In my book, a bigger problem is that the OS X GUI is "cake frosting" over a much more complex underlying system. There's endless system subdirectories filled with mystery files -- some hidden from the Finder. There's certain tasks where the user MUST drop to the command line. The user is much more dependent on installers.
That leads to a system that's significantly more complex to troubleshoot and support than Classic MacOS. No more drag-n-drop of Extension icons, no more Space-Boot, etc. It brings about a certain fear factor among the normal Mac userbase.
And that removes the primary market advantage that MacOS held. Once the system becomes as 'difficult' as Windows or Unix, then all you've got left is lickable graphics.
Yeah. Like IE, Mozilla's practically designed for such a thing. Just look at all the crazy stuff on the "Firebird Extensions" page that installs in 2 clicks.
Were there a substantial userbase on Mozilla, it would be pretty easy to use social engineering and get folks to install something malicious.
The current ones have hardware emulation, yes. But I believe that Intel has announced that feature is going away because you'll get better performance with software.
I can develop and test n-tier applications on my laptop while sitting on the bus if I want.
It's also useful for stuff like COM development, so you can blow away you environment and not have to worry about all the fragile registration stuff. However, they use VMWare for this at work.
Uh, who are these people? Did you read that in a marketing brouchure?
The NT4 legacy users I've seen continue to run NT4 on bare metal. Also, unlike VMWare, VirtualPC has not been positioned as a server product, even under Microsoft management.
If it was integrated in, there's some interesting security stuff they could do -- you could 'sandbox' Win32 apps without breaking the API. That would minimize embarssing mail worms, spyware and so on. Sorta like 'jails' in the Unix world.
I think they also need an x86 emulator for Itanium systems, but I'm not sure if they are going to use Virtual PC or something else.
That might be true, but the major uses of VirtualPC (for PCs) that I've heard about are:
+ MS-DOS for old games. (Connectix even marketed this) + OS/2 for legacy applications.
I think most Linux users of virtualization prefer Linux as the host OS and Windows as guest with VMWare, since they are usually migrating from Windows -> Linux and not the other way around.
Well, as you pointed out, their competitors like Netscape were selling products that could be cracked by anyone, while partial-escrow was only reasonably crackable by the US Government.
All in all, I think that IBM thought the export regs are as stupid as you and I do, and I can see how a foreign government might have reasonably decided that US software wasn't the best option. However, that doesn't mean that there was any big secret there, except for the sake of political scandal.
OK, it sounds like Netscape didn't go with the escrow plan.(http://www.networkcomputing.com/704/704f3mai n.html -- historical information on the WWW, what a stroke of fate).
Microsoft did distribute seperate versions of their products until the restrictions were lifted.
> The big issue was that IBM/Lotus didn't tell their non-US clients about the escrow until they were found out.
That's incorrect. It was documented, publicized in tech rags, and even sold as a feature for international customers. It became an issue maybe because IBM oversold the product, but probably mainly for political reasons. (Both international politics and some typical My Vendor Can Beat Up Your Vendor stuff,)
even makes the interesting case that these folks aren't really for Linux as they are against a whole mound of stuff. Linux just happens to be something they can latch onto
3-4 years ago, Slashdot was certainly biased toward "Pro-Linux Zealots". Young, brash, and full of ideas, "World Domination", the revolution was NOW, Linux was the future. Windows was barely worth mentioning.
Over the years, however, I think the balance has shifted away from "Pro-Linux" to "Anti-Microsoft Zealots". Maybe this was because the revolution wasn't as immediate and swift as expected. Maybe the zealots matured into Priests. Maybe it was the editorial stance that publishes every MS newsbit it can find, or just that Linux was the latest rallying point for the ABM crowd.
Anyway, all of a sudden you get the disaffected OS/2, Amiga, and BeOS lusers jumping on the bandwagon. And hey, the great thing about being an anti-Microsoft zealot is that you don't even really have to stand FOR anything. You can sit there from XP and IE6 and tell the world how terrible Microsoft is, with the aura of inevitable failure for your cause. And I'd disagree that everyone in that crowd is young - some of them have been doing the same online act for years.
