The question was about Word, Excel and PowerPoint. Whether they were sold in a bundle at the time is irrelevant. But if it make you happy, translate Office as "the Microsoft Applications that were later bundled in Office". The time period of the "big scandal" was the early 1990s at the time of the Apple lawsuit and the first FTC/DOJ suits. The end result was the same. The "Secret API" urban legend is just that. (and a way for MS competitors to explain why their stuff is slow)
There's a reason why this wasn't in "Pirates of Silicon Valley" or anywhere else. It is completely and totally wrong. Actually there were four "Mac like" systems in development. VisiCorp's VisiOn, Digital Research's GEM, IBM's TopView and Microsoft's Windows. Of these, VisiOn was by far the most vaporous.
VisiOn was shown first of the four and finally shipped last - well after Windows, TopView and GEM. VisiCorp bundled it with their own apps so nobody wanted to develop for it and the development tools were next to impossible to get. Not only that but it would only work if you used their VisiMouse with it. In fact, it was the classic industry example of bad preannouncement for years.
As for people signing their lives over to MS, the only commercial apps for Windows 1.0 were a graphics app from MicroGrafx and the game Balance of Power, both of which came with the run-time version of Windows bundled with them since neither of them were willing to bet their sales on the pitifully small Windows installed base. Remember that Windows had almost no users until Windows 3.0 which was five years later.
Everybody else either didn't bother or wrote for TopView which with IBM's name was expected to be the winner and didn't require much change to existing apps. There were also a few people writing for GEM.
In the end, TopView flopped because their only real feature was multitasking and QuarterDeck's DesqView did it better, GEM was destroyed by the Apple look and feel lawsuit and VisiOn finally shipped at a ridiculous price years late. Oh, and Microsoft lost money on Windows for five years in a row after it released but kept working on it thus proving that persistance sometimes pays off over good business sense.
Actually, MITS did have a working Altair running at the time of the article. It was lost on the flight out to Popular Electronics when they were shooting the cover photo so the cover just has a dummy box. People were getting Altair kits within weeks of the article. There was a backlog on assembled versions and later kits because MITS never anticipated that level of demand and ran out of parts and people to solder them together.
Actually, they wrote it (along with Monte Davidoff) over a few weeks on the Harvard mainframe using an emulator they also wrote. What was written on the trip down there was the bootstrap loader since Paul Allen forgot to pack the listing. (You had to toggle it in to be able to load a paper tape)
In the case of Kerberos, Microsoft used the method for proprietary extension that was described in the specs. The put proprietary data in a field specifically reserved for proprietary extensions.
In the case of Java, Microsoft provided both compliant and extended versions of calls in all their tools and runtimes and documented which ones would only run on Windows systems. Whether they used the right naming convention is what the meat of the case is about not whether they had the right to extend Java to add Windows specific calls. That is something that Sun specifically allowed.
If Microsoft had to wait for Sun to provide a public standard for anything, they'd still be waiting while Sun tries to find a standards body that wants to be a rubber stamp.
You mean the same way MS Excel was using undocumented O.S. routines which its competitor, Lotus 1-2-3, could not use (was it in Windows 3.1?). I think Lotus won a court case about that, although it wasn't mentioned in the MS Antitrust case.
This is an Urban Legend.
The only time anybody researched and documented the "secret API calls" in Windows software, they found less than a dozen used in all of Office. None of which offered any performance benefit. Almost all of these were duplicated by published calls (the developers used some older beta format calls and didn't replace them). They also found that WordPerfect and Lotus 1-2-3 actually used more of these "secret API calls" than their Microsoft equivalents.
Actually, when Mark Ursino coined the phrase it meant products that existed only in their press releases. At the time a lot of companies would do the press release and possibly a mock up and show that to the press and at trade shows to get orders. They'd then take the money from those orders to pay for tha actual development while putting out announcements about the final version being delayed. So any product that is even in development isn't truly vaporware.
Yet another x86 clone? Why is this worth even discussing? Especially on/. where most of the readers are running an architecture independant OS anyway? Come on people, do something interesting!
Is Microsoft wrong when they talk about inventing Windows?
No, they're not wrong because Microsoft did NOT buy the original Windows code. It was written by Microsoft people at the Microsoft building at 10700 Northup Way in Bellevue, Washington. Perhaps you're thinking of MS-DOS 1.0 which was derived from Seattle Computing's QDOS (and didn't support hard drives, pipes, directories, etc. which Microsoft added)?
But you are missing one exceedingly important point:
Linux != UNIX
Actually, that Linux != Unix was implicit in my point. If Linux == Unix then it would have had an installed base as a cost of that decision. The fact that Linux != Unix is precisely the tragedy. Since it isn't Unix it didn't need to inherit all the baggage but instead of throwing all that legacy out and looking for new ideas, the community embraced the legacy as a shortcut to "legitimacy".
