MS provided IE in 1995, but you had to download Microsoft Plus to get it (or install it off of the CD). They now provide IE but it is bundled with the OS.
Plus! wasn't free. You had to buy it. I can't remember if IE 1.0 was freely available for download (probably not), but 2.0 was (not that anyone sane would have used either of them to do more than download Navigator).
Well, I really doubt anyone was "clamouring" for anything, but if they have some useful additions to the spec then by all means implement them, but let everyone else implement them as well. The competition is in the efficacy of the browser, not who can lock competitors out of the most functionality.
The competition is in who can give the users what they want, the fastest.
I fail to see that pointing the finger at another company and saying they did it too is an adequate defense. (And no-one who was around at the time can forget the crime against humanity that was the "blink" tag!)
The assumption here is that it *needs* defending. The point I'm trying to make is that the industry was proceeding in exactly the same way it (and numerous others) had previously. Competitively.
They can convince those who were not alive or who were not there but it was viewed no differently than the Win3.11 failure for DR.Dos back then.
Windows 3.1 worked fine on DR-DOS.
It wasn't even refuted back then.
Of course not. That's because the computer-using world was, on average, more intelligent and such a stupid idea was just laughed about.
It was just accepted that Excel would work and Lotus would not for a while when a new version of DOS came out.
Excel never existed for DOS.
It is really weird to see how effectively they are papering over history.
I bet you can't find a single reliable piece of evidence supporting your claims. Given how easy it is to find old versions of things like DOS floating, you shouldn't have any trouble at all finding some combination of DOS and Lotus 1-2-3 that doesn't work.
IBM did that and it would have done it without Microsoft.
No, they did not. Compaq, et al, the companies who cloned IBM's PC did that. What enabled the PC clones to be viable ? MS-DOS.
Without MS they would incidentially have offered a better product as the competition was technologically superiour (yes, I have used both).
IBM's intentions were clearly demonstrated when they tried to close up the PC platform again with the PS/2. I'm not sure why this revisionist bullshit about how kind-hearted IBM wanted to give the world a cheap computer that anyone with basic hardware familiarity could slap together a copy of in their garage has suddenly started appearing on Slashdot, but it's exactly that - bullshit.
What you are calling gray areas seem to be easily placed in the OS category. You mention a bunch of "tools" (network stack, CLI, GUI, etc.) that are the embodiment of a computer system or help keep the system running, which to me is the definition of an OS.
There are numerous examples of OSes that have no need for any of those things. Hence the reason they're a "gray area" (although they are definitely closer to the black (or white) end of the scale).
How about shared libraries ? 99% of Internet Explorer's (and its equivalents on other platforms) functionality is basically just a shared library.
Ok, I didn't equate provided with bundled, but it was still free in 1995, and included in Microsoft Plus.
Well, there's little difference that I can see between "provided" and "bundled"...
Just for your information, OS/2 didn't include Web Explorer until 1995, and was replaced by Netscape in 1996. But it was "major platform" that I was querying, personally I would have said "alternative platform". But as I don't know how to define a "major platform" you may well be right.
I was just thinking of the two primary competitors to Windows. Despite low market share in absolute numbers, I would consider it reasonable to call the three of them "major" platforms (in the context of the PC).
The fundamental point remains, however, that "everyone" was doing the same thing at the same time. Singling out Microsoft as somehow being "evil" for responding to customer demands and competitive pressure seems a bit... hypocritical. (Especially since the same people will then usually turn around and talk about how Microsoft is a monopoly who never listens to their customers.)
If they had simply provided a browser I don't think it would have been a problem, it was what they did after they provided it, i.e. trying to subvert standards so the web wouldn't work correctly without their browser that upsets people.
People making this judgement seem to forget some rather important aspects of that timeframe:
* The WWW was exploding. A time of rapid advancement. Why on Earth would anyone expect software vendors in that sort of situation to sit on their hands and wait for a "standards committee" to tell them what functionality their customers were clamouring for that they could and could not give provide ? Remember that whole "competition is good" meme ?
* Netscape were doing *exactly* the same thing. How soon we forget who came up with <blink> and HTML email. Their primary business plan was to leverage being in control of both the server and client.
You don't need Safari to update OSX and you don't need Firefox to update Linux. You need IE to update Windows.
No, you don't. Only the Windows Update website (funnily enough) "requires" IE. Automatic Updates, manual installation of patches, or centralised distribution via WSUS do not. Nor do updates in Vista.
and on that note, IE is also a key part of the copy protection system.
How do you figure ?
And, still, no-one has managed to come up with why using a shared component as a shared component is a bad thing in the first place.
