Considering this might break login and other admin scripts, be extra sure you want to do this. If you administrate a large number of Windows machines you've just made your life potentially much more difficult.
Besides, it would be trivial to convert your typical Outlook virus into JavaScript, PerlScript, or even an VB EXE file. NOTAFIX.
Microsoft has had a security patch out which mitigates the problem for many months. Have you tried it?
The Lotus Notes wire protocol dates back to the mid-80s and has about 10 layers of cruft on top and at least appears to be highly dependant on Notes' document database format. Reverse Engineering it would be unlikely at best -- Probably better to just write a wrapper around the portable Notes C libraries and pay IBM their client seat $ (which if you are using Notes, you probably are doing anyway).
Reverse engineering Exchange is probably just as icky. The first step would to create a free OSF DCE-compatible (and therefore MS-compatible) RPC library. Some Samba folks might be working on this, but nothing yet. Then you would need to figure out what ungodly pipe-filling garbage Exchange/Outlook uses to talk.
It's not my time and effort, but I wouldn't bother. It would be much more effective to build an open source replacement for Exchange/Domino, in particular the calendar and the mail routing functionality. Most of the groupware apps could be better done on a web system with mySQL or whatever. --
It depends. I've worked in environments that get along just fine with the "Meeting at 9AM Monday" e-mail. However, if you are using some program with scheduling, it more polite to send a meeting invitation because the reciever gets an automated notification and they can just push "I Accept" or whatever, and not have to manually update whatever schedule book or software they're using.
There's also the issue of getting executive types to show up at these things, and that almost always demands sending some sort of invitation. Even back in the old days, we'd have to fire up a 3270 emulator and send IBM PROFS invitations to get certain people to show. (We were all happily on a POP3 system.)
There's definately different philosophies at play. Linus somehow manages the whole Linux Kernel project out of his Inbox, but I've found that my life is much easier with groupware tools and much less e-mail. --
There no such thing as a "generic" calendar solution, period.
Sure there's RFCs 2445-7 (iCalendar), but who the hell has product support? The fact is, if you are developing a scheduling application, you pretty much only have a choice between Outlook (Windows and sorta Mac) and Lotus Notes (Windows and Mac and sorta Unix). That's an ugly reality, and sitting on your butt waiting for a open platform is a stupid business decision, because you'd have been waiting for about 15 years now.
It's also pretty much impossible to create "generic" workflow mail routing applications either (no RFCs that I know of!), etc, etc for other groupware type stuff. --
A few years ago, Netscape made a big "enterprise" sales push for it's messaging/calendar platforms. It just so happened that Microsoft and IBM/Lotus were also having a huge sales war at the same time (mostly because there was millions of old ccMail and MS-Mail seats out there ripe for the Y2K picking).
Anyway, Netscape came in a distant 3rd (or perhaps even 4th behind Novell) in that fight, which is one of the big reasons they aren't an independant company anymore.
The platform pretty much requires Netscape Communicator clients, and it's fuzzy how some things like the Calendar will work with Mozilla. Maybe AOL doesn't want to force Netscape onto their own employees.. --
Coleco Adam (sold standalone or as ColecoVision Expansion Module #3)
The "you can turn it into a computer" logic seemed to appeal to my parents too, even though we had a funky greenscreen Apple ][+. The Adam shipped a little too late and with too few features (such as pricy highspeed tapes instead of disks), and never really could compete with Commodore and Atari. --
In the early days, Beta only held 90 minutes (which worked out to 80-something), while VHS held 120m. Well, no shit, you might say, the VHS tape shell was much bigger.
Anyway, that meant that a large percentage of movies would fit on 1 VHS tape but required 2 Beta tapes. Which is not "technically superior" from the movie studio point of view.
Not to mention that Beta's visual superiority wasn't all that apparent on the shitbanger early-80s VCRs and people's bluratron early-80s TVs, and the crappy mastering jobs studios tended to do back then. So it was essentially a non-feature to most users, and there was always LaserDisc, which was/is superior to both tape formats, if you happend to be the sort who cares.
