS, licensed for $50.000 from Seattle Computer Products’ QDOS (1980), a disk-operating-system written in two months (Technically, this is incorrect. Tim Paterson claims it took him 2 man-months over a 4 month period, look at http://patersontech.com/Dos/Micronews/paterson04_10_98.htm), and basically representing a mixture of ‘new’ ideas (especially in media access) and a 16 bit port from CP/M-80 to the 8088/8086 architecture. Of course, much has changed from these ancient times, but the roots are still visible. So after all, although OpenDOS (Caldera: 1997–1999) also learnt from MS-DOS to become most compatible (e.g. taking its own HIDEVICE= and MS-DOS’ play on words DEVICEHIGH=, etc. ;-) ) and emulates MS-DOS’ APIs, it truely is not a clone, as real as DOS can be… To give some foundation to these arguments, here are some of the single-user series’ unique features at that times: DR DOS 3.4x (1988/1989) had the CONFIG.SYS INSTALL= command, and CONFIG.SYS queries by the ?= command, a full screen editor, help screens with /?, XDIR and XDEL, and already was ROMable and supported harddisks up to 512 MiB; DR DOS 5.0 (1990) introduced all these DOS load-high options moving parts of the kernel and drivers into UMBs and the HMA (like MEMMAX, HIDOS=ON, HIDEVICE=, HIINSTALL=, HILOAD, and the /MH parameter), and the graphically DOS-shell (ViewMAX); DR DOS 6.0 (1991) came with data compression (SUPERSTOR at that times), delete tracking software (DELWATCH), NetWare Lite support, a DOS prompt task switcher (TASKMAX), a link facility (FILELINK), CONFIG.SYS SET= and boot menus; been followed by Novell DOS 7 (1993) introducing real multitasking, Personal NetWare, a DOS DPMI server (without a need for MS Windows), Protected Mode DOS drivers (with DPMS), an adaptive cache which could lean memory even to DOS programs, LASTDRIVE=32 etc. Some of the features, like password protection or list files still have not been adapted by the competitors. Well, that’s history and we should now get back to the present - because the story goes on (even if much of the being ahead of time has been lost because of Novell’s trashing the product for two years)… 6.2 Mr. M.DRDOS was originally developed in the 1980s at Charnam Park, an industrial estate in Hungerford, in the far east of Wiltshire, in England (‘’Not Quite True it was in Station Road Hungerford in West Berkshire, didnt move to charnham park till much later’‘ - remark from anonymous poster of our guestbook). It stayed there until about 1996 when Novell dropped DRDOS. In January 1997, under Caldera, development began again in a barn near Andover, in Hampshire, by a group of ex-Novell engineers. OpenDOS 7.01 was released, a quick hack based on the version of Novell DOS which Novell had given them. Novell had unfortunately mislaid the source code of the latest version, which would not be found for nearly a year. In May 1997, I was looking for a company to do a placement year with as the 3rd year of a sandwich course. I was running OpenDOS 7.01 at the time, and I’d been using DRDOS 6.01 since 1992. At the time, I believed that the OpenDOS work was being done in the US, so I didn’t believe my eyes when I found a note on the University noticeboard looking for a Software Engineering undergraduate to work on DRDOS in Andover. I applied for that post and worked for Caldera UK as a contractor from July 1997 until August 1998. I did a little kernel work and have a fair few bugs to my name (The Year 2000-PCI problem of the 7.02 beta, the DR-Mouse fiasco). I also rewrote the guts of FDISK and added support for other partition types. I’m attaching a photo of the inside of the barn where DRDOS and Webspyder were being developed, sometime in early 1998. This photo was taken by Matthias Paul. I’ve concealed all the faces except those of myself and Kevin, who can’t really be seen well enough to identify “Click here” to see the photo! I also have some photos of the outside of the barn but I haven’t scanned them in yet. 7. Microsoft and Caldera - dispelling the myths (2000)See also: FAQ (about DOS, DR-DOS and EDR-DOS) |
Recent Comments