Ah, I'm not talking about the help desk and borked machines, it's the management structure. The political fear is that a year later Mr Tweaky will have his own development department, or the legitimate issue that some half-baked app will get thrown in IT's lap with no budget to support it.
I will single out IT Depts -- On many occassions, I've seen IT actually fight to prevent users from using programming tools (even if it's just VB or Access). If the only programming tool on one's programmable computer is the Excel macro language, people have little other choice but to use it. Usually IT's reasoning is some "enterprise J2EE initiative" or "next years ERP implemenation" or some other phony politics.
(And even if a poor luser gets his hands on VB, good luck getting any sort of rights on a RDBMS system.)
I think this all stems from the nerd prejudice that only they know the secrets of the machine. In most organizations I've been in, there's usually some business analysts or finance people that are capable of putting together solutions to their problems.
Microsoft and DEC announced that they were working on a 64-bit version of NT4/Alpha. Unfortunately it was quashed when Compaq bought DEC.
My impression is that Windows/x64 was delayed due to driver and 32-bit compatibility issues and not because the core OS lacked 64-bit support. (There's still Linux distros that haven't got the 32-bit compatibility totally sorted.)
What do they have to gain (at this point) from actually producing a standards-compliant browser?
Improved Dev Tool support, plus developer eyeballs. You see the hated here from people who believe that IE is old/broken/buggy. In the CSS world it's the reincarnation of Netscape 4..
Plus, MS has spent the last few years hyping XML, so the lack of XHTML support stands out like a sore thumb.
OK, not "dead", but "dying". For all the pain with using webapps, people keep coming back and asking for more of them. I predict this.NET download thing will be about as popular as Java WebStart which does the same thing.
I think one of the big reasons MS put so much work into client-scripting technology is because you really can build an webapp that's 90% as good Win32. They have been around a while, but they are only just showing up in public (like Google Maps).
I gotta call bullshit here. ASP.NET is probably their most popular developer technology and they make an enormous amount of money off it. Client/Server is DEAD even in Microsoft-land.
Web-based applications like maps.google.com were possible in Internet Explorer 4.0 back in 1998. Until very recently, nobody was near them in DOM/Javascript functionality, and they still have an enormous amount of proprietary clientside stuff that FireFox hasn't matched.
The thing MS has finally seemed to realize is that XHTML and CSS actually helps them -- it would make their developer tools work better which in turn makes them more money. Realize that IE is just a tool for them, not an end in itself.
So, stupid was selling operating systems to stupid. What's your point?
"Stupid" just wanted the most featureful applicaiton programs possible. It is not at all stupid to trade abstract features like multitasking for full-featured programs like Lotus 1-2-3 (which would have been impossible to write on the systems that you suggest.)
Just four years later, Amiga came out with AmigaOS
Four years is an eternity in hardware development. It would have been impossible for IBM to ship an affordable 68000-based machine in 1981. And it was only so cheap because the case assembly was junk (by IBM PC standards).
At that time, people were using 4.1BSD and Smalltalk (including GUIs and IDEs).
Well, obviously nobody was running BSD and Smalltalk on $2000 desktop PCs. (What did a PDP-11 cost, $15,000?) However, here's a couple points:
1) One thing that people forget is the "hacker" mentality in early microcomputing. People wanted to run things right on the metal, with as little overhead as possible. Developers rewrote chunks of DOS and BIOS for tiny speed increases.
That is, even if they had BSD, they likely wouldn't want it. Even years later there was an enormous amount of resistance to OS/2 and so on.
2) The OS that Microsoft *really* wanted to push was not MS-DOS, but MS Xenix which was a cut-down version of UNIX.
Unfortunately for everyone, MS's master IBM never was interested, so MS ended up dumping off Xenix. IBM instead chose to develop OS/2, which could be considered an inferior alternative to Unix, and very likely was doing everything possible to prevent the PC from "growing up" and competing with midrange systems.