Although, I would agree that VHS's "won" on the strength of it's manufacturing model , just that back then, Beta and VHS weren't as differenced as people's histories tell it (as in, it wasn't a question of Cost versus Quality). --
Bay Area -- Feinstein, despite being a local hero, is absolutely not a friend to the local tech industry. Time after time, she has sided with law enforcement and hollywood factions against a rational encryption policy, and appears to get most of her money from down south. I have no idea what Boxer is up to, but I take it that she's there because people believe her heart's in the right place.
The guy who ran against Feinstein, Tom Campbell, (he was too much of a maverick to have any chance of winning) was pretty sharp. Stanford prof and congressman from the SV area, and very up on the anti-trust issues with Microsoft and encryption policy. Too bad he's out of government now. There's various other industry functionaries that have been elected to certain posts, but I would say that's more "chamber of commerce" than "technically saavy". --
The reason Microsoft can say that "open source != secure" is the EXACT SAME logic that you are using when you say "OpenBSD==secure" (how? why? doing what? who's the admin? etc?)
People want simple answers to a complex problem. You're playing the same advocacy game that Microsoft is. --
The fact is that ANY business application of any software requires an expert (in-house or consultant) before people should roll it out.
You may think that and be correct. But the fact remains that many small NT shops limp along OK without expert assistatnce, and maybe a little tech support from the guy who sold them the comptuer. You can slap Great Plains on an NT box and have an accounting system.
These small businesses will never pay a Unix admin the money he's worth, nor would they pay a good NT guy the money either. And retention would be a problem in an environment where the guy is rehabing old 486s and refilling the copy machine toner. So, they do the rational thing and limp along with part time or slightly retarded computer help.
Open source software + contracted service would be a great solution for small businesses with no inhouse experience, and cheaper than paying MS licences. However the customer base is never going to pick up the phone and sign a service contract with RedHat or IBM (nor should they). The only real answer is that it's going to take a phalanx of Linux saavy people in the small system integrators that are out there.
And that I don't see. It's easier for the corner comptuer guy to build a computer and slap MS Small Business server or Exchange on it and send the money back to Microsoft. If he could find a skilled Linux admin to hire, the guy is going to figure that he's got better things to do than screwing in IDE drives in his spare time and go and make $90K doing Unix admin for a big corporation.
(As a sidenote, Novell essentially built their business on these Corner Computer Store integrators, but it took a massive channel push and lots of product education. So, it's possible, but it's a long haul for someone.) --
Considering they amount of money spent on development of NT 3.1, 3.5, and 4.0, and the anemic rate of adoption of those products, Windows 2000 isn't looking so bad.
Nobody in the history of personal computers has successfully sold a "modern" desktop operating system to the public. Xenix didn't sell, OS/2 1.x didn't sell, OS/2 2.x didn't sell, and NT hasn't sold well either, and Linux on the desktop is as of yet a pipedream. Considering that sad history, if I was Fat Ballmer, I wouldn't be sweating the W2K sales too hard. --
MS-DOS never ran on Z-80s. Those things probably ran CP/M. Maybe you're talking about MSX machines, but I doubt you know.
BTW, you obviously never dealt with the old IBM. Their "philosophic-nearly-religious" convictions were equal to what you see out of Microsoft today. Shit like OS/2, PS/2, and SAA were the height of their arrogance. Microsoft had the choice to remain as a subcontractor to that death star or move on, and as shitty as they are, the fact that they moved on was a great thing for personal computing in general. --
Office was never playing catch up. Both Word and Excel were best of class programs 12 years ago.
