Jobs sweeps his fingers in the air saying: "You don't need to boot OS X". Apple fans turn to each other saying: "We don't need to boot OS X". Jobs moving his hand again: "Go buy a new Mac" Apple fans: "Let's go buy a new Mac" Jobs sweeping his fingers a final time: "Move along to the Apple store" Apple fans: "Move along, move along"
"But let's be serious. Apple would sell DVDs of OSX at Best Buy in a heartbeat if they thought it could work."
I agree. Apple isn't going to do it because it won't work (i.e. it wouldn't make business sense).
"Companies don't have agendas, people do. I'm sure there are lots of people at Apple that would like to put a dent in the windows desktop monopoly."
A company's agenda (yes, they do have them) is not the aggregate opinion of all its workers, it reflects the views of its leaders. I'm sure Jobs would love to sell more computers by increasing OS X's market share, but the goal is to sell more computers not strike a blow against MS. If he can sell more Apple computers by allowing Windows to run on them, he'll gladly take that as well. The money is just as green.
"Remember, Apple had a huge home and educational market share at one point in time."
Yes, that would be the time before the PC and the Mac were on the market.
1) The reason Apple lost the OS war to Microsoft in the 1990s, despite having the superior OS, is because of the strategic mistake of tying their software to their hardware. Perhaps with another superior OS on their hands, they don't want to repeat the same mistake.
They lost that war in the 1980's, not the 1990's. You underestimate the significance of the 68000 vs. the 8088. The segmentation of the 8088 made it very hard to produce programs that were greater than 64K in size. On the 68000 you could address 1MB of memory effortlessly. The 68000 had a true 16 bit data bus, while the 8088 was only 16 bits on the inside.
If the original Mac OS had been written for the PC market, it would probably be 20% slower and would have taken 20-50% longer to develop.
This extra capability however, came with a price. The 68000 was more expensive than the 8088 and the Mac's 128KB of RAM was at least double what most people had in their PC. Keep in mind that memory was extremely expensive in those days.
So the Mac "lost the OS war" at least partially because it was necessarily more expensive (the lack of compatiblity with PC applications was a factor too).
Licensing the OS to third-parties (assuming there would be any takers) would not have significantly lowered the hardware cost of running the OS.
By the time the 1990's arrived, DOS and eventually Windows were so firmly entrenched that converting people from a PC to a Mac became much harder even though the cost issue was not as great.
In my opinion Jobs did the right thing for Apple when he returned and killed the Mac clone efforts.
Keep in mind that the 9x and ME version of Windows were from a codebase that orginated back in the 8088 days. The 8088 had no memory management so the OS, drivers, and all applications had to share the same address space. If the top free/OSS programmers tried to port Linux to the 8088, it would still crash if a flawed application or driver overwrote OS data structures.
We can fault MS for not producing a stable OS sooner, but on the other hand some people continue to use Windows 95 today despite the fact that more stable Windows versions have been available for a decade. Maybe some people just don't want to buy a new machine.
Perhaps one way to promote Linux is to create a version that supports the classic Windows 95 machine: i.e. 386/486/Pentium 1 8MB of RAM. I know earlier versions of Linux supported this configuration, but I'm talking about incorporating the improvements in installation ease that we've seen in recent years into a lightweight Linux. Or perhaps Linux developers aren't any more interested in the hardware slackers than MS is.
Apple isn't in the OS X business, they are in the computer hardware business. If somebody buys an Apple instead of a Dell so they can run an occasional Mac application, Boot Camp is a success.
Of course, many people want to see Windows market share decrease, but that's their agenda, not Apple's.
Traditionally standards were about codifying existing practice, not creating new technologies. It was typically the market leaders (often monopolies), who created the technology, got it working and then sought to create a standard around it.
Obviously standards created using this traditional approach are going to be adopted more readily than an approach that is designed to be an alternative to the existing leading technologies.
It's probably just a gimmick to boost Red Hat's stock price. The tweaking that Red Hat does to Linux is rather abstract for most investors to appreciate, but buying another company makes it sound like something new is going on.
