Why has no large corporation tried to compete with Microsoft in the OS field?

It's a UNIX even if it's a weird one.

Only with Windows can your company buy cheap-ass Epsin X9001 fax/modem/scanners with drivers that only come on cd disks for win98/xp/7

Back in the day IBM partnered with Microsoft. You bought an IBM machine with Windows on it.

The measure of an OS's success is it's software support. Unless IBM started pouring exclusive application development into this OS, applications that businesses/consumers want, no one will use it.

OS/2 was conceived originally as a joint IBM/Microsoft venture with OS/2 to be the professional OS for servers and workstations while Windows was to be the consumer OS. However, the first version of OS/2 was 16-bit and only ran in text mode and IBM didn't have a real 32-bit GUI-based one until 1992 after which Windows handily stomped them.

Xerox should have, but they are cucks

>IBM fucked themselves back in the PS/2 days even
IBM's PC division which account for a really small part of IBM fucked themselves by releasing the IBM PC-XT without adding custom chips and the likes that would make clone manufacturers' life difficult. IBM shot itself in the foot by making a platform so easily clonable and not doing anything to stop the cloning spree like apple did with the apple IIe before it was too late.
>It's a UNIX even if it's a weird one.
Nope it's your bog standard AT&T UNIX System V SVR4 ported on Power architecture and with a few IBM additions.

>The collaboration between IBM and Microsoft unravelled in 1990, between the releases of Windows 3.0 and OS/2 1.3. During this time, Windows 3.0 became a tremendous success, selling millions of copies in its first year.[17] Much of its success was because Windows 3.0 (along with MS-DOS) was bundled with most new computers.[18] OS/2, on the other hand, was available only as an additional stand-alone software package. In addition, OS/2 lacked device drivers for many common devices such as printers, particularly non-IBM hardware.[19] Windows, on the other hand, supported a much larger variety of hardware. The increasing popularity of Windows prompted Microsoft to shift its development focus from cooperating on OS/2 with IBM to building its own business based on Windows.[20]

>Several technical and practical reasons contributed to this breakup.

>The two companies had significant differences in culture and vision. Microsoft favored the open hardware system approach that contributed to its success on the PC; IBM sought to use OS/2 to drive sales of its own hardware, including systems that could not support the features Microsoft wanted. Microsoft programmers also became frustrated with IBM's bureaucracy and its use of lines of code to measure programmer productivity.[21] IBM developers complained about the terseness and lack of comments in Microsoft's code, while Microsoft developers complained that IBM's code was bloated.[22]

>the open hardware system approach that contributed to its success on the PC
The success of the PC mainly comes from the shiny IBM logo on the case.

Microsoft originally convinced IBM to use an open architecture for the PC because they wanted to sell operating systems to as many customers as possible.

This same gulf in understanding happened later on with OS/2. Microsoft envisioned selling to any licensee who was willing to pay while IBM only cared about supporting their own hardware, so whether OS/2 worked on a Compaq machine or not was irrelevant to them.

Actually it wasn't IBM that made the PC architecture successful or even the early clones like Compaq and Zenith, it was the event of low-cost Asian PC clones in the late 80s that allowed the architecture to take over the consumer market. Meanwhile, the event of 386 PCs killed off any viable competition in the professional market.

The IBM PC wasn't the best or most sophisticated hardware around; most x86 machines of the period were more technically advanced. But it was a nice, well-put together machine that didn't have proprietary hardware and it incorporated some traditional home computer features while if you looked at a lot of business computers in the early 80s, they just had a generic text display and no sound capabilities.

OS/2 wasn't terrible. If microsoft didn't ditch it and go hard into NT instead it probably would've become pretty popular.

Attached: os2w4[1].jpg (1024x768, 777K)

>But it was a nice, well-put together machine that didn't have proprietary hardware and it incorporated some traditional home computer features while if you looked at a lot of business computers in the early 80s, they just had a generic text display and no sound capabilities.

Mmmmuh gaymes.

Attached: soy wojak.png (632x756, 26K)

It was a much better OS than anything pre-Windows 2000, but it required high end hardware and of course was geared towards IBM machines.

It is, the IBM PC's success comes from the professional market. Taiwanese clones wouldn't have been successful at all if they weren't just "doing the same thing that the ones at the office", they were merely riding on the initial success of the platform.
The IBM PC's architecture was only documented to allow 3rd-party expansion hardware. Every single clone used a BIOS which was barely legal (they were reverse-engineered illegally but thanks to a loop-hole in the law they got away with it).
Yes it's actually pretty good. Too bad only 2 special releases actually support multi-processor machines.

Early PC was pretty shitty at muh games, compare what C64 or

*or Atari 400 could do, much better sound and graphics support than CGA and PC speaker.

They didn't reverse engineer anything and none of IBM's copyrighted code was used. BIOS clones were made simply by creating a workalike based on all published info about the BIOS and its functions.