You lost me there, what other "major platform" provided a free browser in 1995?
Windows didn't actually come with IE until 1996, with OSR1.
OS/2 had Web Explorer in 1994 (and from memory Warp came with Navigator, or had a direct link to download it on the desktop).
MacOS had Cyberdog in 1996.
Hell, even if Microsoft were the first to include a web browser, why would that be a bad thing ? Aren't people always complaining they never do anything except copy everyone else ?
MS gave API documentation to WP for their windows client. This API wasn't used by MS Word for windows (at the time MS fanbois said there were no secret API's, but these people didn't appear when MS opened up a lot of secret API documentation. go figure) and the API's given to WP were discontinued when the product moved to RC, requiring WP rewrite WPfW when it was released (because MS didn't tell WP the API they used was broken).
Wordperfect for Windows was utter crap from its very first version in 1992, to the first not-quite-so-crap version in 1997. Are you implying that, for five years, the only company who had sufficient API documentation for Windows 3.x and Windows 95 to write functional applications was Microsoft ? Because a lot of other third party applications released in that time period would suggest otherwise.
Wordperfect on Windows failed for much the same reason Lotus 1-2-3 on Windows did. They were implemented as DOS programs with a shoddy GUI grafted on top, rather than proper Windows programs in line with Windows UI standards, etc.
It's integrated into the OS and into many of its applications. Many things won't work without it. Like Windows Update, Steam, Active Desktop and just about any app that integrates IE to view webpages.
In most circles, modularity and code-reuse are considered *good* things.
You can't really remove it without crippling the rest of the system.
And how much stuff that depends on khtml or WebKit do you think would keep working if you just ripped them out ?
The "tightly bundled" argument is basically complaining because a piece of shared code is being used exactly in the way shared code is supposed to be. I fail to see why it is a valid complaint.
It may be a while before the average person has that setup [...]
I'd be inclined to say the average person will never have that setup, if for no other reason than the space constraints - most people don't live in houses big enough to dedicate a whole room just to watching movies (to say nothing of those living in apartments).
What about Windows 3.1 refusing to run specifically under DR DOS? That is a bit more documented, you will have to erase half of internet to cover that.
Windows 3.1 ran fine under DR-DOS. Even the much ballyhooed warning message during the Windows 3.1 beta - presumably what you were referring to - didn't actually stop it running.
(The weird part about that situation was not the warning message itself - perfectly reasonable for something that relied on specific implementation details of DOS as much as Windows 3.x did - but more the code obfuscating the check. Regardless, it didn't actually stop Windows running on DR-DOS.)
I may be wrong, but I was taught that there is a division between an operating system and the applications that it runs. The OS is supposed to handle things like IO and memory, while the real functionality comes from userland applications (often third party) interacting with the OS.
This distinction is basically meaningless outside of the academic world. Even within it, the lines are very blurry. Does a network stack count as part of the OS ? For which protocols ? How about a command line shell ? How about a GUI shell ? How about just the windowing system ? Where do shared libraries like glibc fit into the picture ? Etc, etc.
There are some things basically everyone can agree are "the OS" (memory management, process scheduling) and there are some things basically everyone can agree are not (word processors, video editing tools). But there is a massive grey area between those two ends of the scale.
A: An OS Vendor who's also trying to sell a competing software to said 90% of their potential customers.
Except no remotely intelligent businessman would expect that many people to spend the money to just up and switch platforms - and if there's one thing Bill Gates was, it was a good businessman.
What's particularly stupid about the whole "... 'til Lotus won't run" myth, is that - just plain common sense aside - historical behaviour demonstrates the exact opposite is true. Microsoft works very, very hard to maintain backwards compatibility, arguably to a fault. The idea they'd be deliberately breaking such a high profile and important application as Lotus 1-2-3 - *especially* back in the '80s when they were far from the only option - is just laughable.
One that would have such market domination that it could get away with it.
In 1985, Microsoft didn't have anything remotely close to the "market domination" necessary to force a mass switch of application. Heck, it's quite arguable whether or not they've ever had it.
Since you obviously weren't there to experience them, let me assure you that both Wordperfect for Windows (especially until the late '90s) and Netscape Navigator 4.0 were, in fact, bad products.
WordPerfect was the word processor of choice for lawyers. The "Reveal Codes" function was very well-liked for formatting legal documents. In some shops it is still the preferred word processor.
Indeed. But there are a lot more people out there than lawyers using word processors.
"DOS ain't done, til Lotus won't Run" was *well* known back in the 80's in my user group.