I'd have to know the specifics to determine if your CEO friend really has a legitimate problem, doesn't know what he's doing, or is just part of the disinformation campaign.
One quick way to increase the pool of qualified applicants is to eliminate the top of the experience range you advertise. For example, instead of asking for sw engineers with 3-5 years experience just state that the job requires a minimum of 3 years experience. Probably 60-70% of sw engineers have more than 5 years experience so you'll see a big increase in applications.
Of course, if the real issue is not a lack of qualified applicants but a lack of cheap, qualified applicants then this approach won't help.
"I'm not after people who can learn "lots" of languages, I'm after people who understand the principles and practice of language structure and design, the challenges of multi-threaded code, distributed code and enterprise challenges. These are the bits that an IT degree helps with."
The reality of university courses is that there is a limited amount of time one can spend on an assignment. CS courses can teach you important principles, but the toy problems solved in classes bear little resemblence to real-world industry problems.
After graduation, it can take years to unlearn some of the absolute "truths" that apply only in the artifical academic environment where there are no issues of cashflow, customers or competitors.
So a corporation that depends on computer science graduates for its business wants to keep their labor costs down in the future by suckering young people into a career that will probably be over by the time they are 50.
The shortage of technical talent in the US has been proclaimed by industry continuously since the 1950's but it has never been true.
Given the absurd compensation given CEOs in the US, perhaps IBM should encourage more business school graduates to try to flood the market with cheap management labor.
"When I took over the IT department where I work, I removed all bloated and over priced software. I don't care why they had it... I can find much better resources within the open source community. This is why I fired all of the temps we had that coded all the crap for windows and Oracle. I hired Perl developers. They are much more productive in an OSS environment."
So everybody else in the company took a few months off while the Perl developers tried to recreate the lost functionality. Your IT tail is wagging the dog.
"So, you would say platform-specific development which is tied to one vendor's bottom line is good?"
Is this your defense of your "old-school" comment?
Anyway, the answer to your question is: it depends.
First of all, Windows isn't going to go away anytime soon. So you probably could write application for it today and be assured that the market will still be there in 10 years.
Second, the idea of platform-independence with a single code base is more an academic goal then a reality. You still have to test on every system you intend to market to and you have to be aware of version problems and scale issues (if Java were really platform-independent there would be a single version that covered everthing from handhelds to enterprise).
Third, sometimes it's better to let the future pay for itself rather than paying for it in the present. These days it's widely believed that you have to design everything so that all future changes are easy to make. The problem with that approach is that the future is seldom what you expect. In addition, many startups have failed because too much time was taken trying to cover the future when there wasn't enough money to pay for it. So if in the future you want to port your application to another platform, you can. You also have the opportunity to customize the application to the new platform so that it's better for your new customers.
Anyway, I'm not saying that aiming for platform-independence is a bad thing in general, but it's not appropriate or necessary in many cases.
"If you want to write an app for windows you pretty much have to use.NET these days."
Not at all. You can still use Visual Basic, C++ from many vendors etc. There's no unique functionality in.NET, it's just an higher-level way to write code.
"I'd think that any new software project would be silly to tie itself to one operating system at the programming language level (even if not in Java: Python, Ruby, etc.)"
Sure. Why go for just 90% of the software market when with additional effort and degraded performance you can approach 100% without recompiling. If the linux users won't buy it anyway, so what?
"Single-platform applications programming is pretty old-school, IMO"
Right. Whenever I come up with an idea I ask the question: Is this old-school? If the answer is yes, I forget all about it no matter how promising it might be.
"Operating systems like OS/2 were able to retain almost 100% compatibility with DOS and 16-bit Windows applications without making the kinds of architectural sacrifices that Microsoft has made over the past 10+ years"
The key word here is "almost". Also keep in mind that the number of users of OS/2 was very, very, very, small. So the degree of compatibility is still really unknown since the majority of DOS and 16-bit Windows applications have never been tested on OS/2.