Well the interesting thing about ZIP is how quickly it took over from other standards. (It became the defacto PC compression standard in less than a year.) This was mainly due to Katz's friendship with influencial BBS sysops, who drummed up the conflict between PK and SEA, I believe.
Keep in mind that includes Long Island and half of NJ and Connecticut, not all of which is very densely populated. (Although perhaps Canada is not so agressive at including exurban areas in their statistics.)
> Mac OS X still runs almost all programs written for System 7
And by "almost all", you mean about 50%. Seriously -- most "System 7" programs weren't even 32-bit clean much less compatible with MacOS 7.5, 8, or 9. (I've got bunch of old 68K programs to prove it.)
Compatibility is actually much better with first-gen 128K Mac programs because they tended not to use undocumented APIs. By the time System 7 came around, developers were using every hack they could find.
The good thing about OS X is that it's 99% compatible with MacOS 9, not that it fixed Apple's only so-so record with backcompatibility.
I think RHEL isn't so much competing with other Linux distro's as with Windows.
Not at all. RedHat is very happy staying in the "Enterprise Unix" niche -- J2EE, financial applications, Unix, Oracle. They're stealing business like mad from UNIX/RISC companies and barely acknowledge Microsoft. Who needs Main Street when you have Wall Street?
RedHat has done almost nothing to compete in the "LAN" or Windows server market -- file & print, directory services, groupware, RAD apps -- they've simply got no answer to this stuff. (SuSE/Novell at least is building a product lineup.)
If you implement an Exchange client only to the documented MAPI spec and documented set of object properties
In order to do this, you need Outlook licenced and installed. You're basically writing an Outlook frontend, not a stand-alone client. These guys aren't really interested in MAPI -- what they want is the native Exchange RPC wire protocol.
The big problem, last I checked, was that that there was no complete open source DCE-RPC implementation (even though it's also a UNIX standard as well as WinNT), and until they have that, it will be very difficult to reverse engineer Exchange RPC.
Ximain's IMAP/HTTP-based connector is probably the "best" approach because it uses documented interfaces which MS put into Exchange exactly for this type of thing.
By late 1995, IBM was already in the process of backing away from OS/2, and Microsoft knew it. At that point all they had to do is sit back and wait and the OS/2 problem would take care of itself. By 1996 it was official.
If anything, OS/2 proves that massive consumer ad campaigns don't necessarily do a very good job of selling software. (Anyone else remember the IBM OS/2 Fiesta Bowl?)
It gets worse than that. In IE you can do window.elementname.value and it usually works properly. (I was agast at this when I found a guy doing it - his argument was that the InterDev typeahead prompted him to do it.)
In MS's defense, a lot of this stuff came about before there was any effort at standardization.
> I can't wait to see XAML(Microsoft's version of XUL).
Yeah, that's kinda my point about the GUI Designer. XAML will come with one, and MS will have more XAML developers on Day 2 than XUL has gained over 3-4 years.
I didn't mean to imply that Mozilla doesn't have 'great stuff' as well, just that Microsoft wouldn't screw their main proponents (developers, developers...) by dropping IE.
BTW, is there a good GUI Designer for XUL yet? XUL always seemed like it could do well in the VB/RAD space if it ever got good tool support, documentation, etc.
> MS would lose very little if everybody switched to Firefox
True, but they would also have nothing to gain. IE could be used as the delivery mechanism for all sorts of new technology (like "Avalon" rendering, for example). So, it benefits them to keep it around, especially when they are trying to migrate their base to something new.
IE is also full of extended technology and features that almost nobody uses except for the hardcore base of Micrsoft developers (VB/Internal types mainly). Stuff like Vector Markup Language, HTML animations, scriptlets, etc. Once given, this stuff can't be taken away.
(The 'smart navigation' seems to be standard hidden iframe/script stuff. It could be made to work on Firefox if MS wished it to be.)
> According to Apple, this eclipses shipments by all other UNIX/UNIX-like system vendors.