This encourages a lot of dull "zero sum" thinking. It's not enough that your side is winning, the other side has to be losing. Which in total lowers the quality of debate.
Note that Netscape and Microsoft also gave the NSA part of their SSL keys as well. So this situation was not unique to Lotus and was mandated by US government regulation for "export" products. According to IBM/Lotus, there was no key escrow for the North American version.
However, that was years ago -- US corporations can now export the full encryption versions of their products.
When a new security hole in a Microsoft product is posted on Slashdot, these people will start talking about how holes are only discovered "because Windows is more popular than Linux" or similar, which of course is an old and tired claim, especially considering the fact that open-source product Apache, which is more widely used than any other web server, has had far fewer security holes than Microsoft's IIS.
Well, I don't want to get into it, but I think it's well known that the *ongoing* problems with IIS and MS-SQL worms come not from professionally managed servers. but from unpatched client systems running this software, which wouldn't be included in any server survey. (I'll withhold any assessment of how secure these products are or aren't.)
I've probably seen the Netcraft survey misinterpreted a dozen different ways on this board. It counts Internet domains, not webservers, software installations, or just "servers". It's this misuse of statistics that shows a taint of zealotry, or at least uncritical thinking. But, hey, there's a reason that the most popular troll starts with "Netcraft Confirms It....".
But now, I've refused to accept your assertion about Apache, and according to your post, I'm spreading Microsoft FUD. Yeah, that really elevated the debate.
Obviously, I've been around slashdot for a while and I've picked up some of the unspoken rules:
1) Pro-Linux points don't need to be substantiated as well as Pro-Microsoft points. This occurs both at the top end of the comment scale and the bottom end. So, if you are going to be not totally critical of Microsoft products, get your facts straight.
1a) The corollary of this that Pro-MS posts are the easiest way to troll slashdot. I've seen one liner "Windows XP is the best" AC posts that hook a dozen Linux 'zealots'.
2) Popularity arguments aren't very fruitful. If the core readership accepted the fact they were outnumbered 99-1, they wouldn't be here. (You see this a lot with Mac or Gnome topics too.) Best bet is to just avoid them.
3) Anything someone doesn't like to hear is "FUD" or a "troll". This is slashdot's equivalent of Godwin's Law and should be ignored as noise, except for the fact that people can vote with the moderation system.
The point of UNIX is that you could toss that SparcStation and go buy hardware from a vendor that documents their framebuffer (haha), with fairly low software migration costs.
At least that's how the UNIX vendors sold it.
Sun made lots of money selling "Open Systems" against DEC and IBM in their day, but now they are watching their customers switch seemlessly to Linux. So, they weren't entirely lying.
The overall integration between the "Mac side" and the "Unix side" is pretty poor on OS X, in my book.
Either you have Mac Carbon/Cocoa programs that have virtually zero Unix integration, or you've got Unix programs that have the really rough, incomplete GUI front-ends typically found on a Unix system.
Just compare the functionality Windows "MMC" GUI for IIS versus the lame Start/Stop button for OS X's Apache server. A sysadmin might prefer the OS X/Unix config file approach, but regular users can't make head-nor-tail of getting Apache configured. It's kinda sad to see Microsoft out-Mac the Mac by providing a high-level of GUI integration, while Apple runs against 20 years of their advertising by telling people to go to the command line.
It's probably easier to think of MacOS X as a proprietary OS design that happens to have a Unix-compatibility module duct-taped on the side.
As a Windows workstation, it was pretty much worthless because other than MS Office and some CAD software, there weren't any native applications. Everything had to be run through emulation, which killed the price/performance.
As a Windows server, the Alpha had much better software support -- almost every major Windows server package ran on it. My company at the time was considering buying them, but DEC's salesmen were too incompetent to get us a working demo machine.
As enkidu said, the UNIX software situation wasn't that great either. The DEC was used by vendors for benchmarks, but always was a "Tier 2" platform.
Eventually they did release a $1200 "entry level" Alpha called the Multia, but those didn't sell well either. Eventually they became popular among Linux users scrounging the clearance deals.
DR's Gary Kildall sat down at an IBM PC supplied by IBM and, using a secret code, got it to pop up a Digital Research copyright notice.