The DOJ didn't "fold". They just decided not to bother using taxpayer's money to force MS to provide Sun's JVM in Windows. MS paid the protection money to Sun (again), so Sun stopped lobbying against them. MS paid protection money to AOL, same result. MS paid money to Real, same result.
You didn't really believe the orignal DOJ case was all about the consumer did you? The case came about through the efforts of competitors and once they received their money no further action was needed or taken. Business as usual in Washington.
Yes, I remember how you couldn't turn a page in a computer magazine without and ad from MS saying how much better Windows NT 4.0 was than Netware.
But seriously, marketing had little to do with it. Netware had a near monopoly at one time and the market was theirs to lose. Many businesses found that MS's solution was cheaper and easier than Netware. Many companies at that time needed only the ability to share files across a network and that's all they used Netware for anyway.
Your interpretation makes sense, that's probably what she meant to say.
On the other hand, I doubt that the founding fathers believed that the "unalienable Rights" they spoke of in the Declaration of Independence could be suspended by individuals and institutions in the private sector. They understood a potential threat from government but perhaps underestimated the private sector's greed and overestimated the judicial system's ability to resist it.
"The First Amendment applies only to the government (public); and not to (private) citizens or corporations of private citizens."
I'm confused. The above statement is the misunderstanding you are trying to correct. Right? You're not suggesting (I hope) that the First Ammendment doesn't apply to private citizens. Private citizens are the only ones we can be sure that the Bill of Rights applies to. The fact that courts have said that it applies to corporations as well proves that a judge doesn't have to be liberal to be activist.
Slashdot Mirror
Ouch. It was supposed to be funny, but one of the "can't take a joke" crowd got to it first.
Jobs sweeps his fingers in the air saying: "You don't need to boot OS X".
Apple fans turn to each other saying: "We don't need to boot OS X".
Jobs moving his hand again: "Go buy a new Mac"
Apple fans: "Let's go buy a new Mac"
Jobs sweeping his fingers a final time: "Move along to the Apple store"
Apple fans: "Move along, move along"
"But let's be serious. Apple would sell DVDs of OSX at Best Buy in a heartbeat if they thought it could work."
I agree. Apple isn't going to do it because it won't work (i.e. it wouldn't make business sense).
"Companies don't have agendas, people do. I'm sure there are lots of people at Apple that would like to put a dent in the windows desktop monopoly."
A company's agenda (yes, they do have them) is not the aggregate opinion of all its workers, it reflects the views of its leaders. I'm sure Jobs would love to sell more computers by increasing OS X's market share, but the goal is to sell more computers not strike a blow against MS. If he can sell more Apple computers by allowing Windows to run on them, he'll gladly take that as well. The money is just as green.
"Remember, Apple had a huge home and educational market share at one point in time."
Yes, that would be the time before the PC and the Mac were on the market.
1) The reason Apple lost the OS war to Microsoft in the 1990s, despite having the superior OS, is because of the strategic mistake of tying their software to their hardware. Perhaps with another superior OS on their hands, they don't want to repeat the same mistake.
They lost that war in the 1980's, not the 1990's. You underestimate the significance of the 68000 vs. the 8088. The segmentation of the 8088 made it very hard to produce programs that were greater than 64K in size. On the 68000 you could address 1MB of memory effortlessly. The 68000 had a true 16 bit data bus, while the 8088 was only 16 bits on the inside.
If the original Mac OS had been written for the PC market, it would probably be 20% slower and would have taken 20-50% longer to develop.
This extra capability however, came with a price. The 68000 was more expensive than the 8088 and the Mac's 128KB of RAM was at least double what most people had in their PC. Keep in mind that memory was extremely expensive in those days.
So the Mac "lost the OS war" at least partially because it was necessarily more expensive (the lack of compatiblity with PC applications was a factor too).
Licensing the OS to third-parties (assuming there would be any takers) would not have significantly lowered the hardware cost of running the OS.
By the time the 1990's arrived, DOS and eventually Windows were so firmly entrenched that converting people from a PC to a Mac became much harder even though the cost issue was not as great.