The problem with this argument is that only a small percentage of Mac systems are used in the "Unix Market" (where Sun and RedHat live). [For the sake of argument, define Unix Market as application servers, financial systems, high-end RDBMS, web hosting, engineering workstations, etc.]
The vast majority of Mac systems are still in Apple's traditional markets of creative and home desktops where the users run Mac programs and may not even be aware that there's Unix buried down there. Only lately has their been a slight uptick of Apple Unix users (in academia, biotech, clusters)
So you can play your meanigless numbers game, but for the most part Apple is not in serious compeition for Sun and RedHat. When their salesmen show up, they are bidding against each other (and Novell, IBM, etc) and almost never Apple.
The stupid part of IBM (the mini/mainframe side) is still trying to charge $200k for an AS/400 --- sorry, "iServer" --- that is comparable to a $5k HP Linux box.
Traditionally that "stupid part" is where IBM made all of it's money. (Although the Stupids are the customers, not IBM.)
My guess is that a huge chunk of IBM's Services revenue actually comes from servicing their Mainframe/Mini platforms, moreso than from $5K Linux boxes. So, its sort of an accounting trick -- take traditional proprietary platform revenue and make it sound like new services revenue to wall street. (DEC used to do the same thing.)
So, you can't really kill the golden gooses -- get rid of the AS/400, and the Services side is going to plummet (See DEC under Compaq management.)
Slashdot Mirror
There were LAN Manager MCSEs. I used to work with one, his serial # was in the 1000 range.
This was in the mid-90s, when MCSEs were few-and-far-between and generally earned more than equivilant UNIX Admins.
Ah, I'm not talking about the help desk and borked machines, it's the management structure. The political fear is that a year later Mr Tweaky will have his own development department, or the legitimate issue that some half-baked app will get thrown in IT's lap with no budget to support it.
> Not to single out IT departments
I will single out IT Depts -- On many occassions, I've seen IT actually fight to prevent users from using programming tools (even if it's just VB or Access). If the only programming tool on one's programmable computer is the Excel macro language, people have little other choice but to use it. Usually IT's reasoning is some "enterprise J2EE initiative" or "next years ERP implemenation" or some other phony politics.
(And even if a poor luser gets his hands on VB, good luck getting any sort of rights on a RDBMS system.)
I think this all stems from the nerd prejudice that only they know the secrets of the machine. In most organizations I've been in, there's usually some business analysts or finance people that are capable of putting together solutions to their problems.
Microsoft and DEC announced that they were working on a 64-bit version of NT4/Alpha. Unfortunately it was quashed when Compaq bought DEC.
My impression is that Windows/x64 was delayed due to driver and 32-bit compatibility issues and not because the core OS lacked 64-bit support. (There's still Linux distros that haven't got the 32-bit compatibility totally sorted.)
What do they have to gain (at this point) from actually producing a standards-compliant browser?
..
Improved Dev Tool support, plus developer eyeballs. You see the hated here from people who believe that IE is old/broken/buggy. In the CSS world it's the reincarnation of Netscape 4
Plus, MS has spent the last few years hyping XML, so the lack of XHTML support stands out like a sore thumb.
OK, not "dead", but "dying". For all the pain with using webapps, people keep coming back and asking for more of them. I predict this .NET download thing will be about as popular as Java WebStart which does the same thing.
I think one of the big reasons MS put so much work into client-scripting technology is because you really can build an webapp that's 90% as good Win32. They have been around a while, but they are only just showing up in public (like Google Maps).
I gotta call bullshit here. ASP.NET is probably their most popular developer technology and they make an enormous amount of money off it. Client/Server is DEAD even in Microsoft-land.
Web-based applications like maps.google.com were possible in Internet Explorer 4.0 back in 1998. Until very recently, nobody was near them in DOM/Javascript functionality, and they still have an enormous amount of proprietary clientside stuff that FireFox hasn't matched.
