I'll second that. I've had a FreeNAS box up as a media storage server and backup for over a year and it's been great. 400MHz Pentium machine, RAID 1 on 2 400MB IDE drives, plus a 3rd 400MB IDE drive that is backup for my main workstation (which itself has RAID 1, so I'm not as worried about RAID on that backup).
My wife's Macbook gets backed up automatically every night using Superduper onto the NAS's CIFS share, and as does my Mac desktop. My web server and code repository (Linux) are backed up via streaming tar over ssh to the the NAS. Using syslog from the NAS share to the Linux box to monitor the health automatically, so I don't have to go through the web interface, just get an email if anything goes wrong (never has).
Only cost was the drives (about $400 if I recall correctly), and about a day to install and configure everything and set up the backups on all my machines. Getting the SSH right was actually the toughest, because it was hard to figure out where to store the authorized_keys files on the NAS where it would survive a reboot (sorry, don't recall how I finally solved that).
What he didn't mention is that this would run at night too, as opposed to a solar panel that only works during the day.
Sure, but this will only run when there is wind. In any case you are going to need some kind of energy storage, whether batteries for small scale use, or pumped water for larger scale.
With a small scale system like this, you could also combine it with solar panels and a battery and get luggable power generation that would work in most places.
I'm thinking about building one myself to get some numbers and see how well it scales
Cool! I'm sure a lot of people would love to see a project page for a DIY wind generator of this sort!
I find it interesting that you are a mathematician but have ignored the ongoing efforts to unify General Relativity and Quantum Mechanics. M theory seems to have a lot of promise toward resolving that conflict. It's a particularly interesting area of mathematics and physics.
Also, all of human language revolves around slapping names on things, so it's hardly reasonable to fault science for doing that as well. Science generally goes one step further, and provides pretty concise definitions of things (often in mathematical terms), asks testable questions, and often provides verifiable answers to those questions. Just because it hasn't yet provided verifiable answers to some questions doesn't mean it can't or won't.
I do agree, however, that some questions are intrinsically untestable. But even in those cases science and mathematics can help us gather evidence, even if it is non-conclusive.
Maybe. An elegant design reflects a deep understanding of everything it touches. Intensive study is necessary, but it goes beyond that. You have to know it so well that you instinctively feel what works and what doesn't. You can't grok something that way without caring a great deal about it. And while one person usually has a guiding vision, it takes the intense focus of lots of people to get the best possible outcome.
That's when the magic happens. The design starts to seem purely asthetic, because the functional design seamlessly helps you do what you wanted, without calling attention to itself. It's only if you stop and think about the amount of complexity that's hidden (beneath the apparent simplicity) that you really start to appreciate how elegant that design is.
So, like a religion? Well, perhaps like the good bits.
"There is no average common machine. Example: The mac mini is slightly underspec for a developer ( mainly: harddisk sucks, only 2 GB memory max ) and the design is completely irrelevant: we have all plenty of lost space under the desk. My company buys beige ibm/dell boxes with the same spec as the mini and roughly the same price, but the fact that the dell/ibm come with standard disk in a standard ugly box is seen as a benefit, unlike in my livingroom."
This is a great point that explains a lot about why Apple has not sold more machines. I'm a long time Mac user and developer and have often found it a bit frustrating to have to choose between inexpensive machines with little expandability or go into very high priced machines. And then got burned when I bought the original PCI-X based dual G5, and the expandability I expected to have vanished as the rest of the industry settled on PCI-Express.
All that said, Apple's strong point is in the laptop market, where expandibility isn't such an issue. As long as people can add RAM and maybe a new hard drive they seem to be happy. Apple is a very strong competitor in the laptop market, and that's the market that's seeing most of the growth.
These seems to me to be the perfect way for Open Source to make rapid progress and gain further acceptance. By targetting key industries that are only served by expensive software packages that are poorly supported or require expensive support contracts, Open Source can provide a obvious and undeniable cost and quality improvement over closed source software. This is doubly so for industries where the needs are well understood. In addition to library management software, I would suggest that class scheduling and enrollment/registration software might be another area. Universities and schools pay millions for this software, and it's usually pretty primative stuff. Inventory management and cash register software might be another area.
While it's true that commercial software developers would be hesitant to only control one half of the system (the server, while the client is open and modifiable), there are some factors that could make this useful for a commercial software developer regardless.