In fact, as you mention, most of the new features have made them much worse products for the mid-tier to experienced user. This means that there's a greater opportunity NOW for a challenger (say StarOffice) to underprice and outperform the marketleader than at any time since Microsoft stole market from Lotus and the Perpetually Sucky WordPerfect. --
Realistically, nobody but Microsoft and The Gartner Group expected mass conversions of corporate desktops to W2K within the first 2 years. And considering the history of NDS, minimal ActiveDirectory adoption isn't a big shock either.
So to say that "hasn't done very well" all depends on your expectations. From my POV, it's done extremely well in attracting power users and frustrated Win9x users. Obviously if you have a stable NT4 network (and most don't), there's little benefit, but if MS was honest they would have admitted that. (As they seem to be doing for WinXP vs. Win2K.) --
1987. That was when IBM was trying force it's POS 286-specific OS (OS/2) and it's proprietary PC hardware (PS/2) down everyone's throats as an end-to-end solution to sell more mainframes.
Thank god Microsoft bucked them. As bad as things are, we're better off than if IBM would have won that one. --
I think Linux has a real problem, and that is the elitist attitude found among developers. "Why don't they just read the man pages", etc. There is a real market for a distribution which can easily replace a newbie's MS Windows OS, however, it seems that no one wants to build it. Why? The geeks would rather keep it for themselves.
There's nothing wrong with an elitist attitude and wanting to "keep it for yourself".
The problem is the intellectual dishonesty involved in ripping on Microsoft (and Apple) because they make a consumer-level product while steadfastly refusing to adapt Unix to the consumer market.
It's a "Put up or Shut up" situation, and Unix community's reaction is to do neither, even though Unix OSes have the capability to be better than Windows, both technically and in terms of user interface. --
Do you know of any hardware that uses the same driver for both 9x and 2000? I think that WDM is still a pie-in-the-sky. There's a huge segment of non-name and low-end hardware that still doesn't run on 2000 and never will.
Windows 95 was specifically designed to be back-compatible with some very old drivers. XP doesn't have that luxury. Microsoft's solution is to market the product primarily for new OEM installs and not upgrades, but that doesn't solve the labelprinter issue. --
(1) SW vendors will just "require" their software be installed and run with administrator right.
Hell, even Microsoft was doing this until a couple years ago, so it's no wonder the OEMs are busy supporting only Win9x. It's amazing how poorly MS has historically supported their own OS (NT).
MS will cave into pressure from developpers wanting unrestricted access to resources and users wanting to run their old programs/games that MS will create "Windows Classic", fourth in the 95/98/ME line of single user OSes.
Wouldn't shock me one bit. Recall that both NT4 and Win2000 were supposed to be the "unified system", and then MS always squeezed out another version of DOSWin.
The other alternative is that OEMs give the finger to MS and continue to ship WinME. If a large variety of crap off-the-shelf software and crap hardware (label printers, business card scanners, that ilk) won't run under WinXP, it will drive support costs through the fucking roof.
Don't forget that Win95 shipped being compatible with 98% of Windows 3.1 programs and compatible with around 70-80% of DOS/Win3.1 drivers. That made the conversion relatively painless. Can't say that's going to happen with WinXP. (Although I really wish MS would go nazi and make it happen.) --
There has actually been numerous lawsuits and fed investigations of the car stereo situation. So far, Detroit has gotten away with what they've been doing (for example, the oval-shaped stereo in Tauruses), but don't oversimplify the situation. --
IBM shouldn't have had to do an 'internal audit' to find that out -- it's right in their financial reports. And, good thing too, because Lou Gerstner basically saved the company by focusing on services.
The thing about consulting is that, unlike software or hardware, it only scales linearly - to double your revenue, you generally need to double your consulting force. That means there's a practical limit to profits that can be made in a particular segement. So, IBM is raking in the dough doing Windows 2000 migraiton consutlting, but they need to expand their offerings with things like Linux and Java.
However, if you've dealt with IBM recently, you'll know that services ain't their main cup of tea yet. They primarily use it as part of a sales bundle to move software or hardware. (Such as, you get a cut on these boxes and licences, just give us the support contract.)