The thing MS has finally seemed to realize is that XHTML and CSS actually helps them -- it would make their developer tools work better which in turn makes them more money. Realize that IE is just a tool for them, not an end in itself.
So, stupid was selling operating systems to stupid. What's your point?
"Stupid" just wanted the most featureful applicaiton programs possible. It is not at all stupid to trade abstract features like multitasking for full-featured programs like Lotus 1-2-3 (which would have been impossible to write on the systems that you suggest.)
Just four years later, Amiga came out with AmigaOS
Four years is an eternity in hardware development. It would have been impossible for IBM to ship an affordable 68000-based machine in 1981. And it was only so cheap because the case assembly was junk (by IBM PC standards).
At that time, people were using 4.1BSD and Smalltalk (including GUIs and IDEs).
Well, obviously nobody was running BSD and Smalltalk on $2000 desktop PCs. (What did a PDP-11 cost, $15,000?) However, here's a couple points:
1) One thing that people forget is the "hacker" mentality in early microcomputing. People wanted to run things right on the metal, with as little overhead as possible. Developers rewrote chunks of DOS and BIOS for tiny speed increases.
That is, even if they had BSD, they likely wouldn't want it. Even years later there was an enormous amount of resistance to OS/2 and so on.
2) The OS that Microsoft *really* wanted to push was not MS-DOS, but MS Xenix which was a cut-down version of UNIX.
Unfortunately for everyone, MS's master IBM never was interested, so MS ended up dumping off Xenix. IBM instead chose to develop OS/2, which could be considered an inferior alternative to Unix, and very likely was doing everything possible to prevent the PC from "growing up" and competing with midrange systems.
Yes, but NYCers are basically subsidizing everyone out in the sticks. Plus you'll pay $50/month.
Well the interesting thing about ZIP is how quickly it took over from other standards. (It became the defacto PC compression standard in less than a year.) This was mainly due to Katz's friendship with influencial BBS sysops, who drummed up the conflict between PK and SEA, I believe.
NYC Metro has 4,092 ppl/sqmi.
Keep in mind that includes Long Island and half of NJ and Connecticut, not all of which is very densely populated. (Although perhaps Canada is not so agressive at including exurban areas in their statistics.)
> 2800 miles from Nova Scotia to Vancouver
Bogus calculation because most of that 2800x100 miles has nothing in it.
The population densities of Canadian cities are generally MUCH higher than american cities.
For example, Metro Toronto has 6,857 people per square mile. Metro Chicago only has 3,641 ppl/sqmi
http://www.demographia.com
> Mac OS X still runs almost all programs written for System 7
And by "almost all", you mean about 50%. Seriously -- most "System 7" programs weren't even 32-bit clean much less compatible with MacOS 7.5, 8, or 9. (I've got bunch of old 68K programs to prove it.)
Compatibility is actually much better with first-gen 128K Mac programs because they tended not to use undocumented APIs. By the time System 7 came around, developers were using every hack they could find.
The good thing about OS X is that it's 99% compatible with MacOS 9, not that it fixed Apple's only so-so record with backcompatibility.
I hate to say it, but the Winner is probably Lotus Notes R5, who allowed had Web Browsing (IE control) in "early 1999".
Download the .NET SDK from msdn.microsoft.com and use the Documentation viewer to surf the web with tabs.
These tabs are actually better in some respect than Firefox's -- you can drag'n'drop them around and they maintain Z-order correctly.
I think RHEL isn't so much competing with other Linux distro's as with Windows.
Not at all. RedHat is very happy staying in the "Enterprise Unix" niche -- J2EE, financial applications, Unix, Oracle. They're stealing business like mad from UNIX/RISC companies and barely acknowledge Microsoft. Who needs Main Street when you have Wall Street?
RedHat has done almost nothing to compete in the "LAN" or Windows server market -- file & print, directory services, groupware, RAD apps -- they've simply got no answer to this stuff. (SuSE/Novell at least is building a product lineup.)