MMOGs are selling a service, not software. The majority of MMOG income is from subscription fees, and the primary reason people pay for them is because they enjoy the content and they want to be able to play with their friends. Some MMOs forgo retail box sales altogether and just have a free download and sell subscriptions (or digital objects).
There's not much reason to worry about people replacing the server to play for free, either. Developing and operating MMO servers and creating good content is expensive. Even with an Open Source client it would be tough for someone to just create a new server that was compatible, reliable under load and had better content while significantly undercutting prices. This means that alternate servers would remain unable to attract the majority of the community.
Furthermore, if you create your own custom art, sounds and animations, those key elements of the client would not necessarily be available to competitor. They are data and so could remain proprietary without violating the GPL, so users would need a license to use them. Even client side scripts might be able to remain closed source, but that seems like more of a gray area.
MMO development costs continue to climb, and the availablility of an Open Source MMO engine could be a very attractive alternative to developing your own or licensing a commercial one.
I think you misunderstand what the word "literally" means. I'd love to see you draw a screen full of anything faster by hand than a video card can render that frame to the screen. Won't happen.
More multi-core functionality in Xcode and related tools would be pretty cool.
What else would you like it to do? It already distributes compiles very effectively across all the processors & cores you have. OS X itself does a fine job distributing load where possible. Of course, if people write single threaded CPU intensive apps, there's not a lot the compiler or the OS can do about that.
Are you looking for some way to have the compiler extract parallelism from the code implicitly without the developer having to write multithreaded code? That would be nice I'll admit.
Math is too hard! Until the mathematicians make it more usable, it will never gain acceptance in the Real World(TM).
While undeniably funny, this is decidedly not insightful. First, math is hard, and thus the widespread use of calculators and computers to make it easier.
Here's my summary of TFA: "Somebody needs to do everything for me, including all of my thinking."
This couldn't be further from what TFA actually says. A better summary would be "don't make me have to spend time fiddling with the cruft, make it easy to use so I can get on with thinking about the important stuff."
A good user interface will get out of the way and let you get your job done, and perhaps more important it will have a shallow, near linear learning curve. A bad user interface imposes itself between you and the problem space, and is characterized by a steep learning curve. The command line is a great example of this. It requires a large amount of knowlege to do rather simple things compared to a GUI. Command lines have their place -- for those who have the knowlege required and need to do complicated things, they are quite useful.
The beauty of rational and objective thought is, I DON'T HAVE TO CARE WHO VINT CERF WORKS FOR! Vint Cerf has laid out his proposals and assertions, as has Google and the monopolistic telecomm companies. As a rational person, I can decide the veracity of their statements based on the other information at my disposal. I can never know when or if a liar is lying, so the questioning of motives is moot.
It's an interesting arguement, but one that doesn't hold much water in the real world. Neither you nor I have the time to research the veracity of every statement we read or hear. What's more, the information we would need to do so is rarely, if ever, completely available to us.
Understanding a person's trustworthiness helps guide us as to whether it's even worth considering their claims. Sure, it may be irrelevant who Vint Cerf works for, or it may not. It really depends on Vint Cerf himself. I don't know Vint Cerf, so I can't particularly judge how honest he is. But I can notice who signs his paychecks and presume he generally favors ideas that align with his financial interests.
Do you, as a selp proclaimed rationalist, investigate every faith healing you see on TV? Chances are you've probably formed a general opinion of faith healers and so you tend to distrust their actions based on prior assumptions. Presumably there was some rational thought and consideration that when into forming those assumptions in the first place, and that's fine.
As for this case in particular, I actually trust him a bit more knowing he's working for Google.
All that said, demoting the grandparent in promotion of rational thought is not rational.
I wouldn't call the DMCA and the (un)Patriot(ic) act "submarine legislation" - they had quite vocal critics that had damn good arguments, but the people in power were not listening to the critics.
I certainly agree about the Patriot Act, I wouldn't call it "submarine legislation". But I'd bet you that most people in the US have never heard of the DMCA. Unless, of course, they read slashdot. Compare this to say "activist judges" (or the Patriot Act), which pretty much everyone in the US has heard of, and I think it's fair to say that as DMCA is at least at periscope depth.