It's case won, Digital Research received monetary compensation and the right to clone MS-DOS. This is why Microsoft never sued DR over DR-DOS, but used every other means to destroy it. The settlement was under a strict non- disclosure agreement, so few even know DR sued, never mind that they won.
Now, this is third hard information, but the story was repeated in various print publications of the day as well as histories of Microsoft.
Just last year I was working with a company that had just migrated off "Q&A" -- they were running their sales database off it on W2000. (Fortunately, Q&A could export to DBase format, so they were able to get their data out).
This practice always bothered me. Just because they sent a form into the SEC doesn't automatically make their name meaningless, at least in terms of ordinary conversation.
I think the reasoning behind these changes is based on the abbreviation style used by the New York Times and Wall Street Journal.
For example, AT&T was until recently the American Telephone and Telegraph Corporation, which was abbreviated "A.T.&T.", or worse, spelled out as "American Telephone". Those extra periods bother the marketing types, so they rename the corporation to ensure that their brandname is consistently represented.
(Even worse for marketing, I think the NYT would use "Sco" instead of "S.C.O." because most people pronounce it as a word.)
Back in the day, Apple computers were loaded with custom chips that gave them unique capabilities. The downside to this design was that it limited Apple's ability to manufacture machines.
So, they basically had more potential customers than they had computers. There's two ways they could deal with this situation:
a) Move to an 'open' architecture and bring in 3rd party manufacturing b) Keep raising prices until the demand curve falls off.
Scully chose Plan B, which pretty much permenently doomed them to a nitch player. The upside is that their profits were so high that they built that $4 Billion bank account that people are always talking about. Apple is really more of a mutual fund now days than a computer manufacturer.
There's a history of Apple by Jim Carlton that covers the decision not to allow 'cloning' in great detail.
I disagree, it was more like an early version of Visual Basic
I'd agree -- Except that Apple never really positioned HyperCard very well, so it primarily got used as a "toy" educational language.
Meanwhile, VB was aimed directly at corporate programming and had support for DB Access, "grids" and so on. This was painfully obvious at the job I had 10 years ago where several thousand Macs were dumped because they didn't make a good "client-server" platform.
And now Sun and Linux devolve into a sad war of FUD and Counter-FUD...
It has been explitictly shown that they are funding them.
Sun was in a pretty shitty situation. They have been telling everyone for years that they have "full rights" to their UNIX. Except when they took a closer look, they really didn't. Oops.
SCO owned some drivers and other bits. Oh, and SCO was just about to go insane. So they paid them off in order to make their claims of "full rights" and "indemnification" retroactively true. They tried to put a good face on it by getting some stock options and calling it an investment.
Note that having SCO around doesn't really help Sun at all in the long term. IBM specifically got involved with SCO System V back in the 90s (Project Monterey) because customers were asking for something "more like Solaris". If/When SCO finally tanks, Sun will be pretty much the sole owner of UNIX.
I give the author a lot of credit for addressing the philsophical underpinnings of the Unix Versus Windows platform conflict. Too frequently, the "IT guys" portray this as a simple matter of switching the software stack, when in fact they are advocating a radically different technology management ideology.
The author at least makes an attempt to address this idelogical difference from a practical, if biased, perspective. However, by putting it in the context of "the MCSE", it's skirting around the greater management issues involved.
I though this point in particular was facinating:
Use of the Unix Business Architecture eliminates all of these problems:
First it removes the user empowerment lie fundamental to the client-server architecture by eliminating the pretence that the desktop is a personal, rather than corporate, resource and thus positions the IT group for an honest working relationship with users.
Does anyone see clamoring for a "honest relationship" from the End User side? If anything IT has moved to more of a "service" role with users as it's "customers" -- especially in executive-heavy US corporate culture. A change in platforms is not going to reinstate a great "honest" reverence for the almighty BOFH power-relationship.
Furthermore, it ignores the reason that corporate computing moved to the "Personal" model to begin with -- IT departments were widely seen as not being flexible enough to provide the solutions that end users needed in the trenches. Sure, there's a massive cost savings with the Host-Terminal model, but there's also a large opportunity cost associated with it, and that's fundementally a high-level business decision.