Then you shouldn't have any trouble at all coming up with some documented examples of incompatibilities between 1-2-3 and DOS. Or even any evidence at all, really, that would support such an idiotic idea.
Windows 95 did it all over again by certifying Word which cheated and used invalid API's.
Evidence ? Even if it were true, what exactly is it "cheating" at ?
Indeed, IIRC they even had an internal slogan -- "it's not done til Lotus won't run", or something like that.
Which - even ignoring the utter lack of even the slightest actual evidence of this ever being true - would have sounded even dumber when it first surfaced back in the mid-80s than it does today. What sane OS vendor would lock out 90% of its potential customers by not running their primary application ?
I disagree. I noticed MS being evil with the introduction of Windows 95, when the then-standard Word Perfect oddly didn't seem to run properly under Windows.
Which wouldn't have had anything at all to do with the abominable implementations on Windows at all, right ?
Not to mention, when 1995 rolled around, Word Perfect was well on its way out (and with good reason). The aforementioned almost incomprehensibly bad Windows implementations had sealed its fate. By the time the first semi-decent version of Wordperfect for Windows was released in mid-1997, the game was well and truly over.
Shortly thereafter came MSN and the introduction of the free Internet Explorer and the beginnings of Netscape's death. That was several years before Ballmer entered the picture.
Indeed. Providing a free web browser - just like every other major platform of the day did - was the very embodiment of "evil".
Wordperfect and Navigator are textbook examples of bad products being displaced in the market by better ones (although the first few Wordperfect for Windows iterations were orders of magnitude worse than even Navigator 4.0).
Slashdot Mirror
MS provided IE in 1995, but you had to download Microsoft Plus to get it (or install it off of the CD). They now provide IE but it is bundled with the OS.
Plus! wasn't free. You had to buy it. I can't remember if IE 1.0 was freely available for download (probably not), but 2.0 was (not that anyone sane would have used either of them to do more than download Navigator).
Well, I really doubt anyone was "clamouring" for anything, but if they have some useful additions to the spec then by all means implement them, but let everyone else implement them as well. The competition is in the efficacy of the browser, not who can lock competitors out of the most functionality.
The competition is in who can give the users what they want, the fastest.
I fail to see that pointing the finger at another company and saying they did it too is an adequate defense. (And no-one who was around at the time can forget the crime against humanity that was the "blink" tag!)
The assumption here is that it *needs* defending. The point I'm trying to make is that the industry was proceeding in exactly the same way it (and numerous others) had previously. Competitively.
They can convince those who were not alive or who were not there but it was viewed no differently than the Win3.11 failure for DR.Dos back then.
Windows 3.1 worked fine on DR-DOS.
It wasn't even refuted back then.
Of course not. That's because the computer-using world was, on average, more intelligent and such a stupid idea was just laughed about.
It was just accepted that Excel would work and Lotus would not for a while when a new version of DOS came out.
Excel never existed for DOS.
It is really weird to see how effectively they are papering over history.
I bet you can't find a single reliable piece of evidence supporting your claims. Given how easy it is to find old versions of things like DOS floating, you shouldn't have any trouble at all finding some combination of DOS and Lotus 1-2-3 that doesn't work.
IBM did that and it would have done it without Microsoft.
No, they did not. Compaq, et al, the companies who cloned IBM's PC did that. What enabled the PC clones to be viable ? MS-DOS.
Without MS they would incidentially have offered a better product as the competition was technologically superiour (yes, I have used both).
IBM's intentions were clearly demonstrated when they tried to close up the PC platform again with the PS/2. I'm not sure why this revisionist bullshit about how kind-hearted IBM wanted to give the world a cheap computer that anyone with basic hardware familiarity could slap together a copy of in their garage has suddenly started appearing on Slashdot, but it's exactly that - bullshit.
What you are calling gray areas seem to be easily placed in the OS category. You mention a bunch of "tools" (network stack, CLI, GUI, etc.) that are the embodiment of a computer system or help keep the system running, which to me is the definition of an OS.
There are numerous examples of OSes that have no need for any of those things. Hence the reason they're a "gray area" (although they are definitely closer to the black (or white) end of the scale).
How about shared libraries ? 99% of Internet Explorer's (and its equivalents on other platforms) functionality is basically just a shared library.
Dev 1: Let's reuse the browser as the update manager!
Your straw men aren't particularly germane.
Separate cleanly. Can I separate the update functionality, or Active Desktop (ugh) from IE?
Of course not. No more than you can separate any pieces of dependent functionality.
The "update functionality" (which is actually just the Windows Update web page) is an ActiveX control. Hence, you need ActiveX to use it.