I hate Microsoft as much as the next slashdotter, but being an IT manager, I wouldn't want the headache of using one OS.
You are a rare and valuable IT manager. Most IT managers seem to argue the opposite -- they want a monoculture so they only have to learn and manage one thing.
I'd rather work with an IT manager like you any day.
There are an awful lot of people out there that would like to run OS X but are held back by their existing investment in Windows apps, or the (often mistaken) belief that they'll have trouble finding categories of applications on the Mac platform.
And those people are largely being served by Boot Camp -- sure not having to reboot to run would be a big improvement, but that may make it too easy, leading developers to believe that there is no reason to port.
This means Mac market share jumps significantly. And those people will prefer to run OS X native apps to Windows apps, given the choice. And for most categories of software that that most users want to run, there *is* a choice. Those vendors that choose to only write windows apps wont sell to these people.
The jump in market share would be good for Apple, no question. Whether it would ultimately be good for OS X is another question. There is no real choice in most cases for things like Microsoft Office -- businesses require it, especially Outlook because it can talk to Exchange server. It would be a rare IT department that would allow someone to use alternate apps like Open Office or Star Office, even assuming they had fully native OS X ports.
Also, some vendors might realise that in this scenario, for many categories of software, they can split an application into an engine running on the Windows API, and the GUI running on the OS X API, the two on the same machine via some method of IPC - perhaps simple TCP/IP. Which means a single code base for the engine, and porting only the GUI.
Having built applications both ways, I can tell you it's a lot more work to make the GUI run in a separate process space. You'd be far better off to use something like wxWidgets or SDL to isolate yourself from the details of the OS and write a single body of platform neutral code above it.
Despite all this, I can see a way that having this could actually work well. Imagine that OS X became really really good at running Windows apps, but left them with the WinXP look and feel (much like running an OS 9 app in OS X today) -- people would want ports because it would make their app look and feel better, but they could still use their legacy windows apps if they needed to.
The other part of this would be to provide the Mac OS X Cocoa APIs running on Windows with the full Windows look and feel. Then developers could write to the Mac APIs and get Windows versions for free. Take it a step further and do the same for Linux and you'd have a real incentive for developers to develop on and for the Mac first.
So what happens if my show ends, and I'm just flipping channels trying to find something interesting. Does it lock me in to the first commercial I come to?
I'm going to patent a device that will detect if you changed to the channel in the middle of the commercial and let you flip through it. If some else beats me to the patent, then this should serve as prior art!
I feel that dead open source or dead closed source ends up being the same issue.
You may feel that way, but there are some substantial differences.
Open Source can't really "die" as long as someone has a copy of the source. It can only be "mostly dead". Someone who has the source can always modify it to work elsewhere. At very least the source can be used to understand clearly what the file format is so they can retrieve the data.
A closed source application can die because it the company decides to stop producing it or goes out of business. IANAL but I'm pretty sure that even in the later case you can't legally use that software -- it doesn't revert to the public domain just because the company is gone.
So consider a dead closed source app with an open file format. If all you need is the data and there's another application that supports the format, then no problem. Unless the apps didn't quite read the spec the same way, or don't run on the same platform, or the other apps are way more expensive.
I don't have anything against open file formats, but it's no substitute for open source. It is better nothing though.
If a report came and featured a council comprosed of the equilivant anti-OSS people (ie headed by a microsoft spokesperson) people here would be screaming bloody murder.
Why hasn't the US already switched away from oil? Because it's cheap compared to competitive technologies. Even adding in the war subsidy (a hundred or two billion dollars a year), I think you'd only add a dollar or so to the price of gas
This comment is right on the mark: Oil is cheap compared to competitive technologies. You also correctly recognize the cost of the war as an external cost of oil -- arguably the US would have little interest in the Middle East if it weren't for oil, certainly not enough interest to go to war. That's what economists call an external cost. External costs cause the free market to operate inefficiently, because the costs aren't borne by the buyer and/or seller.
However, what you are missing is that there are a lot of other external costs of oil: damages and cleanup from oil spills and oil production is a smaller one, but global climate change is a huge one. It's hard to even begin estimating the costs of global climate change, but if current predictions are even close to correct the the costs will absolutely dwarf the costs of the war in Iraq.
Then India and China should have chosen to be the advanced countries rather than be the ones catching up.