Slashdot Mirror
In my book, a bigger problem is that the OS X GUI is "cake frosting" over a much more complex underlying system. There's endless system subdirectories filled with mystery files -- some hidden from the Finder. There's certain tasks where the user MUST drop to the command line. The user is much more dependent on installers.
That leads to a system that's significantly more complex to troubleshoot and support than Classic MacOS. No more drag-n-drop of Extension icons, no more Space-Boot, etc. It brings about a certain fear factor among the normal Mac userbase.
And that removes the primary market advantage that MacOS held. Once the system becomes as 'difficult' as Windows or Unix, then all you've got left is lickable graphics.
Or atleast hijack Mozilla/etc.
Yeah. Like IE, Mozilla's practically designed for such a thing. Just look at all the crazy stuff on the "Firebird Extensions" page that installs in 2 clicks.
Were there a substantial userbase on Mozilla, it would be pretty easy to use social engineering and get folks to install something malicious.
The current ones have hardware emulation, yes. But I believe that Intel has announced that feature is going away because you'll get better performance with software.
I can develop and test n-tier applications on my laptop while sitting on the bus if I want.
It's also useful for stuff like COM development, so you can blow away you environment and not have to worry about all the fragile registration stuff. However, they use VMWare for this at work.
Uh, who are these people? Did you read that in a marketing brouchure?
The NT4 legacy users I've seen continue to run NT4 on bare metal. Also, unlike VMWare, VirtualPC has not been positioned as a server product, even under Microsoft management.
If it was integrated in, there's some interesting security stuff they could do -- you could 'sandbox' Win32 apps without breaking the API. That would minimize embarssing mail worms, spyware and so on. Sorta like 'jails' in the Unix world.
I think they also need an x86 emulator for Itanium systems, but I'm not sure if they are going to use Virtual PC or something else.
That might be true, but the major uses of VirtualPC (for PCs) that I've heard about are:
+ MS-DOS for old games. (Connectix even marketed this)
+ OS/2 for legacy applications.
I think most Linux users of virtualization prefer Linux as the host OS and Windows as guest with VMWare, since they are usually migrating from Windows -> Linux and not the other way around.
Well, as you pointed out, their competitors like Netscape were selling products that could be cracked by anyone, while partial-escrow was only reasonably crackable by the US Government.
All in all, I think that IBM thought the export regs are as stupid as you and I do, and I can see how a foreign government might have reasonably decided that US software wasn't the best option. However, that doesn't mean that there was any big secret there, except for the sake of political scandal.
OK, it sounds like Netscape didn't go with the escrow plan.(http://www.networkcomputing.com/704/704f3mai n.html -- historical information on the WWW, what a stroke of fate).
Microsoft did distribute seperate versions of their products until the restrictions were lifted.
> The big issue was that IBM/Lotus didn't tell their non-US clients about the escrow until they were found out.
That's incorrect. It was documented, publicized in tech rags, and even sold as a feature for international customers. It became an issue maybe because IBM oversold the product, but probably mainly for political reasons. (Both international politics and some typical My Vendor Can Beat Up Your Vendor stuff,)
even makes the interesting case that these folks aren't really for Linux as they are against a whole mound of stuff. Linux just happens to be something they can latch onto
3-4 years ago, Slashdot was certainly biased toward "Pro-Linux Zealots". Young, brash, and full of ideas, "World Domination", the revolution was NOW, Linux was the future. Windows was barely worth mentioning.
Over the years, however, I think the balance has shifted away from "Pro-Linux" to "Anti-Microsoft Zealots". Maybe this was because the revolution wasn't as immediate and swift as expected. Maybe the zealots matured into Priests. Maybe it was the editorial stance that publishes every MS newsbit it can find, or just that Linux was the latest rallying point for the ABM crowd.
Anyway, all of a sudden you get the disaffected OS/2, Amiga, and BeOS lusers jumping on the bandwagon. And hey, the great thing about being an anti-Microsoft zealot is that you don't even really have to stand FOR anything. You can sit there from XP and IE6 and tell the world how terrible Microsoft is, with the aura of inevitable failure for your cause. And I'd disagree that everyone in that crowd is young - some of them have been doing the same online act for years.