In my opinion Jobs did the right thing for Apple when he returned and killed the Mac clone efforts.
Keep in mind that the 9x and ME version of Windows were from a codebase that orginated back in the 8088 days. The 8088 had no memory management so the OS, drivers, and all applications had to share the same address space. If the top free/OSS programmers tried to port Linux to the 8088, it would still crash if a flawed application or driver overwrote OS data structures.
We can fault MS for not producing a stable OS sooner, but on the other hand some people continue to use Windows 95 today despite the fact that more stable Windows versions have been available for a decade. Maybe some people just don't want to buy a new machine.
Perhaps one way to promote Linux is to create a version that supports the classic Windows 95 machine: i.e. 386/486/Pentium 1 8MB of RAM. I know earlier versions of Linux supported this configuration, but I'm talking about incorporating the improvements in installation ease that we've seen in recent years into a lightweight Linux. Or perhaps Linux developers aren't any more interested in the hardware slackers than MS is.
I doubt that Linux fans would want to be put in the same unflattering category as Windows users - they want to be insulted separately.
Apple isn't in the OS X business, they are in the computer hardware business. If somebody buys an Apple instead of a Dell so they can run an occasional Mac application, Boot Camp is a success.
Of course, many people want to see Windows market share decrease, but that's their agenda, not Apple's.
Traditionally standards were about codifying existing practice, not creating new technologies. It was typically the market leaders (often monopolies), who created the technology, got it working and then sought to create a standard around it.
Obviously standards created using this traditional approach are going to be adopted more readily than an approach that is designed to be an alternative to the existing leading technologies.
It's probably just a gimmick to boost Red Hat's stock price. The tweaking that Red Hat does to Linux is rather abstract for most investors to appreciate, but buying another company makes it sound like something new is going on.
I'd have to know the specifics to determine if your CEO friend really has a legitimate problem, doesn't know what he's doing, or is just part of the disinformation campaign.
One quick way to increase the pool of qualified applicants is to eliminate the top of the experience range you advertise. For example, instead of asking for sw engineers with 3-5 years experience just state that the job requires a minimum of 3 years experience. Probably 60-70% of sw engineers have more than 5 years experience so you'll see a big increase in applications.
Of course, if the real issue is not a lack of qualified applicants but a lack of cheap, qualified applicants then this approach won't help.
It's nice that you got to vent your frustrations, but your post seems to have little to do with mine.
"I'm not after people who can learn "lots" of languages, I'm after people who understand the principles and practice of language structure and design, the challenges of multi-threaded code, distributed code and enterprise challenges. These are the bits that an IT degree helps with."
The reality of university courses is that there is a limited amount of time one can spend on an assignment. CS courses can teach you important principles, but the toy problems solved in classes bear little resemblence to real-world industry problems.
After graduation, it can take years to unlearn some of the absolute "truths" that apply only in the artifical academic environment where there are no issues of cashflow, customers or competitors.
So a corporation that depends on computer science graduates for its business wants to keep their labor costs down in the future by suckering young people into a career that will probably be over by the time they are 50.
The shortage of technical talent in the US has been proclaimed by industry continuously since the 1950's but it has never been true.
Given the absurd compensation given CEOs in the US, perhaps IBM should encourage more business school graduates to try to flood the market with cheap management labor.
"When I took over the IT department where I work, I removed all bloated and over priced software. I don't care why they had it... I can find much better resources within the open source community. This is why I fired all of the temps we had that coded all the crap for windows and Oracle. I hired Perl developers. They are much more productive in an OSS environment."
So everybody else in the company took a few months off while the Perl developers tried to recreate the lost functionality. Your IT tail is wagging the dog.
Wow, clothes don't make the programmer but how he dresses-up his code does? Coding conventions are the tux and tails of the software world.
If you want to be anal about non-functional factors, don't blame the man for doing the same thing.
"So, you would say platform-specific development which is tied to one vendor's bottom line is good?"
Is this your defense of your "old-school" comment?