Active Desktop uses IE to render. Hence, you need IE to use it.
Complaining you can't "separate" them is like complaining you can't rip Qt out of a KDE install and not have anything break.
Ok, I didn't equate provided with bundled, but it was still free in 1995, and included in Microsoft Plus.
Well, there's little difference that I can see between "provided" and "bundled"...
Just for your information, OS/2 didn't include Web Explorer until 1995, and was replaced by Netscape in 1996. But it was "major platform" that I was querying, personally I would have said "alternative platform". But as I don't know how to define a "major platform" you may well be right.
I was just thinking of the two primary competitors to Windows. Despite low market share in absolute numbers, I would consider it reasonable to call the three of them "major" platforms (in the context of the PC).
The fundamental point remains, however, that "everyone" was doing the same thing at the same time. Singling out Microsoft as somehow being "evil" for responding to customer demands and competitive pressure seems a bit... hypocritical. (Especially since the same people will then usually turn around and talk about how Microsoft is a monopoly who never listens to their customers.)
If they had simply provided a browser I don't think it would have been a problem, it was what they did after they provided it, i.e. trying to subvert standards so the web wouldn't work correctly without their browser that upsets people.
People making this judgement seem to forget some rather important aspects of that timeframe:
* The WWW was exploding. A time of rapid advancement. Why on Earth would anyone expect software vendors in that sort of situation to sit on their hands and wait for a "standards committee" to tell them what functionality their customers were clamouring for that they could and could not give provide ? Remember that whole "competition is good" meme ?
* Netscape were doing *exactly* the same thing. How soon we forget who came up with <blink> and HTML email. Their primary business plan was to leverage being in control of both the server and client.
You don't need Safari to update OSX and you don't need Firefox to update Linux. You need IE to update Windows.
No, you don't. Only the Windows Update website (funnily enough) "requires" IE. Automatic Updates, manual installation of patches, or centralised distribution via WSUS do not. Nor do updates in Vista.
and on that note, IE is also a key part of the copy protection system.
How do you figure ?
And, still, no-one has managed to come up with why using a shared component as a shared component is a bad thing in the first place.
Don't forget the rest of us who don't shoot for "theater theme" and just have our panel mounted on the wall and a THX system tucked away out of sight.
Well, yeah, but that's not really offering the same experience as a wall-sized movie screen and a dozen speakers, is it ? :)
You lost me there, what other "major platform" provided a free browser in 1995?
Windows didn't actually come with IE until 1996, with OSR1.
OS/2 had Web Explorer in 1994 (and from memory Warp came with Navigator, or had a direct link to download it on the desktop).
MacOS had Cyberdog in 1996.
Hell, even if Microsoft were the first to include a web browser, why would that be a bad thing ? Aren't people always complaining they never do anything except copy everyone else ?
MS gave API documentation to WP for their windows client. This API wasn't used by MS Word for windows (at the time MS fanbois said there were no secret API's, but these people didn't appear when MS opened up a lot of secret API documentation. go figure) and the API's given to WP were discontinued when the product moved to RC, requiring WP rewrite WPfW when it was released (because MS didn't tell WP the API they used was broken).
Wordperfect for Windows was utter crap from its very first version in 1992, to the first not-quite-so-crap version in 1997. Are you implying that, for five years, the only company who had sufficient API documentation for Windows 3.x and Windows 95 to write functional applications was Microsoft ? Because a lot of other third party applications released in that time period would suggest otherwise.
Wordperfect on Windows failed for much the same reason Lotus 1-2-3 on Windows did. They were implemented as DOS programs with a shoddy GUI grafted on top, rather than proper Windows programs in line with Windows UI standards, etc.
It's integrated into the OS and into many of its applications. Many things won't work without it. Like Windows Update, Steam, Active Desktop and just about any app that integrates IE to view webpages.
In most circles, modularity and code-reuse are considered *good* things.
You can't really remove it without crippling the rest of the system.
And how much stuff that depends on khtml or WebKit do you think would keep working if you just ripped them out ?
The "tightly bundled" argument is basically complaining because a piece of shared code is being used exactly in the way shared code is supposed to be. I fail to see why it is a valid complaint.
It may be a while before the average person has that setup [...]
I'd be inclined to say the average person will never have that setup, if for no other reason than the space constraints - most people don't live in houses big enough to dedicate a whole room just to watching movies (to say nothing of those living in apartments).
As far as I know, this is impossible in Linux, hence it's not business ready IMHO.
It's all _doable_ on Linux, it just involves a lot more work because most of it you need to DIY.
The problem here is that most MCSE's are still only capable of setting up a fisher-price level network.