This encourages a lot of dull "zero sum" thinking. It's not enough that your side is winning, the other side has to be losing. Which in total lowers the quality of debate.
Note that Netscape and Microsoft also gave the NSA part of their SSL keys as well. So this situation was not unique to Lotus and was mandated by US government regulation for "export" products. According to IBM/Lotus, there was no key escrow for the North American version.
However, that was years ago -- US corporations can now export the full encryption versions of their products.
The primary reason for that is because WMP is using the overlay hardware in your video card. DRM is just a side-benefit.
Try minimizing then maximizing the player -- you can sometimes get a screenshot that way.
When a new security hole in a Microsoft product is posted on Slashdot, these people will start talking about how holes are only discovered "because Windows is more popular than Linux" or similar, which of course is an old and tired claim, especially considering the fact that open-source product Apache, which is more widely used than any other web server, has had far fewer security holes than Microsoft's IIS.
Well, I don't want to get into it, but I think it's well known that the *ongoing* problems with IIS and MS-SQL worms come not from professionally managed servers. but from unpatched client systems running this software, which wouldn't be included in any server survey. (I'll withhold any assessment of how secure these products are or aren't.)
I've probably seen the Netcraft survey misinterpreted a dozen different ways on this board. It counts Internet domains, not webservers, software installations, or just "servers". It's this misuse of statistics that shows a taint of zealotry, or at least uncritical thinking. But, hey, there's a reason that the most popular troll starts with "Netcraft Confirms It....".
But now, I've refused to accept your assertion about Apache, and according to your post, I'm spreading Microsoft FUD. Yeah, that really elevated the debate.
Obviously, I've been around slashdot for a while and I've picked up some of the unspoken rules:
1) Pro-Linux points don't need to be substantiated as well as Pro-Microsoft points. This occurs both at the top end of the comment scale and the bottom end. So, if you are going to be not totally critical of Microsoft products, get your facts straight.
1a) The corollary of this that Pro-MS posts are the easiest way to troll slashdot. I've seen one liner "Windows XP is the best" AC posts that hook a dozen Linux 'zealots'.
2) Popularity arguments aren't very fruitful. If the core readership accepted the fact they were outnumbered 99-1, they wouldn't be here. (You see this a lot with Mac or Gnome topics too.) Best bet is to just avoid them.
3) Anything someone doesn't like to hear is "FUD" or a "troll". This is slashdot's equivalent of Godwin's Law and should be ignored as noise, except for the fact that people can vote with the moderation system.
The point of UNIX is that you could toss that SparcStation and go buy hardware from a vendor that documents their framebuffer (haha), with fairly low software migration costs.
At least that's how the UNIX vendors sold it.
Sun made lots of money selling "Open Systems" against DEC and IBM in their day, but now they are watching their customers switch seemlessly to Linux. So, they weren't entirely lying.
Nobody said UNIX couldn't be proprietary!
Actually, the entire UNIX market says it's not proprietary. (well, except for SCO)
Open Standards, independent certification, multiple vendors, etc.
Yeah, I meant to note that MMC is a bloated piece of shit, and study of bad UI design with it's right-clicks and it's mega-tabbed dialogs.
However, I think the point stands that for casual Mac-type users, some GUI is better than no GUI.
The overall integration between the "Mac side" and the "Unix side" is pretty poor on OS X, in my book.
Either you have Mac Carbon/Cocoa programs that have virtually zero Unix integration, or you've got Unix programs that have the really rough, incomplete GUI front-ends typically found on a Unix system.
Just compare the functionality Windows "MMC" GUI for IIS versus the lame Start/Stop button for OS X's Apache server. A sysadmin might prefer the OS X/Unix config file approach, but regular users can't make head-nor-tail of getting Apache configured. It's kinda sad to see Microsoft out-Mac the Mac by providing a high-level of GUI integration, while Apple runs against 20 years of their advertising by telling people to go to the command line.
It's probably easier to think of MacOS X as a proprietary OS design that happens to have a Unix-compatibility module duct-taped on the side.