Where exactly did China and India get the chance to choose to be poor rather than rich?
Imagine if someone did a study that showed that internet usage was linked to obesity.
This is an extremely flawed analogy, because the harm (obesity) falls on the actor (the internet user). Whereas with global climate change, the harm falls on everyone. A better example would be if pouring chemicals into the groundwater were linked to increased cancer rates in anyone who drank water. Strangely enough, the people affected in those cases tend to get upset and bring lawsuits that are extremely costly to the perpetrators. In some cases it's even outright illegal. That sounds fair to me.
The concern I have about the lite edition is it's going to increase the cost of Windows development and testing. Basically, this is just one more version that people will have test on to make sure everything behaves correctly: Win95, Win2k, WinXP, Vista--, and Vista. It was bad enough when we just had to test on and fix Win95 specific bugs. I also suspect it will force people to design for the lowest common denominator, unless of course the Vista-- version is as unpopular as Window ME.
And apparently working as a Techical Director at Electronic Arts is even better! Of course there are occasionally tight deadlines and the hours are a wee bit on the long side.
What Boot Camp does is remove the barrier to adoption. There are a number of Windows users who would like to switch, but need access to Software X or don't want to give up Game Y
I agree. But I also think there's an additional barrier. Business that would otherwise be willing to buy a Mac to allow an individual to work according to their preference often won't do so because they are worried that if the individual leaves, they will be stuck with a Mac that nobody wants to use. This way, the business can simply be reassured that they machine can be converted to run Windows. Sure there are many other significant barriers to business adoption, but that's one fewer.
The reason Apple doesn't want to sell OS X to PC users (aside from the obvious ties with their more lucrative hardware business) is that OS X simply wouldn't be as stable or bulletproof in the PC world as it has been in the Mac world.
Nonsense. The reason Apple doesn't want to sell OS X to PC users is because they make much more money selling hardware than software. Apple is in business to make a profit (and fortunately they seem to believe that producing a great product is the right way to do that). They aren't going to intentionally do things that reduce their profits.
Don't forget that Apple has already been down the road of licensing the OS. It nearly killed them. People starting buying Power Computing machines because they ran Mac OS as well or better than Apple hardware, and significantly cut into their sales -- sure they were getting OS license fees, but at the same time it was causing Apple's market share to plumet. Even though between Apple and Power Computing the Mac OS market share was growing, the press saw Apple's market share going down and started sounding the death knells. This helped convince developers and consumers that Apple was irrelevant and (combined with many other factors, including increasing quality of Windows) they were in real trouble. It finally took Microsoft making a deal with Apple to keep producing Office for the next 5 years to reduce the hemoraging enough for a turnaround (which started with killing the licensing and bring out the iMac).
There are some differences today, Apple is hip because of the iPod and OS X, but it would still be a really tough battle to get to the point where OS X licenses replaced the lost revenue from hardware sales. That not to say that this could never happen, but I would say Apple would have to have 10%+ market share and growing before it would be worth the risk.
Slashdot Mirror
I'll second that. I've had a FreeNAS box up as a media storage server and backup for over a year and it's been great. 400MHz Pentium machine, RAID 1 on 2 400MB IDE drives, plus a 3rd 400MB IDE drive that is backup for my main workstation (which itself has RAID 1, so I'm not as worried about RAID on that backup).
My wife's Macbook gets backed up automatically every night using Superduper onto the NAS's CIFS share, and as does my Mac desktop. My web server and code repository (Linux) are backed up via streaming tar over ssh to the the NAS. Using syslog from the NAS share to the Linux box to monitor the health automatically, so I don't have to go through the web interface, just get an email if anything goes wrong (never has).
Only cost was the drives (about $400 if I recall correctly), and about a day to install and configure everything and set up the backups on all my machines. Getting the SSH right was actually the toughest, because it was hard to figure out where to store the authorized_keys files on the NAS where it would survive a reboot (sorry, don't recall how I finally solved that).
What he didn't mention is that this would run at night too, as opposed to a solar panel that only works during the day.
Sure, but this will only run when there is wind. In any case you are going to need some kind of energy storage, whether batteries for small scale use, or pumped water for larger scale.
With a small scale system like this, you could also combine it with solar panels and a battery and get luggable power generation that would work in most places.