Most *people* are only capable of setting up a "fisher-price level of network". That some of them are MCSEs is incidental.
This is largely because most networks only _need_ to be "fisher-price level".
What about Windows 3.1 refusing to run specifically under DR DOS? That is a bit more documented, you will have to erase half of internet to cover that.
Windows 3.1 ran fine under DR-DOS. Even the much ballyhooed warning message during the Windows 3.1 beta - presumably what you were referring to - didn't actually stop it running.
(The weird part about that situation was not the warning message itself - perfectly reasonable for something that relied on specific implementation details of DOS as much as Windows 3.x did - but more the code obfuscating the check. Regardless, it didn't actually stop Windows running on DR-DOS.)
I may be wrong, but I was taught that there is a division between an operating system and the applications that it runs. The OS is supposed to handle things like IO and memory, while the real functionality comes from userland applications (often third party) interacting with the OS.
This distinction is basically meaningless outside of the academic world. Even within it, the lines are very blurry. Does a network stack count as part of the OS ? For which protocols ? How about a command line shell ? How about a GUI shell ? How about just the windowing system ? Where do shared libraries like glibc fit into the picture ? Etc, etc.
There are some things basically everyone can agree are "the OS" (memory management, process scheduling) and there are some things basically everyone can agree are not (word processors, video editing tools). But there is a massive grey area between those two ends of the scale.
And how joyous it would have been if IE were not so tightly integrated with Windows following that theory.
Except IE is no more "tightly bundled" into Windows than the equivalents are into OSX and KDE- and GNOME-based distros.
A: An OS Vendor who's also trying to sell a competing software to said 90% of their potential customers.
Except no remotely intelligent businessman would expect that many people to spend the money to just up and switch platforms - and if there's one thing Bill Gates was, it was a good businessman.
What's particularly stupid about the whole "... 'til Lotus won't run" myth, is that - just plain common sense aside - historical behaviour demonstrates the exact opposite is true. Microsoft works very, very hard to maintain backwards compatibility, arguably to a fault. The idea they'd be deliberately breaking such a high profile and important application as Lotus 1-2-3 - *especially* back in the '80s when they were far from the only option - is just laughable.
One that would have such market domination that it could get away with it.
In 1985, Microsoft didn't have anything remotely close to the "market domination" necessary to force a mass switch of application. Heck, it's quite arguable whether or not they've ever had it.
They weren't bad products.
Since you obviously weren't there to experience them, let me assure you that both Wordperfect for Windows (especially until the late '90s) and Netscape Navigator 4.0 were, in fact, bad products.
WordPerfect was the word processor of choice for lawyers. The "Reveal Codes" function was very well-liked for formatting legal documents. In some shops it is still the preferred word processor.
Indeed. But there are a lot more people out there than lawyers using word processors.
Has MS done anything innovative in the past few years?
Has anyone ? Without knowing your measure of what counts as "innovation", it's an impossible question to answer.
"DOS ain't done, til Lotus won't Run" was *well* known back in the 80's in my user group.
Then you shouldn't have any trouble at all coming up with some documented examples of incompatibilities between 1-2-3 and DOS. Or even any evidence at all, really, that would support such an idiotic idea.
Windows 95 did it all over again by certifying Word which cheated and used invalid API's.
Evidence ? Even if it were true, what exactly is it "cheating" at ?
Indeed, IIRC they even had an internal slogan -- "it's not done til Lotus won't run", or something like that.
Which - even ignoring the utter lack of even the slightest actual evidence of this ever being true - would have sounded even dumber when it first surfaced back in the mid-80s than it does today. What sane OS vendor would lock out 90% of its potential customers by not running their primary application ?
I disagree. I noticed MS being evil with the introduction of Windows 95, when the then-standard Word Perfect oddly didn't seem to run properly under Windows.
Which wouldn't have had anything at all to do with the abominable implementations on Windows at all, right ?
Not to mention, when 1995 rolled around, Word Perfect was well on its way out (and with good reason). The aforementioned almost incomprehensibly bad Windows implementations had sealed its fate. By the time the first semi-decent version of Wordperfect for Windows was released in mid-1997, the game was well and truly over.
Shortly thereafter came MSN and the introduction of the free Internet Explorer and the beginnings of Netscape's death. That was several years before Ballmer entered the picture.
Indeed. Providing a free web browser - just like every other major platform of the day did - was the very embodiment of "evil".
Wordperfect and Navigator are textbook examples of bad products being displaced in the market by better ones (although the first few Wordperfect for Windows iterations were orders of magnitude worse than even Navigator 4.0).