The one thing that every (and I do mean every) existing OS designer and developer would kill for is the chance to do their work without an installed base so they could innovate. The tragedy of Linux is that the first really popular alternate OS was built by a community who's sole goal is getting an installed base even if they have to grab somebody elses. The end result being a non-mature OS with a thirty year old stack of legacy installed base compatibility to lock themselves down with.
OK, here's a better example. You spend 5 years of your life and $100,000 creating your wonderful new product Foo. BigCorp, Inc. makes a copy of your product and names it Bar.
They then spend their remaining $99,998 of their $100,000 product budget on marketing and distribution and sales reps for BigCorp Bar which becomes a big seller. People think that Foo is just a cheap knock-off of BigCorp Bar and since you can't make it as cheaply as they can, your "knock-off" is even more expensive.
BigCorp, Inc. got all their development costs for free and now has a successful product. You are out of business, lost 5 years of your life and $100,000.
Of course, if you think engineers and inventors and innovators don't deserve being paid but marketing and sales do, then, by all means, think intellectual property is blatently stupid. Personally, I'd prefer to reward the people who actually create something.
Actually, IBM had a much stronger hold than Microsoft ever had. They virtually owned all hardware and software planet wide and even controlled a good portion of the US educational system. For a while there was talk of privatizing the department of education and handing it over to SRA. It was only their lack of understanding of new markets combined with inter-division rivalries that brought them down.
No, but they should do some due diligence to make sure that new systems will work with existing OSes. At keast if they want to claim that the CPU is backwards-compatbile with x86.
I'm guessing you don't know what a CPUID is. It is the identifier for a specific processor family. It is the one thing that CAN'T be backward compatible.
NASA did a study back in the '60s that showed that using tones for status information was a very effective method for important (but non-critical) information. This technique was used on the Apollo program. After a while the normal sounds become part of the background noise but when something changes, the brain picks up on it very quickly. It has the advantage that the human does it as a background task.
That's called backward compatibility. Windows 98 only does that if you have real-mode drivers and have them loaded in your CONFIG.SYS file with a DEVICE=xxxxxxxx.SYS command when you install Windows 98 and if Windows 98 doesn't have a replacement protected-mode driver that it can use. So if you depend on them you can still use them, if you don't then the boot sequences skips the real-mode driver load process.
Of course, that isn't true about Windows ME. By 2000 there were few enough users that still needed backward compatibility with their old hardware that Microsoft could drop the "real-mode pause" step. Back in 1998 there were too many. Oh, and there are users out there complaining about dropping the real mode driver support in Windows ME anyway.
NT used to run on x86, Alpha, MIPS, and PPC, but those were gradually killed off so that only Intel remains.
You forgot the Fairchild/Intergraph Clipper which was also a Windows NT platform processor.
The reason Windows NT was killed off on those processors was that nobody bought them. In most of those cases (IBM, DEC/Compaq and Integraph for sure - I can't remember about MIPS) they were killed off by the hardware vendors who wanted to drop support costs for their almost totally non-selling products while Microsoft wanted to keep them going.
Oh, and I didn't get it from scanning a book. I was in that part of the industry at the time.
The "IBM is a hardware company" smoke screen is an old M$ propaganda mantra.
The classic IBM statement was:
IBM is a sales company. Making hardware is just to give sales something to sell. Making software is just to make selling the hardware easier.
Believe me, if they could make sales without hardware or software they'd love to. Since selling support and services is a way to make sales without all that nasty engineering staff, they love it.
In reality linux is the marketing only OS. Let's look at a typical linux distro or dev shop.
Marketing - $alary plus $tock
Venture Capital - lots of $tock
Management - $alary plus $tock
Developers - mostly "the thanks of the community"
Testers - almost totally "the thanks of the community"
Let's be honest, in a Linux distro or dev shop, the money you get is inversely proportional to your tech level. (The "mostly" above is because most of the code involved wasn't developed in house). And what's amazing is that a lot of Linux techies actually don't notice that this is just a way of putting the money back in the non-tech hands.
The fact that this is NOT indeed the case is why many of us never bothered with NT. The compatibility isn't "perfect enough" to justify selling out for the ability to "run everything".
And that is the only reason why Windows 9x (Windows ME) still exists. If the backward compatibility were there for things like games then Windows 2000 would be the only Microsoft desktop OS. When Windows "Whistler" ships then you can expect to see the 9x OSs phased out.
As a starting point, I have the greatest respect for RMS and the GNU projects. Their philosophy may be pie-in-the-sky but seems thought through and sincere. The thought experiment is finding out whether there's a way to have a way to create non-gratis but yet "free" software. As it currently exists, my gut tells me that the GNU people care about "free" but most of the Linux community are much more concerned with gratis. I'd love to see how true that gut feeling is.