They are thinking that Linux is an especially good sales tool against Sun accounts (Linux/Netfinity or Linux/S390 Cluster on the low end, maybe move some high profit RS/6000s on the high end.) However, their profit margin is MUCH higher on the classic IBM stuff (RS/6000, AS/400), so to a great degree Linux is just a door opener for them. --
Mac Word was a great program, and I guess I get along because it still hiding somewhere beneth 2GB of do-dads in modern versions of Word.
Forget about MacPaint -- the program I really miss from the old Mac was SuperPaint (a combination of MacPaint and MacDraw with a few other features like layers). It's virutally impossible to find a nice low-end usable progam like this anymore. Everythings big-n-bloated (CorelDraw, Photoshop), or retarded (MS graphics progs).
My comment about business trends goes to the fact that Microsoft is redefining their strategy. They're raising prices on the brink of an IT spending recession, and it's pretty certain that they think they know what they are doing, but it's not clear anyone else does. Picking winners in the next few years is going to be tough. --
Re:you've fallen for MS strategy
on
Mozilla 0.9 Out
·
· Score: 1
Apologies -- I was looking at the lack of activity on http://bugzilla.mozilla.org/show_bug.cgi?id=18722 . Look forward to trying Moz XSLT out. --
I read somewhere quite a while back, that Windows (the 95 family) is a loss leader. It could be in Microsoft's best interests to let Linux win.
Microsoft makes $Billions/year on Windows, and most of that is probably from Windows 9x. There's no way it's a loss-leader. In fact, it's one of the handful of products they make that's profitable.
(Which is not to say that there aren't markets where MS sells below cost. Apparently for the low-end home machines you see at electronics stores, Windows is sold to the OEMs at a very steep discount.) --
Slashdot Mirror
Considering this might break login and other admin scripts, be extra sure you want to do this. If you administrate a large number of Windows machines you've just made your life potentially much more difficult.
Besides, it would be trivial to convert your typical Outlook virus into JavaScript, PerlScript, or even an VB EXE file. NOTAFIX.
Microsoft has had a security patch out which mitigates the problem for many months. Have you tried it?
--
The Lotus Notes wire protocol dates back to the mid-80s and has about 10 layers of cruft on top and at least appears to be highly dependant on Notes' document database format. Reverse Engineering it would be unlikely at best -- Probably better to just write a wrapper around the portable Notes C libraries and pay IBM their client seat $ (which if you are using Notes, you probably are doing anyway).
Reverse engineering Exchange is probably just as icky. The first step would to create a free OSF DCE-compatible (and therefore MS-compatible) RPC library. Some Samba folks might be working on this, but nothing yet. Then you would need to figure out what ungodly pipe-filling garbage Exchange/Outlook uses to talk.
It's not my time and effort, but I wouldn't bother. It would be much more effective to build an open source replacement for Exchange/Domino, in particular the calendar and the mail routing functionality. Most of the groupware apps could be better done on a web system with mySQL or whatever.
--
It depends. I've worked in environments that get along just fine with the "Meeting at 9AM Monday" e-mail. However, if you are using some program with scheduling, it more polite to send a meeting invitation because the reciever gets an automated notification and they can just push "I Accept" or whatever, and not have to manually update whatever schedule book or software they're using.
There's also the issue of getting executive types to show up at these things, and that almost always demands sending some sort of invitation. Even back in the old days, we'd have to fire up a 3270 emulator and send IBM PROFS invitations to get certain people to show. (We were all happily on a POP3 system.)
There's definately different philosophies at play. Linus somehow manages the whole Linux Kernel project out of his Inbox, but I've found that my life is much easier with groupware tools and much less e-mail.
--
There no such thing as a "generic" calendar solution, period.