That this was possible was because the PC used vectored interrupts so BIOS clones could easily be made without pointing to the same ROM addresses as the IBM BIOS, while on the Apple II, the ROM addresses were absolute and thus it was harder to get away with copying it.

It was about comparable to the Apple II, however the Apple II had the advantage of years of experienced programmers.

youtube.com/watch?v=h5Q9loiBJfs

youtube.com/watch?v=I9W5rmvM2ns

Note how the Apple II version has nice, smooth 60 fps animation while the PC is a choppy, flickery piece of shit.

But the Phoenix BIOS that allowed the first PC clones to hit the market without the manufacturer being sued by IBM WAS the result of the IBM BIOS reverse-engineering.

In fact this is not true Compaq had their own BIOS for the Compaq Portable which they developed in-house. IBM initially filed a lawsuit, but Compaq were prepared for this and the case was quickly dropped.

en.wikipedia.org/wiki/Phoenix_Technologies

Phoenix introduced their BIOS in 1984, more than a year after the Compaq Portable went on sale and they were the first company to offer an "off the shelf" BIOS to any clone manufacturer that didn't want to develop one themselves. According to the article, some companies did rip off IBM's code and got sued for it, but Phoenix never did this and their BIOSes were always clean room implementations.

Compaq had their own BIOS and Zenith and HP and Tandy (sometimes they also used Phoenix BIOSes depending on the machine) and probably some others too I forget.

Used to have a Presario 6000 and it would display a huge, giant Compaq logo on the screen on power up.

Compaq, HP, Zenith, Tandy, Epson, Acer, DTK, AST, IBM, NCR, Olivetti, Osicom, Tandon, Toshiba, and Wang had their own BIOSes. This is according to "Secrets of Windows 3.1".

"Ha ha ha ha, Wang!"

Attached: maxresdefault.jpg (1920x1080, 109K)

It amazes me to think how many PC manufacturers there once were compared to today.

They laid every single bit of the hardware bare and printed the full pre-assembled BIOS listing on the back of their technical manual. Most manufacturers just copied it instead of reimplementing it, but they hardly had to do any reverse engineering at all.
classiccomputers.info/down/IBM/IBM_PC_5150/IBM_5150_Technical_Reference_6025005_AUG81.pdf

Correct. They printed BIOS listings up to the AT at least (the PS/2 line did not have BIOS listings) but Compaq and other clone makers made it explicit policy to not look at any of the BIOS code so as to not intentionally or unintentionally steal it.

>they're more successful than ever

holy fuck you're deluded

>Most manufacturers just copied it instead of reimplementing it
Copying that BIOS listing to assemble your own would result in your ass getting sued by IBM.
IBM communicated these informations for hardware expansion and software to be made for the IBM PC, not for other companies to clones their computer.

They may have not intended for it to happen, but you can bet Bill Gates sure wanted to because every PC clone produced trebled the size of his Cayman Islands bank account.

Meanwhile Apple even went so far as to sue DRI for trying to put GEM on the Mac.

The Mac System ROMs were an extension of the OS and very tightly integrated with it. They never documented how to bring the system up from the bare metal and a huge mess would result from having to support multiple OSes. Also Apple wanted to ensure a consistent user experience and compatibility with different Mac models, which would be completely ruined by third party OSes.

Anyway, the PS/2 line tried to force everything back into proprietary hardware and software, but it backfired and cost IBM millions.

MS barely makes money from OS sales

I wouldn’t ve surprised to see them drop support entirely for consumer windows

>However, the first version of OS/2 was 16-bit and only ran in text mode and

In the purest sense of the word. It was 100% an OS designed for 286 machines and had no 386 support at all while Windows from 2.x onward could use various 32-bit features even though the OS kernel was 16-bit until 95.

In sectors that are barely profitable and that can be done by everyone the biggest company wins.
There is no money in better OSes. Also, network effect.

z/OS and z/VSE aren't UNIX/linux/DOS-based (well, not DOS-based in the MS-DOS sense) and do have their own segment that microsoft can't have access to even if they wanted to. But like said, money isn't in the OS market, but the aditional software and services.

"Secrets of Windows 3.1" lists at least 160 different PC manufacturers/brands and that's just for machines sold in North America, not counting European or Asian ones.

It's actually incredible to think just how many PC brands there were in the early 90s considering that there were not as many computers around back then compared with today.

Some of those industrial machines and not all consumer ones. For example, Tangent is still around and makes industrial/medical PCs.

I honestly don't understand why Microsoft still makes almost all their profits out of Windows and Office, when they have had the oppertunity to expand into so many markets.

And why the fuck is there no proper Microsoft alternative to Adobe products?

>I honestly don't understand why Microsoft still makes almost all their profits out of Windows and Office, when they have had the oppertunity to expand into so many markets.
They're lazy and run purely on inertia since Bill Gates stepped down in 2000.