The current "Open Source" model seems to guarantee that the money made is inversely related to the technical expertise involved. Marketers and Venture Capitalists make money but programmers and architects are expected to do it "for the good of the community". This cannot be what anybody in the technical community intends but it has been the reality and does fit the historical model of technical economies.
On the BIOS issue, if we carry it to the extreme, the layout of the chips and the logic design of the chips themselves are just frozen software as well. There's no real difference except in packaging since boolean logic is boolean logic whether the AND is expressed in a gate or a bit. Since it is unlikely that whole new architectures will be made due to the economic and commonality issues it seems that there will always be some blind spot here. Must software be free when it is on a disk, on an EPROM, on a ROM, incorporated in the logic design of a dedicated support chip, in the processor architecture?
Slashdot Mirror
The question was about Word, Excel and PowerPoint. Whether they were sold in a bundle at the time is irrelevant. But if it make you happy, translate Office as "the Microsoft Applications that were later bundled in Office". The time period of the "big scandal" was the early 1990s at the time of the Apple lawsuit and the first FTC/DOJ suits. The end result was the same. The "Secret API" urban legend is just that. (and a way for MS competitors to explain why their stuff is slow)
VisiOn was shown first of the four and finally shipped last - well after Windows, TopView and GEM. VisiCorp bundled it with their own apps so nobody wanted to develop for it and the development tools were next to impossible to get. Not only that but it would only work if you used their VisiMouse with it. In fact, it was the classic industry example of bad preannouncement for years.
As for people signing their lives over to MS, the only commercial apps for Windows 1.0 were a graphics app from MicroGrafx and the game Balance of Power, both of which came with the run-time version of Windows bundled with them since neither of them were willing to bet their sales on the pitifully small Windows installed base. Remember that Windows had almost no users until Windows 3.0 which was five years later.
Everybody else either didn't bother or wrote for TopView which with IBM's name was expected to be the winner and didn't require much change to existing apps. There were also a few people writing for GEM.
In the end, TopView flopped because their only real feature was multitasking and QuarterDeck's DesqView did it better, GEM was destroyed by the Apple look and feel lawsuit and VisiOn finally shipped at a ridiculous price years late. Oh, and Microsoft lost money on Windows for five years in a row after it released but kept working on it thus proving that persistance sometimes pays off over good business sense.
Actually, MITS did have a working Altair running at the time of the article. It was lost on the flight out to Popular Electronics when they were shooting the cover photo so the cover just has a dummy box. People were getting Altair kits within weeks of the article. There was a backlog on assembled versions and later kits because MITS never anticipated that level of demand and ran out of parts and people to solder them together.
Actually, they wrote it (along with Monte Davidoff) over a few weeks on the Harvard mainframe using an emulator they also wrote. What was written on the trip down there was the bootstrap loader since Paul Allen forgot to pack the listing. (You had to toggle it in to be able to load a paper tape)
In the case of Java, Microsoft provided both compliant and extended versions of calls in all their tools and runtimes and documented which ones would only run on Windows systems. Whether they used the right naming convention is what the meat of the case is about not whether they had the right to extend Java to add Windows specific calls. That is something that Sun specifically allowed.
If Microsoft had to wait for Sun to provide a public standard for anything, they'd still be waiting while Sun tries to find a standards body that wants to be a rubber stamp.
This is an Urban Legend.
The only time anybody researched and documented the "secret API calls" in Windows software, they found less than a dozen used in all of Office. None of which offered any performance benefit. Almost all of these were duplicated by published calls (the developers used some older beta format calls and didn't replace them). They also found that WordPerfect and Lotus 1-2-3 actually used more of these "secret API calls" than their Microsoft equivalents.
Actually, when Mark Ursino coined the phrase it meant products that existed only in their press releases. At the time a lot of companies would do the press release and possibly a mock up and show that to the press and at trade shows to get orders. They'd then take the money from those orders to pay for tha actual development while putting out announcements about the final version being delayed. So any product that is even in development isn't truly vaporware.
Yet another x86 clone? Why is this worth even discussing? Especially on /. where most of the readers are running an architecture independant OS anyway? Come on people, do something interesting!
Examples:
No, they're not wrong because Microsoft did NOT buy the original Windows code. It was written by Microsoft people at the Microsoft building at 10700 Northup Way in Bellevue, Washington. Perhaps you're thinking of MS-DOS 1.0 which was derived from Seattle Computing's QDOS (and didn't support hard drives, pipes, directories, etc. which Microsoft added)?
Linux != UNIX
Actually, that Linux != Unix was implicit in my point. If Linux == Unix then it would have had an installed base as a cost of that decision. The fact that Linux != Unix is precisely the tragedy. Since it isn't Unix it didn't need to inherit all the baggage but instead of throwing all that legacy out and looking for new ideas, the community embraced the legacy as a shortcut to "legitimacy".
