The Software That Failed To Compete With Windows 347
harrymcc writes "When Microsoft shipped Windows 1.0 back in November 1985 — it turned 25 on Saturday — it wasn't clear that its much-delayed windowing add-on for DOS was going to succeed. After all, it was a late arrival to a market that was already teeming with ambitious competitors. A quarter-century later, it's worth remembering the early Windows rivals that didn't make it: Visi On, Top View, GEM, DESQview, and more."
OS/2 (Score:5, Insightful)
They left out the most viable competitor.
OS/2 (Score:3, Insightful)
Re:So ... (Score:2, Insightful)
Both.
The "woe" part being, that no replacement for microsoft will succeed, unless it has the same blinding ambition and greed that microsoft had, and the others lacked. This was proven by the fact that the other competitors were "nice", but did not have those qualities, so were dominated.
Re:OS/2 (Score:5, Insightful)
Given that this is a list of "Windows' Failed Rivals", OS/2 rightfully isn't on that list... IBM continued to release new OS/2 versions for nearly a decade after its initial release.
SideKick (by Intuit) (Score:2, Insightful)
This was perhaps the Enabler for Windows. It addressed the primary multi-tasking via
a terminate-and-stay-resident pop-up that had a calculator, todo list, and the like.
By solving this problem for Word Perfect, Lotus and DB3 users, it delayed the
adoption of windowing environments for another 2-3 years till Windows 3.0
Hard to forget hell. (Score:5, Insightful)
"Twenty-five years and two days later, it’s not just hard to remember an era in which Windows wasn’t everywhere"
Bullshit - As a C64 and Atari ST veteran, twenty-five years later it's painful to remember the extraordinary effort it took to lose to windows. I had better graphics playing Neuromancer on the C64 than windows managed for a decade, and let's not even talk about comparing Star Flight on the ST vs the DOS version.
Jack Tramiel should be strung up for crimes against computing.
{sigh} - Pug
Re:So ... (Score:2, Insightful)
A windows replacement can succeed if it's pushed out by Microsoft. The factors at play were nothing to do with quality of the product. Microsoft "skillfully" pushed its stuff out in such a way that no one else could play in the same market for long. And yeah, OS/2 was effectively stolen by Microsoft and made into Windows NT. I miss OS/2... it was way too good. I wonder what it would be like today if they continued to develop it.
Re:OS/2 (Score:1, Insightful)
That just means it took longer to fail than the other competitors.
Re:X-Windows? (Score:3, Insightful)
Re:OS/2 (Score:4, Insightful)
"I believe Windows 1.0 predates OS/2 by a bit."
You're right, but OS/2 is worth mentioning anyway. I tried it back in the day, and really liked it. It was a 32 bit os when Windows was still only 16 bit and REXX was a really powerful shell language, much more so than Batch. I'm really sorry it couldn't survive. Although it gave it quite a go. I think I've read comments from /. readers who still use it.
Re:Hard to forget hell. (Score:4, Insightful)
So many people have posted that what you need to succeed against Microsoft is simply greed. I think Jack Tramiel is evidence that this is not true. Greed != Business Acumen.
Windows did fail... Totally. (Score:5, Insightful)
Windows 1.0 was a total failure. Nobody used it. I worked at a computer store at the time and people would ask us to take it off the drives of the compter because they had no use for it.
Windows 2.0 was also a total failure.
Only when Windows 386 and WIndows 3.0 came out was Windows usable. Even then most people didn't use it. It just slowed down their dos programs.
Only when Windows 3.11 came out did WIndows become popular. Mostly to run DOS apps. Windows won because Microsoft just gave it away for the longest time. Almost nobody would have paid for it. That is why all the others failed. Most people wouldn't pay for a program to run programs!
Microsoft used the drug dealer method to win market share. But to call any version of Windows before 3.0 as not a failure is just not valid.
Re:Desqview (Score:4, Insightful)
Same experience here. Nothing at the time other than DESQview was offering decent multitasking for tasks like BBSes. Windows was a joke in comparison.
Eventually I gave up DESQview, but it was a painful transition and I bitterly resented Microsoft for winning in the market with their inferior product.