Sure there's RFCs 2445-7 (iCalendar), but who the hell has product support? The fact is, if you are developing a scheduling application, you pretty much only have a choice between Outlook (Windows and sorta Mac) and Lotus Notes (Windows and Mac and sorta Unix). That's an ugly reality, and sitting on your butt waiting for a open platform is a stupid business decision, because you'd have been waiting for about 15 years now.
It's also pretty much impossible to create "generic" workflow mail routing applications either (no RFCs that I know of!), etc, etc for other groupware type stuff.
--
A few years ago, Netscape made a big "enterprise" sales push for it's messaging/calendar platforms. It just so happened that Microsoft and IBM/Lotus were also having a huge sales war at the same time (mostly because there was millions of old ccMail and MS-Mail seats out there ripe for the Y2K picking).
Anyway, Netscape came in a distant 3rd (or perhaps even 4th behind Novell) in that fight, which is one of the big reasons they aren't an independant company anymore.
The platform pretty much requires Netscape Communicator clients, and it's fuzzy how some things like the Calendar will work with Mozilla. Maybe AOL doesn't want to force Netscape onto their own employees..
--
Coleco Adam (sold standalone or as ColecoVision Expansion Module #3)
The "you can turn it into a computer" logic seemed to appeal to my parents too, even though we had a funky greenscreen Apple ][+. The Adam shipped a little too late and with too few features (such as pricy highspeed tapes instead of disks), and never really could compete with Commodore and Atari.
--
In the early days, Beta only held 90 minutes (which worked out to 80-something), while VHS held 120m. Well, no shit, you might say, the VHS tape shell was much bigger.
Anyway, that meant that a large percentage of movies would fit on 1 VHS tape but required 2 Beta tapes. Which is not "technically superior" from the movie studio point of view.
Not to mention that Beta's visual superiority wasn't all that apparent on the shitbanger early-80s VCRs and people's bluratron early-80s TVs, and the crappy mastering jobs studios tended to do back then. So it was essentially a non-feature to most users, and there was always LaserDisc, which was/is superior to both tape formats, if you happend to be the sort who cares.
Although, I would agree that VHS's "won" on the strength of it's manufacturing model , just that back then, Beta and VHS weren't as differenced as people's histories tell it (as in, it wasn't a question of Cost versus Quality).
--
Bay Area -- Feinstein, despite being a local hero, is absolutely not a friend to the local tech industry. Time after time, she has sided with law enforcement and hollywood factions against a rational encryption policy, and appears to get most of her money from down south. I have no idea what Boxer is up to, but I take it that she's there because people believe her heart's in the right place.
The guy who ran against Feinstein, Tom Campbell, (he was too much of a maverick to have any chance of winning) was pretty sharp. Stanford prof and congressman from the SV area, and very up on the anti-trust issues with Microsoft and encryption policy. Too bad he's out of government now. There's various other industry functionaries that have been elected to certain posts, but I would say that's more "chamber of commerce" than "technically saavy".
--
The reason Microsoft can say that "open source != secure" is the EXACT SAME logic that you are using when you say "OpenBSD==secure" (how? why? doing what? who's the admin? etc?)
People want simple answers to a complex problem. You're playing the same advocacy game that Microsoft is.
--
The fact is that ANY business application of any software requires an expert (in-house or consultant) before people should roll it out.
You may think that and be correct. But the fact remains that many small NT shops limp along OK without expert assistatnce, and maybe a little tech support from the guy who sold them the comptuer. You can slap Great Plains on an NT box and have an accounting system.
These small businesses will never pay a Unix admin the money he's worth, nor would they pay a good NT guy the money either. And retention would be a problem in an environment where the guy is rehabing old 486s and refilling the copy machine toner. So, they do the rational thing and limp along with part time or slightly retarded computer help.
Open source software + contracted service would be a great solution for small businesses with no inhouse experience, and cheaper than paying MS licences. However the customer base is never going to pick up the phone and sign a service contract with RedHat or IBM (nor should they). The only real answer is that it's going to take a phalanx of Linux saavy people in the small system integrators that are out there.