The one thing that every (and I do mean every) existing OS designer and developer would kill for is the chance to do their work without an installed base so they could innovate. The tragedy of Linux is that the first really popular alternate OS was built by a community who's sole goal is getting an installed base even if they have to grab somebody elses. The end result being a non-mature OS with a thirty year old stack of legacy installed base compatibility to lock themselves down with.
They then spend their remaining $99,998 of their $100,000 product budget on marketing and distribution and sales reps for BigCorp Bar which becomes a big seller. People think that Foo is just a cheap knock-off of BigCorp Bar and since you can't make it as cheaply as they can, your "knock-off" is even more expensive.
BigCorp, Inc. got all their development costs for free and now has a successful product. You are out of business, lost 5 years of your life and $100,000.
Of course, if you think engineers and inventors and innovators don't deserve being paid but marketing and sales do, then, by all means, think intellectual property is blatently stupid. Personally, I'd prefer to reward the people who actually create something.
Actually, IBM had a much stronger hold than Microsoft ever had. They virtually owned all hardware and software planet wide and even controlled a good portion of the US educational system. For a while there was talk of privatizing the department of education and handing it over to SRA. It was only their lack of understanding of new markets combined with inter-division rivalries that brought them down.
OK. How do you make an identifier for a product that is unique to it but also identical to its predecessor?
Perhaps you should have looked up what a CPUID was rather than how known broken code uses it!
I'm guessing you don't know what a CPUID is. It is the identifier for a specific processor family. It is the one thing that CAN'T be backward compatible.
NASA did a study back in the '60s that showed that using tones for status information was a very effective method for important (but non-critical) information. This technique was used on the Apollo program. After a while the normal sounds become part of the background noise but when something changes, the brain picks up on it very quickly. It has the advantage that the human does it as a background task.
That's called backward compatibility. Windows 98 only does that if you have real-mode drivers and have them loaded in your CONFIG.SYS file with a DEVICE=xxxxxxxx.SYS command when you install Windows 98 and if Windows 98 doesn't have a replacement protected-mode driver that it can use. So if you depend on them you can still use them, if you don't then the boot sequences skips the real-mode driver load process.
Of course, that isn't true about Windows ME. By 2000 there were few enough users that still needed backward compatibility with their old hardware that Microsoft could drop the "real-mode pause" step. Back in 1998 there were too many. Oh, and there are users out there complaining about dropping the real mode driver support in Windows ME anyway.
You forgot the Fairchild/Intergraph Clipper which was also a Windows NT platform processor.
The reason Windows NT was killed off on those processors was that nobody bought them. In most of those cases (IBM, DEC/Compaq and Integraph for sure - I can't remember about MIPS) they were killed off by the hardware vendors who wanted to drop support costs for their almost totally non-selling products while Microsoft wanted to keep them going.
Oh, and I didn't get it from scanning a book. I was in that part of the industry at the time.
Since some people on "enterprise networks" are likely to be reading this, a lot of us.
The classic IBM statement was:
Believe me, if they could make sales without hardware or software they'd love to. Since selling support and services is a way to make sales without all that nasty engineering staff, they love it.In reality linux is the marketing only OS. Let's look at a typical linux distro or dev shop.
- Marketing - $alary plus $tock
- Venture Capital - lots of $tock
- Management - $alary plus $tock
- Developers - mostly "the thanks of the community"
- Testers - almost totally "the thanks of the community"
Let's be honest, in a Linux distro or dev shop, the money you get is inversely proportional to your tech level. (The "mostly" above is because most of the code involved wasn't developed in house). And what's amazing is that a lot of Linux techies actually don't notice that this is just a way of putting the money back in the non-tech hands.And that is the only reason why Windows 9x (Windows ME) still exists. If the backward compatibility were there for things like games then Windows 2000 would be the only Microsoft desktop OS. When Windows "Whistler" ships then you can expect to see the 9x OSs phased out.
The current "Open Source" model seems to guarantee that the money made is inversely related to the technical expertise involved. Marketers and Venture Capitalists make money but programmers and architects are expected to do it "for the good of the community". This cannot be what anybody in the technical community intends but it has been the reality and does fit the historical model of technical economies.
On the BIOS issue, if we carry it to the extreme, the layout of the chips and the logic design of the chips themselves are just frozen software as well. There's no real difference except in packaging since boolean logic is boolean logic whether the AND is expressed in a gate or a bit. Since it is unlikely that whole new architectures will be made due to the economic and commonality issues it seems that there will always be some blind spot here. Must software be free when it is on a disk, on an EPROM, on a ROM, incorporated in the logic design of a dedicated support chip, in the processor architecture?