If you implement an Exchange client only to the documented MAPI spec and documented set of object properties
In order to do this, you need Outlook licenced and installed. You're basically writing an Outlook frontend, not a stand-alone client. These guys aren't really interested in MAPI -- what they want is the native Exchange RPC wire protocol.
The big problem, last I checked, was that that there was no complete open source DCE-RPC implementation (even though it's also a UNIX standard as well as WinNT), and until they have that, it will be very difficult to reverse engineer Exchange RPC.
Ximain's IMAP/HTTP-based connector is probably the "best" approach because it uses documented interfaces which MS put into Exchange exactly for this type of thing.
> Microsoft never formally refuted the ad
By late 1995, IBM was already in the process of backing away from OS/2, and Microsoft knew it. At that point all they had to do is sit back and wait and the OS/2 problem would take care of itself. By 1996 it was official.
If anything, OS/2 proves that massive consumer ad campaigns don't necessarily do a very good job of selling software. (Anyone else remember the IBM OS/2 Fiesta Bowl?)
It gets worse than that. In IE you can do window.elementname.value and it usually works properly. (I was agast at this when I found a guy doing it - his argument was that the InterDev typeahead prompted him to do it.)
In MS's defense, a lot of this stuff came about before there was any effort at standardization.
> I can't wait to see XAML(Microsoft's version of XUL).
Yeah, that's kinda my point about the GUI Designer. XAML will come with one, and MS will have more XAML developers on Day 2 than XUL has gained over 3-4 years.
I didn't mean to imply that Mozilla doesn't have 'great stuff' as well, just that Microsoft wouldn't screw their main proponents (developers, developers...) by dropping IE.
BTW, is there a good GUI Designer for XUL yet? XUL always seemed like it could do well in the VB/RAD space if it ever got good tool support, documentation, etc.
> MS would lose very little if everybody switched to Firefox
True, but they would also have nothing to gain. IE could be used as the delivery mechanism for all sorts of new technology (like "Avalon" rendering, for example). So, it benefits them to keep it around, especially when they are trying to migrate their base to something new.
IE is also full of extended technology and features that almost nobody uses except for the hardcore base of Micrsoft developers (VB/Internal types mainly). Stuff like Vector Markup Language, HTML animations, scriptlets, etc. Once given, this stuff can't be taken away.
(The 'smart navigation' seems to be standard hidden iframe/script stuff. It could be made to work on Firefox if MS wished it to be.)
> According to Apple, this eclipses shipments by all other UNIX/UNIX-like system vendors.
The problem with this argument is that only a small percentage of Mac systems are used in the "Unix Market" (where Sun and RedHat live). [For the sake of argument, define Unix Market as application servers, financial systems, high-end RDBMS, web hosting, engineering workstations, etc.]
The vast majority of Mac systems are still in Apple's traditional markets of creative and home desktops where the users run Mac programs and may not even be aware that there's Unix buried down there. Only lately has their been a slight uptick of Apple Unix users (in academia, biotech, clusters)
So you can play your meanigless numbers game, but for the most part Apple is not in serious compeition for Sun and RedHat. When their salesmen show up, they are bidding against each other (and Novell, IBM, etc) and almost never Apple.
The stupid part of IBM (the mini/mainframe side) is still trying to charge $200k for an AS/400 --- sorry, "iServer" --- that is comparable to a $5k HP Linux box.
Traditionally that "stupid part" is where IBM made all of it's money. (Although the Stupids are the customers, not IBM.)
My guess is that a huge chunk of IBM's Services revenue actually comes from servicing their Mainframe/Mini platforms, moreso than from $5K Linux boxes. So, its sort of an accounting trick -- take traditional proprietary platform revenue and make it sound like new services revenue to wall street. (DEC used to do the same thing.)
So, you can't really kill the golden gooses -- get rid of the AS/400, and the Services side is going to plummet (See DEC under Compaq management.)