Attached: wang.jpg (640x480, 63K)

Go to any super market and look what brand the cash register is. There's a good chance it's IBM

Wow is summer here already?

Uhm sweety it's July. Of course it's summer.

Most of it now comes from azure and office 365.
There are no markets for them to expand to that isn't already taken up by dominant competitors.

They did. It was called OS/2 and it was great for its time

Attached: 1200px-Os2logo.svg.png (1200x1200, 168K)

wait for Fuchsia

FWIW, OS/2 had multi-core support long before any wangblows release, and had some fantastic features that windows didn't add for 20+ years, such as multiple desktops

Only a few versions of OS/2 actually supported SMP (and not "multi-core" like you're saying), like OS/2 Warp server 4 SMP and OS/2 Warp server advanced. OS/2 Warp 3 and 4 in their client version don't support SMP though.

templeos

>Nope it's your bog standard AT&T UNIX System V SVR4 ported on Power architecture and with a few IBM additions.
Have you ever worked with AIX? It's a lot more than just "few" IBM additions, to the point some people wonder if it should be even called UNIX.
It's a weird crossover between SysV, BSD, and IBM's mainframe OSs.

>Have you ever worked with AIX?
Yes. My first mission at my current job was about the maintenance of a huge application running on RS/6000 servers. It featured the ksh shell you'd find on a post SVR3 System V install. Switched back to z/OS after that, but I did use it for a a bunch of months.
>It's a weird crossover between SysV, BSD, and IBM's mainframe OSs
System V already included stuff from BSD starting with SVR3, and no, AIX has almost nothing in common with either z/OS or z/VSE.

>they're more successful than ever

>Switched back to z/OS after that
but this invalidates all your arguments because you're clearly brainwashed by IBM :^)

>AIX has almost nothing in common with either z/OS or z/VSE
It draws concepts from them, like ODM, binary logging, and smit. Those don't come from UNIX, but are original IBM stuff.
(OK, I know that other Unices have tools like smit, but smit takes it to the next level along with the indecipherable management commands.)

>but this invalidates all your arguments because you're clearly brainwashed by IBM :^)
Well if I was brainwashed by IBM I'd be an eclipse lover, seeing how many stuff they shit out that are base on eclipse (hell, even ibm notes is now based on eclipse).
Seriously, their eclipse-based rational developper suite (cross-developpement platform for z/OS, AIX and OS/400 running on windows) is pure garbage -- it's slow as hell, breaks up all the time (it's the first time I've seen software displaying an error message because an error occured while displaying an error message) and mess up with your work (workspace corruption happened quite a few time during those months of work, downloading those ~8GB of cobol sources multiple times sure is tedious). During the time I was working on that application, I saw that most of my older co-workers used the pacbase server when they needed to search for the cause of a problem. I didn't have pacbase accesses though so I looked at the sources directly on the AIX dev server (using a vt420 terminal emulator, alway found it fun to end up emulating at the office what I already own at home). Unfortunately, all the modifications had to be done on the rational suite.

If such demand existed, Linux, would have more than single digit percentage of the user market.

>eclipse-based rational
I feel like those two words should never be together in one sentence.
Last time I used something eclipse-based it was a native client of HP's ticketing tool. It somehow managed to be slower than the web interface.

Nobody remembers this meme
>call it OS/2 connect
>even though network support was completely missing

Attached: bigdesk.gif (1024x768, 62K)

Are you talking about quality center? ALM? Some people at work are forced to use those too. Hopefuly I alway managed to dodge it.

PS/2 was a fascinating business mess. IBM never had the perception of the computer becoming a commodity and laughed at the clones while producing a system that was superior but did not care about compatibility.

It seems that compatibility (and price) always wins over better OS design and hardware, at least with proprietary software. IBM always had the elitist vision of the computer being an expensive backroom cabinet with them being the key provider of solutions. They rejected a multicorp dominant, distributed computing world until it became reality. At that point, they were just another mediocre player in that area. If IBM did not try to build their typical cathedral with PS/2 it would be a major platform today.

Service Manager 9

>ibm is a dead company
t. retard

AIX, mainframes.

IBM is also focused more on Software more these days, but you probably don't hear of it since most of you are NEETs.

>>they're more successful than ever
They literally reported their first quarter of growth in six years in January of this year.

Haven't seen that one yet.
Don't forget about OS/400, it still has a small but solid user-base.

There will be a OS written in Js before the next decade is out.

CVS senior IT analyst here, this is correct. almost all CVS POS registers are IBM or joint IBM-Toshiba systems. our mainframes are proprietary Toshiba/IBM SMART systems

An OS is worthless unless you get driver support from all major manufacturers

Can you come up with a OS that's just as simple to use as windows and still having some advantage/difference over it?