And that I don't see. It's easier for the corner comptuer guy to build a computer and slap MS Small Business server or Exchange on it and send the money back to Microsoft. If he could find a skilled Linux admin to hire, the guy is going to figure that he's got better things to do than screwing in IDE drives in his spare time and go and make $90K doing Unix admin for a big corporation.
(As a sidenote, Novell essentially built their business on these Corner Computer Store integrators, but it took a massive channel push and lots of product education. So, it's possible, but it's a long haul for someone.)
--
Considering they amount of money spent on development of NT 3.1, 3.5, and 4.0, and the anemic rate of adoption of those products, Windows 2000 isn't looking so bad.
Nobody in the history of personal computers has successfully sold a "modern" desktop operating system to the public. Xenix didn't sell, OS/2 1.x didn't sell, OS/2 2.x didn't sell, and NT hasn't sold well either, and Linux on the desktop is as of yet a pipedream. Considering that sad history, if I was Fat Ballmer, I wouldn't be sweating the W2K sales too hard.
--
Word for the Mac kicked ass. Agreed that Word For Windows 1.x and 2.x weren't very good.
--
MS-DOS never ran on Z-80s. Those things probably ran CP/M. Maybe you're talking about MSX machines, but I doubt you know.
BTW, you obviously never dealt with the old IBM. Their "philosophic-nearly-religious" convictions were equal to what you see out of Microsoft today. Shit like OS/2, PS/2, and SAA were the height of their arrogance. Microsoft had the choice to remain as a subcontractor to that death star or move on, and as shitty as they are, the fact that they moved on was a great thing for personal computing in general.
--
Office was never playing catch up. Both Word and Excel were best of class programs 12 years ago.
In fact, as you mention, most of the new features have made them much worse products for the mid-tier to experienced user. This means that there's a greater opportunity NOW for a challenger (say StarOffice) to underprice and outperform the marketleader than at any time since Microsoft stole market from Lotus and the Perpetually Sucky WordPerfect.
--
Realistically, nobody but Microsoft and The Gartner Group expected mass conversions of corporate desktops to W2K within the first 2 years. And considering the history of NDS, minimal ActiveDirectory adoption isn't a big shock either.
So to say that "hasn't done very well" all depends on your expectations. From my POV, it's done extremely well in attracting power users and frustrated Win9x users. Obviously if you have a stable NT4 network (and most don't), there's little benefit, but if MS was honest they would have admitted that. (As they seem to be doing for WinXP vs. Win2K.)
--
1987. That was when IBM was trying force it's POS 286-specific OS (OS/2) and it's proprietary PC hardware (PS/2) down everyone's throats as an end-to-end solution to sell more mainframes.
Thank god Microsoft bucked them. As bad as things are, we're better off than if IBM would have won that one.
--
I think Linux has a real problem, and that is the elitist attitude found among developers. "Why don't they just read the man pages", etc. There is a real market for a distribution which can easily replace a newbie's MS Windows OS, however, it seems that no one wants to build it. Why? The geeks would rather keep it for themselves.
There's nothing wrong with an elitist attitude and wanting to "keep it for yourself".
The problem is the intellectual dishonesty involved in ripping on Microsoft (and Apple) because they make a consumer-level product while steadfastly refusing to adapt Unix to the consumer market.
It's a "Put up or Shut up" situation, and Unix community's reaction is to do neither, even though Unix OSes have the capability to be better than Windows, both technically and in terms of user interface.
--
WDM (Work on 98, ME, 2000, XP)
Do you know of any hardware that uses the same driver for both 9x and 2000? I think that WDM is still a pie-in-the-sky. There's a huge segment of non-name and low-end hardware that still doesn't run on 2000 and never will.
Windows 95 was specifically designed to be back-compatible with some very old drivers. XP doesn't have that luxury. Microsoft's solution is to market the product primarily for new OEM installs and not upgrades, but that doesn't solve the labelprinter issue.