I'm thinking about building one myself to get some numbers and see how well it scales
Cool! I'm sure a lot of people would love to see a project page for a DIY wind generator of this sort!
I find it interesting that you are a mathematician but have ignored the ongoing efforts to unify General Relativity and Quantum Mechanics. M theory seems to have a lot of promise toward resolving that conflict. It's a particularly interesting area of mathematics and physics.
Also, all of human language revolves around slapping names on things, so it's hardly reasonable to fault science for doing that as well. Science generally goes one step further, and provides pretty concise definitions of things (often in mathematical terms), asks testable questions, and often provides verifiable answers to those questions. Just because it hasn't yet provided verifiable answers to some questions doesn't mean it can't or won't.
I do agree, however, that some questions are intrinsically untestable. But even in those cases science and mathematics can help us gather evidence, even if it is non-conclusive.
Sounds like religion
Maybe. An elegant design reflects a deep understanding of everything it touches. Intensive study is necessary, but it goes beyond that. You have to know it so well that you instinctively feel what works and what doesn't. You can't grok something that way without caring a great deal about it. And while one person usually has a guiding vision, it takes the intense focus of lots of people to get the best possible outcome.
That's when the magic happens. The design starts to seem purely asthetic, because the functional design seamlessly helps you do what you wanted, without calling attention to itself. It's only if you stop and think about the amount of complexity that's hidden (beneath the apparent simplicity) that you really start to appreciate how elegant that design is.
So, like a religion? Well, perhaps like the good bits.
"There is no average common machine. Example: The mac mini is slightly underspec for a developer ( mainly: harddisk sucks, only 2 GB memory max ) and the design is completely irrelevant: we have all plenty of lost space under the desk. My company buys beige ibm/dell boxes with the same spec as the mini and roughly the same price, but the fact that the dell/ibm come with standard disk in a standard ugly box is seen as a benefit, unlike in my livingroom."
This is a great point that explains a lot about why Apple has not sold more machines. I'm a long time Mac user and developer and have often found it a bit frustrating to have to choose between inexpensive machines with little expandability or go into very high priced machines. And then got burned when I bought the original PCI-X based dual G5, and the expandability I expected to have vanished as the rest of the industry settled on PCI-Express.
All that said, Apple's strong point is in the laptop market, where expandibility isn't such an issue. As long as people can add RAM and maybe a new hard drive they seem to be happy. Apple is a very strong competitor in the laptop market, and that's the market that's seeing most of the growth.
These seems to me to be the perfect way for Open Source to make rapid progress and gain further acceptance. By targetting key industries that are only served by expensive software packages that are poorly supported or require expensive support contracts, Open Source can provide a obvious and undeniable cost and quality improvement over closed source software. This is doubly so for industries where the needs are well understood. In addition to library management software, I would suggest that class scheduling and enrollment/registration software might be another area. Universities and schools pay millions for this software, and it's usually pretty primative stuff. Inventory management and cash register software might be another area.
While it's true that commercial software developers would be hesitant to only control one half of the system (the server, while the client is open and modifiable), there are some factors that could make this useful for a commercial software developer regardless.
MMOGs are selling a service, not software. The majority of MMOG income is from subscription fees, and the primary reason people pay for them is because they enjoy the content and they want to be able to play with their friends. Some MMOs forgo retail box sales altogether and just have a free download and sell subscriptions (or digital objects).
There's not much reason to worry about people replacing the server to play for free, either. Developing and operating MMO servers and creating good content is expensive. Even with an Open Source client it would be tough for someone to just create a new server that was compatible, reliable under load and had better content while significantly undercutting prices. This means that alternate servers would remain unable to attract the majority of the community.
Furthermore, if you create your own custom art, sounds and animations, those key elements of the client would not necessarily be available to competitor. They are data and so could remain proprietary without violating the GPL, so users would need a license to use them. Even client side scripts might be able to remain closed source, but that seems like more of a gray area.
MMO development costs continue to climb, and the availablility of an Open Source MMO engine could be a very attractive alternative to developing your own or licensing a commercial one.
I can literally draw the screen by hand faster
I think you misunderstand what the word "literally" means. I'd love to see you draw a screen full of anything faster by hand than a video card can render that frame to the screen. Won't happen.
More multi-core functionality in Xcode and related tools would be pretty cool.