Anyway, the answer to your question is: it depends.
First of all, Windows isn't going to go away anytime soon. So you probably could write application for it today and be assured that the market will still be there in 10 years.
Second, the idea of platform-independence with a single code base is more an academic goal then a reality. You still have to test on every system you intend to market to and you have to be aware of version problems and scale issues (if Java were really platform-independent there would be a single version that covered everthing from handhelds to enterprise).
Third, sometimes it's better to let the future pay for itself rather than paying for it in the present. These days it's widely believed that you have to design everything so that all future changes are easy to make. The problem with that approach is that the future is seldom what you expect. In addition, many startups have failed because too much time was taken trying to cover the future when there wasn't enough money to pay for it. So if in the future you want to port your application to another platform, you can. You also have the opportunity to customize the application to the new platform so that it's better for your new customers.
Anyway, I'm not saying that aiming for platform-independence is a bad thing in general, but it's not appropriate or necessary in many cases.
"Not so -- educated estimates in 1995/1996 had OS/2 sitting between 15% and 20% of the total desktop market"
I don't know what you mean by educated estimates, but I'd have to see some real evidence before I'd buy those numbers.
"If you want to write an app for windows you pretty much have to use .NET these days."
.NET, it's just an higher-level way to write code.
Not at all. You can still use Visual Basic, C++ from many vendors etc. There's no unique functionality in
"I'd think that any new software project would be silly to tie itself to one operating system at the programming language level (even if not in Java: Python, Ruby, etc.)"
Sure. Why go for just 90% of the software market when with additional effort and degraded performance you can approach 100% without recompiling. If the linux users won't buy it anyway, so what?
"Single-platform applications programming is pretty old-school, IMO"
Right. Whenever I come up with an idea I ask the question: Is this old-school? If the answer is yes, I forget all about it no matter how promising it might be.
"Operating systems like OS/2 were able to retain almost 100% compatibility with DOS and 16-bit Windows applications without making the kinds of architectural sacrifices that Microsoft has made over the past 10+ years"
The key word here is "almost". Also keep in mind that the number of users of OS/2 was very, very, very, small. So the degree of compatibility is still really unknown since the majority of DOS and 16-bit Windows applications have never been tested on OS/2.
The DOJ didn't "fold". They just decided not to bother using taxpayer's money to force MS to provide Sun's JVM in Windows. MS paid the protection money to Sun (again), so Sun stopped lobbying against them. MS paid protection money to AOL, same result. MS paid money to Real, same result.
You didn't really believe the orignal DOJ case was all about the consumer did you? The case came about through the efforts of competitors and once they received their money no further action was needed or taken. Business as usual in Washington.
"You are confused. The rights are defined properly. The private sector can't suspend your rights, but your rights don't trump their property rights"
Hey buddy, I'm just responding to what other posters are saying and not qualifying their arguments for them.
"for no other reason than marketing.."
Yes, I remember how you couldn't turn a page in a computer magazine without and ad from MS saying how much better Windows NT 4.0 was than Netware.
But seriously, marketing had little to do with it. Netware had a near monopoly at one time and the market was theirs to lose. Many businesses found that MS's solution was cheaper and easier than Netware. Many companies at that time needed only the ability to share files across a network and that's all they used Netware for anyway.
Your interpretation makes sense, that's probably what she meant to say.
On the other hand, I doubt that the founding fathers believed that the "unalienable Rights" they spoke of in the Declaration of Independence could be suspended by individuals and institutions in the private sector. They understood a potential threat from government but perhaps underestimated the private sector's greed and overestimated the judicial system's ability to resist it.
"The First Amendment applies only to the government (public); and not to (private) citizens or corporations of private citizens."
I'm confused. The above statement is the misunderstanding you are trying to correct. Right? You're not suggesting (I hope) that the First Ammendment doesn't apply to private citizens. Private citizens are the only ones we can be sure that the Bill of Rights applies to. The fact that courts have said that it applies to corporations as well proves that a judge doesn't have to be liberal to be activist.