--
* SQL Server (code from IBM and Oracle);
You have any evidence for this? I've heard a lot of accusations against Microsoft, and this one is a first....
BTW, both Caldera and Stac sued Microsoft and settled. Case closed, let it drop.
--
(1) SW vendors will just "require" their software be installed and run with administrator right.
Hell, even Microsoft was doing this until a couple years ago, so it's no wonder the OEMs are busy supporting only Win9x. It's amazing how poorly MS has historically supported their own OS (NT).
MS will cave into pressure from developpers wanting unrestricted access to resources and users wanting to run their old programs/games that MS will create "Windows Classic", fourth in the 95/98/ME line of single user OSes.
Wouldn't shock me one bit. Recall that both NT4 and Win2000 were supposed to be the "unified system", and then MS always squeezed out another version of DOSWin.
The other alternative is that OEMs give the finger to MS and continue to ship WinME. If a large variety of crap off-the-shelf software and crap hardware (label printers, business card scanners, that ilk) won't run under WinXP, it will drive support costs through the fucking roof.
Don't forget that Win95 shipped being compatible with 98% of Windows 3.1 programs and compatible with around 70-80% of DOS/Win3.1 drivers. That made the conversion relatively painless. Can't say that's going to happen with WinXP. (Although I really wish MS would go nazi and make it happen.)
--
There has actually been numerous lawsuits and fed investigations of the car stereo situation. So far, Detroit has gotten away with what they've been doing (for example, the oval-shaped stereo in Tauruses), but don't oversimplify the situation.
--
IBM shouldn't have had to do an 'internal audit' to find that out -- it's right in their financial reports. And, good thing too, because Lou Gerstner basically saved the company by focusing on services.
The thing about consulting is that, unlike software or hardware, it only scales linearly - to double your revenue, you generally need to double your consulting force. That means there's a practical limit to profits that can be made in a particular segement. So, IBM is raking in the dough doing Windows 2000 migraiton consutlting, but they need to expand their offerings with things like Linux and Java.
However, if you've dealt with IBM recently, you'll know that services ain't their main cup of tea yet. They primarily use it as part of a sales bundle to move software or hardware. (Such as, you get a cut on these boxes and licences, just give us the support contract.)
They are thinking that Linux is an especially good sales tool against Sun accounts (Linux/Netfinity or Linux/S390 Cluster on the low end, maybe move some high profit RS/6000s on the high end.) However, their profit margin is MUCH higher on the classic IBM stuff (RS/6000, AS/400), so to a great degree Linux is just a door opener for them.
--
Mac Word was a great program, and I guess I get along because it still hiding somewhere beneth 2GB of do-dads in modern versions of Word.
Forget about MacPaint -- the program I really miss from the old Mac was SuperPaint (a combination of MacPaint and MacDraw with a few other features like layers). It's virutally impossible to find a nice low-end usable progam like this anymore. Everythings big-n-bloated (CorelDraw, Photoshop), or retarded (MS graphics progs).
My comment about business trends goes to the fact that Microsoft is redefining their strategy. They're raising prices on the brink of an IT spending recession, and it's pretty certain that they think they know what they are doing, but it's not clear anyone else does. Picking winners in the next few years is going to be tough.
--
Apologies -- I was looking at the lack of activity on http://bugzilla.mozilla.org/show_bug.cgi?id=18722 . Look forward to trying Moz XSLT out.
--
I read somewhere quite a while back, that Windows (the 95 family) is a loss leader. It could be in Microsoft's best interests to let Linux win.
Microsoft makes $Billions/year on Windows, and most of that is probably from Windows 9x. There's no way it's a loss-leader. In fact, it's one of the handful of products they make that's profitable.
(Which is not to say that there aren't markets where MS sells below cost. Apparently for the low-end home machines you see at electronics stores, Windows is sold to the OEMs at a very steep discount.)
--