Microsoft's business model ... (Score:3, Insightful)
... is to own everything from the application down to (and in some cases including) the hardware. It was inevitable that add-ons to DOS were not going to be allowed to survive. The only viable UIs have been those on top of other (non Microsoft controlled) O/Ss. And they have been viable only because Microsoft hasn't been able to kill them off. Yet.
Captcha: penguin
Re:Hard to forget hell. (Score:4, Insightful)
Bullshit - As a C64 and Atari ST veteran, twenty-five years later it's painful to remember the extraordinary effort it took to lose to windows. I had better graphics playing Neuromancer on the C64 than windows managed for a decade, and let's not even talk about comparing Star Flight on the ST vs the DOS version.
Seriously. I never had one -- I was an Apple II fanatic for reasons (obviously) unrelated to its graphics capabilities -- but the Atari ST was an amazing piece of hardware, way ahead of its time, and in retrospect, I can see that it was clearly the best of the 8-bit era. This was a machine with three microprocessors: one general purpose CPU and separate processors for both sound and video. And it was cheaper than most of its competitors. It probably would have been vastly successful if the Atari name hadn't been so firmly associated with games.
I wonder how old the author of TFA is. It's not hard to remember life before Windows at all. I remember life before DOS, back when the first pull-down menus were implemented in WordStar -- a text editor by today's standards -- solely as an aid to learning the key commands.
Hardware and software have come a long way since then, but it came at the expense of losing the rich variety of the early personal computer era, to the point that people now have passionate arguments about the barely perceptible differences between Mac and PC GUIs.
Hm, if I'm not mistaken, this is where I should tell someone to get off my lawn. ;)
The real question is: why just one big incumbant ? (Score:3, Insightful)
There are plenty of motor car manufacturers, and most people don't just drive a Ford (or whatever). So why is the computing market so different ? I don't believe that it is down to manufacturing capacity, ie s/ware is so much easier to make many of once you have the first copy; if that was so then the many smaller manufacturers, the list is huge [wikipedia.org].
I think that the key is standards, everyone wants the same - especially file formats. The way that MS got to where it is was by taking everyone else's standards and keeping its own as secret as it could. Whatever reasons: it is something that we should learn from and stop from happening again.
Disclaimer: my desktop has always been Unix based since 1986, Linux for the last 15 years.
OK. I'll speak the truth and take the hit. (Score:5, Insightful)
The lessons of "just okay" (Score:3, Insightful)
I think sometimes the geeks forget the Marketing adage that most enduring products are functionally "just okay." Typically a successful product uses lots of cash to drown their competitors. Might makes right.
Someone somewhere said "Early to bed. Early to rise. Advertise Advertise Advertise"
Re:OS/2 (Score:3, Insightful)
Depends on what you call a failure. Considering government and banks depended on it heavily for almost 20 years sounds like somewhat of a success to me.
Re:The real question is: why just one big incumban (Score:2, Insightful)
*WE* made it happen that way. We as software devs for the longest time would make 3 different versions and then maybe sell a Mac ver and a Sun/HP Unix ver. We got tired of the 5+ different software configurations *ON TOP OF* the zillions of hardware configurations. Just for our sanity we picked windows. Love em or hate em MS was everywhere. Eventually the only software that was everywhere was windows. But we didnt care. We were too busy selling tons of software...
It should also be noted (Score:3, Insightful)
That one isn't inherently better than the other. At first glance, it sounds like preemptive multi-tasking is the way to go since that is what all our desktops use now and since it is far more stable. However that is only true in an environment flush with resources, as our computers are. It incurs a good deal of overhead, which is why it was more problematic on older hardware. You could do it, but you paid a performance price. That's part of the reason you saw CMT not just on Windows but on things like MacOS as well. If programs behave themselves, it can be a much lower amount of overhead, and that mattered on those slower processors.
For just an idea of how slow they were consider that it took almost all of a 486 to play a stereo 128k MP3. I remember when I first started playing with them and in Windows 95, it wasn't possible. Even running nothing but an MP3 player the overhead from the OS (which wasn't fully preemptive itself) was too much, I had to turn it down to mono or reduce quality to play. To get full stereo I had to drop to DOS and play it with Cubic Player. Now of course we can play them in the background with less than 1% CPU time on a single core.