What else would you like it to do? It already distributes compiles very effectively across all the processors & cores you have. OS X itself does a fine job distributing load where possible. Of course, if people write single threaded CPU intensive apps, there's not a lot the compiler or the OS can do about that.
Are you looking for some way to have the compiler extract parallelism from the code implicitly without the developer having to write multithreaded code? That would be nice I'll admit.
Well, according to TFA this latest vote was pretty much along party lines, Republicans voting against net neutrality and Democrats voting for it.
Math is too hard! Until the mathematicians make it more usable, it will never gain acceptance in the Real World(TM).
While undeniably funny, this is decidedly not insightful. First, math is hard, and thus the widespread use of calculators and computers to make it easier.
Here's my summary of TFA: "Somebody needs to do everything for me, including all of my thinking."
This couldn't be further from what TFA actually says. A better summary would be "don't make me have to spend time fiddling with the cruft, make it easy to use so I can get on with thinking about the important stuff."
A good user interface will get out of the way and let you get your job done, and perhaps more important it will have a shallow, near linear learning curve. A bad user interface imposes itself between you and the problem space, and is characterized by a steep learning curve. The command line is a great example of this. It requires a large amount of knowlege to do rather simple things compared to a GUI. Command lines have their place -- for those who have the knowlege required and need to do complicated things, they are quite useful.
The beauty of rational and objective thought is, I DON'T HAVE TO CARE WHO VINT CERF WORKS FOR!
Vint Cerf has laid out his proposals and assertions, as has Google and the monopolistic telecomm companies. As a rational person, I can decide the veracity of their statements based on the other information at my disposal. I can never know when or if a liar is lying, so the questioning of motives is moot.
It's an interesting arguement, but one that doesn't hold much water in the real world. Neither you nor I have the time to research the veracity of every statement we read or hear. What's more, the information we would need to do so is rarely, if ever, completely available to us.
Understanding a person's trustworthiness helps guide us as to whether it's even worth considering their claims. Sure, it may be irrelevant who Vint Cerf works for, or it may not. It really depends on Vint Cerf himself. I don't know Vint Cerf, so I can't particularly judge how honest he is. But I can notice who signs his paychecks and presume he generally favors ideas that align with his financial interests.
Do you, as a selp proclaimed rationalist, investigate every faith healing you see on TV? Chances are you've probably formed a general opinion of faith healers and so you tend to distrust their actions based on prior assumptions. Presumably there was some rational thought and consideration that when into forming those assumptions in the first place, and that's fine.
As for this case in particular, I actually trust him a bit more knowing he's working for Google.
All that said, demoting the grandparent in promotion of rational thought is not rational.
I wouldn't call the DMCA and the (un)Patriot(ic) act "submarine legislation" - they had quite vocal critics that had damn good arguments, but the people in power were not listening to the critics.
I certainly agree about the Patriot Act, I wouldn't call it "submarine legislation". But I'd bet you that most people in the US have never heard of the DMCA. Unless, of course, they read slashdot. Compare this to say "activist judges" (or the Patriot Act), which pretty much everyone in the US has heard of, and I think it's fair to say that as DMCA is at least at periscope depth.
It was funny till you explained the crap out of it and added 3 extra lines of sig
I hate Microsoft as much as the next slashdotter, but being an IT manager, I wouldn't want the headache of using one OS.
You are a rare and valuable IT manager. Most IT managers seem to argue the opposite -- they want a monoculture so they only have to learn and manage one thing.
I'd rather work with an IT manager like you any day.
There are an awful lot of people out there that would like to run OS X but are held back by their existing investment in Windows apps, or the (often mistaken) belief that they'll have trouble finding categories of applications on the Mac platform.
And those people are largely being served by Boot Camp -- sure not having to reboot to run would be a big improvement, but that may make it too easy, leading developers to believe that there is no reason to port.
This means Mac market share jumps significantly. And those people will prefer to run OS X native apps to Windows apps, given the choice. And for most categories of software that that most users want to run, there *is* a choice. Those vendors that choose to only write windows apps wont sell to these people.
The jump in market share would be good for Apple, no question. Whether it would ultimately be good for OS X is another question. There is no real choice in most cases for things like Microsoft Office -- businesses require it, especially Outlook because it can talk to Exchange server. It would be a rare IT department that would allow someone to use alternate apps like Open Office or Star Office, even assuming they had fully native OS X ports.