As a Windows workstation, it was pretty much worthless because other than MS Office and some CAD software, there weren't any native applications. Everything had to be run through emulation, which killed the price/performance.
As a Windows server, the Alpha had much better software support -- almost every major Windows server package ran on it. My company at the time was considering buying them, but DEC's salesmen were too incompetent to get us a working demo machine.
As enkidu said, the UNIX software situation wasn't that great either. The DEC was used by vendors for benchmarks, but always was a "Tier 2" platform.
Eventually they did release a $1200 "entry level" Alpha called the Multia, but those didn't sell well either. Eventually they became popular among Linux users scrounging the clearance deals.
Now, this is third hard information, but the story was repeated in various print publications of the day as well as histories of Microsoft.
Just last year I was working with a company that had just migrated off "Q&A" -- they were running their sales database off it on W2000. (Fortunately, Q&A could export to DBase format, so they were able to get their data out).
So, yes, some people really do use the DOS Box.
SCO doesn't stand for anything here
This practice always bothered me. Just because they sent a form into the SEC doesn't automatically make their name meaningless, at least in terms of ordinary conversation.
I think the reasoning behind these changes is based on the abbreviation style used by the New York Times and Wall Street Journal.
For example, AT&T was until recently the American Telephone and Telegraph Corporation, which was abbreviated "A.T.&T.", or worse, spelled out as "American Telephone". Those extra periods bother the marketing types, so they rename the corporation to ensure that their brandname is consistently represented.
(Even worse for marketing, I think the NYT would use "Sco" instead of "S.C.O." because most people pronounce it as a word.)
Back in the day, Apple computers were loaded with custom chips that gave them unique capabilities. The downside to this design was that it limited Apple's ability to manufacture machines.
So, they basically had more potential customers than they had computers. There's two ways they could deal with this situation:
a) Move to an 'open' architecture and bring in 3rd party manufacturing
b) Keep raising prices until the demand curve falls off.
Scully chose Plan B, which pretty much permenently doomed them to a nitch player. The upside is that their profits were so high that they built that $4 Billion bank account that people are always talking about. Apple is really more of a mutual fund now days than a computer manufacturer.
There's a history of Apple by Jim Carlton that covers the decision not to allow 'cloning' in great detail.
I disagree, it was more like an early version of Visual Basic
I'd agree -- Except that Apple never really positioned HyperCard very well, so it primarily got used as a "toy" educational language.
Meanwhile, VB was aimed directly at corporate programming and had support for DB Access, "grids" and so on. This was painfully obvious at the job I had 10 years ago where several thousand Macs were dumped because they didn't make a good "client-server" platform.
And now Sun and Linux devolve into a sad war of FUD and Counter-FUD...
It has been explitictly shown that they are funding them.
Sun was in a pretty shitty situation. They have been telling everyone for years that they have "full rights" to their UNIX. Except when they took a closer look, they really didn't. Oops.
SCO owned some drivers and other bits. Oh, and SCO was just about to go insane. So they paid them off in order to make their claims of "full rights" and "indemnification" retroactively true. They tried to put a good face on it by getting some stock options and calling it an investment.
Note that having SCO around doesn't really help Sun at all in the long term. IBM specifically got involved with SCO System V back in the 90s (Project Monterey) because customers were asking for something "more like Solaris". If/When SCO finally tanks, Sun will be pretty much the sole owner of UNIX.
The author at least makes an attempt to address this idelogical difference from a practical, if biased, perspective. However, by putting it in the context of "the MCSE", it's skirting around the greater management issues involved.
I though this point in particular was facinating:
Does anyone see clamoring for a "honest relationship" from the End User side? If anything IT has moved to more of a "service" role with users as it's "customers" -- especially in executive-heavy US corporate culture. A change in platforms is not going to reinstate a great "honest" reverence for the almighty BOFH power-relationship.
Furthermore, it ignores the reason that corporate computing moved to the "Personal" model to begin with -- IT departments were widely seen as not being flexible enough to provide the solutions that end users needed in the trenches. Sure, there's a massive cost savings with the Host-Terminal model, but there's also a large opportunity cost associated with it, and that's fundementally a high-level business decision.