Just something for people to consider with regards to cooperative vs preemptive. Preemptive works great and is really what you want on a desktop computer where arbitrary code can be executed because it keeps problematic code from running away with resources, and also just makes programming a bit easier (you have to be careful when programming something for a co-op system that will be expected to use as much resources as it can get, yet still cede control properly). However it does incur overhead to make happen, and when you talk a slow enough system, it is a non-trivial amount.
Re:Windows did fail... Totally. (Score:3, Insightful)
Windows 1.0 was a total failure. Nobody used it. I worked at a computer store at the time and people would ask us to take it off the drives of the compter because they had no use for it. Windows 2.0 was also a total failure. Only when Windows 386 and WIndows 3.0 came out was Windows usable. Even then most people didn't use it. It just slowed down their dos programs. Only when Windows 3.11 came out did WIndows become popular. Mostly to run DOS apps. Windows won because Microsoft just gave it away for the longest time. Almost nobody would have paid for it. That is why all the others failed. Most people wouldn't pay for a program to run programs! Microsoft used the drug dealer method to win market share. But to call any version of Windows before 3.0 as not a failure is just not valid.
I call shenanigans!
* Windows 1.0 was MS-DOS EXEC. It didn't have an installation. Also, what drives are you referring to? As I recall hard drives were pretty scarce in 1985 (heck, even into 1988 when IDE really got going), as most XTs (and early ATs) were dual floppy systems!
* Yet Windows 2.0 manged to be successful enough that Apple sued Microsoft (in a 189 point lawsuit) over the same look & feel they "borrowed" from Xerox.
* Also, Windows/386 was a version of Windows 2.1. So much for it being a failure.
* Exactly how did running Windows 3.0 slow down DOS programs when you had to shell into Windows from DOS? Unless you put Win (or Win: to avoid the spashscreen) into your autoexec.bat, it was a manual process to load Windows!
* For that matter, why run DOS programs on Windows 3.11? You still had to shell to it from DOS, though by this time some companies had begun changing the autoexec.bat on their machines (Blackship, Fast Data and Dell come to mind).
BUT! By the time it was released (31 December 1993), Microsoft Office for Windows was already on version 3, and 4 was out a few months later. Nevermind the competing products like Lotus Smartsuite 1994, cc:Mail/Microsoft Mail or even AutoCAD . Or a little thing called Mosaic, which of course led to Internet Explorer... which also ran on Windows 3.11... as did Netscape. Have you ever heard of Novell Netware or Windows NT 3.51? WfW was the corporate client du jour for *years* (they bought it, mostly) and it's success paved the way for Windows 95.
As opposed to what... using bright, shiny polychromatic plastic cases?
Re:OK. I'll speak the truth and take the hit. (Score:4, Insightful)
The question I want answered is "When will Windows be ready for the desktop?", because I guarantee you my Linux box blows the doors of of any Windows machine hands down, and does it all without being a Malware fest.
Re:Windows did fail... Totally. (Score:3, Insightful)
I'll disagree with you there. MS kinda had to give away Windows 3.0, but one change from 3.0 to 3.1 made all the difference in the world: TrueType fonts with WYSIWYG printer output. That was truly the birth of the desktop publishing revolution. The new simplicity of being able to create good looking documents, handouts and brochures *without* having to know any arcane printer commands meant you no longer needed the WordPerfect Guru secretary who knew all the ins and outs of the printer command codes.
Re:Deskmate (Score:3, Insightful)
" TSR...hacks to make your machine use memory above 640k..."
" It is hard for people to understand just how messy things were in those days..."
No, things were NOT that messy in those days. They were that messy in the DOS/Windows world.
Other systems of the time had device drivers that abstracted printing details from apps. They had no 640K barriers. They had preemptive multitasking. They had device independent APIs, and abstracted container file formats.
Don't confuse the mess that was DOS and early Windows with the state of the industry as a whole. It was not that bad, it's just that for whatever reason, everyone chose to support the system that was an architectural clusterfuck.
Re:It should also be noted (Score:1, Insightful)
> At first glance, it sounds like preemptive multi-tasking is the way to go ...
> However that is only true in an environment flush with resources,
It's probably more accurate to say that it's only true on a system with an MMU.
PMT isn't inherently more resource-hungry than CMT; most of the perceived difference comes from that fact that CMT systems typically switch tasks far less frequently than PMT systems. You could get the same effect with a PMT system simply by using longer time-slices.
The problem with trying to implement PMT on a system without an MMU is that the OS cannot reallocate memory as it sees fit. Once it gives RAM to a task, it can't take it away it until the task says so, because there's no mechanism (i.e. page fault) by which it can return the memory when the task needs it.
With CMT, tasks are supposed to acquire temporary buffers and release them before yielding the CPU. With PMT and an MMU, the OS just confiscates the memory as it sees fit and gives it back when needed.
Also, reliability tends to suck if you implement PMT without an MMU (I'm looking at you, Amiga). Programmers access shared data without bothering to lock it, which works most of the time but results in obscure bugs when the data changes at just the wrong time. With CMT, data only changes when it's expected to change, i.e. either when the program modifies it, or when it yields the CPU. With an MMT, the OS can force the application to go through the proper channels.
Perspective from the CAD / DTP viewpoint (Score:3, Insightful)
I've owned and used Top View, GEM, DESQview and Windows 1.0 and all later editions. And I think the real reason Windows won was simple - Drivers.
I was running Lotus 123, Word Perfect, Ventura Publisher, and AutoCAD. I had expensive ($3000+) graphics cards, a 21" monitor and a laser printer (when they were $5000 beasts). Every time a new software release came out, I had to wait months for drivers to appear for the graphics card and printer. Sometimes they never arrived.
When Windows appeared, it wasn't very useful. But they always seemed to have drivers. I switched to Ami Pro, Excel and PageMaker because they all ran on a system (Windows) that had drivers for all my equipment. It was wonderful to be out of the waiting-for-drivers quandary. When Windows 386 appeared, I could run my DOS apps in a Window and not have to switch back-and-forth to DOS.
I'm pretty sure the younger crowd would have no idea what we went through. Every single app either ran at 640x480 (pretty bad on a 21" monitor) or had to have custom drivers. And you only had text printing - no graphics - without drivers. And you only had text printing if your printer emulated the IBM Graphics Printer.
Pretty soon, the hardware vendors started noticing that the availability of Windows driver became a binary decision for consumers - graphics boards with just Windows drivers would sell, while devices without became hard to sell. Companies that focused on Windows-only got the jump on those that had to write dozens of drivers.
Stop and think about the effort of keeping track of drivers for graphics, printer, mouse, modem, keyboard, sound card for EVERY app. And then do it again for each new release of every app. This is why Windows won - at least in my opinion.
Re:OS/2 (Score:3, Insightful)
So, in a somewhat obtuse way, if you want to map out the code legacy, OS/2 lives on.
No it doesn't. NT was not developed from OS/2.
Re:Desqview (Score:3, Insightful)
I stand by my assertion, Microsoft was way behind the curve in real multi-tasking. Arguably, it wasn't until at least Windows '98 where Windows had any meaningful multi-tasking -- almost 20 years after they licensed the technology from someone else.
Windows 98 had the same pre-emptive (for 32-bit processes) multitasking capabilities as Windows 95, NT had fully pre-emptive multitasking two years before that.
DOS, Windows 3.x and OS/2 1.x did not, due to a combination of hardware constraints and legacy support.
Re:Windows did fail... Totally. (Score:3, Insightful)
Only when Windows 386 and WIndows 3.0 came out was Windows usable. Even then most people didn't use it. It just slowed down their dos programs.
Only when Windows 3.11 came out did WIndows become popular. Mostly to run DOS apps. Windows won because Microsoft just gave it away for the longest time. Almost nobody would have paid for it. That is why all the others failed. Most people wouldn't pay for a program to run programs!
Actually, Windows 3.0 was the (surprising to everyone, including Microsoft) turning point, and largely the reason that today we have Windows NT and not OS/2 NT.
Also, people were more than happy to pay for a "program to run programs" when it gave them things like multitasking, WYSIWYG and consistent printer support (then later, TT fonts and networking).