Also, some vendors might realise that in this scenario, for many categories of software, they can split an application into an engine running on the Windows API, and the GUI running on the OS X API, the two on the same machine via some method of IPC - perhaps simple TCP/IP. Which means a single code base for the engine, and porting only the GUI.
Having built applications both ways, I can tell you it's a lot more work to make the GUI run in a separate process space. You'd be far better off to use something like wxWidgets or SDL to isolate yourself from the details of the OS and write a single body of platform neutral code above it.
Despite all this, I can see a way that having this could actually work well. Imagine that OS X became really really good at running Windows apps, but left them with the WinXP look and feel (much like running an OS 9 app in OS X today) -- people would want ports because it would make their app look and feel better, but they could still use their legacy windows apps if they needed to.
The other part of this would be to provide the Mac OS X Cocoa APIs running on Windows with the full Windows look and feel. Then developers could write to the Mac APIs and get Windows versions for free. Take it a step further and do the same for Linux and you'd have a real incentive for developers to develop on and for the Mac first.
So what happens if my show ends, and I'm just flipping channels trying to find something interesting. Does it lock me in to the first commercial I come to?
I'm going to patent a device that will detect if you changed to the channel in the middle of the commercial and let you flip through it. If some else beats me to the patent, then this should serve as prior art!
I feel that dead open source or dead closed source ends up being the same issue.
You may feel that way, but there are some substantial differences.
Open Source can't really "die" as long as someone has a copy of the source. It can only be "mostly dead". Someone who has the source can always modify it to work elsewhere. At very least the source can be used to understand clearly what the file format is so they can retrieve the data.
A closed source application can die because it the company decides to stop producing it or goes out of business. IANAL but I'm pretty sure that even in the later case you can't legally use that software -- it doesn't revert to the public domain just because the company is gone.
So consider a dead closed source app with an open file format. If all you need is the data and there's another application that supports the format, then no problem. Unless the apps didn't quite read the spec the same way, or don't run on the same platform, or the other apps are way more expensive.
I don't have anything against open file formats, but it's no substitute for open source. It is better nothing though.
If a report came and featured a council comprosed of the equilivant anti-OSS people (ie headed by a microsoft spokesperson) people here would be screaming bloody murder.
Not if the report said the same thing.
Why hasn't the US already switched away from oil? Because it's cheap compared to competitive technologies. Even adding in the war subsidy (a hundred or two billion dollars a year), I think you'd only add a dollar or so to the price of gas
This comment is right on the mark: Oil is cheap compared to competitive technologies. You also correctly recognize the cost of the war as an external cost of oil -- arguably the US would have little interest in the Middle East if it weren't for oil, certainly not enough interest to go to war. That's what economists call an external cost. External costs cause the free market to operate inefficiently, because the costs aren't borne by the buyer and/or seller.
However, what you are missing is that there are a lot of other external costs of oil: damages and cleanup from oil spills and oil production is a smaller one, but global climate change is a huge one. It's hard to even begin estimating the costs of global climate change, but if current predictions are even close to correct the the costs will absolutely dwarf the costs of the war in Iraq.
Then India and China should have chosen to be the advanced countries rather than be the ones catching up.
Where exactly did China and India get the chance to choose to be poor rather than rich?
Imagine if someone did a study that showed that internet usage was linked to obesity.
This is an extremely flawed analogy, because the harm (obesity) falls on the actor (the internet user). Whereas with global climate change, the harm falls on everyone. A better example would be if pouring chemicals into the groundwater were linked to increased cancer rates in anyone who drank water. Strangely enough, the people affected in those cases tend to get upset and bring lawsuits that are extremely costly to the perpetrators. In some cases it's even outright illegal. That sounds fair to me.
The concern I have about the lite edition is it's going to increase the cost of Windows development and testing. Basically, this is just one more version that people will have test on to make sure everything behaves correctly: Win95, Win2k, WinXP, Vista--, and Vista. It was bad enough when we just had to test on and fix Win95 specific bugs. I also suspect it will force people to design for the lowest common denominator, unless of course the Vista-- version is as unpopular as Window ME.
Hey, didn't you see that CNN/Money picked software engineer as the best job in America.
And apparently working as a Techical Director at Electronic Arts is even better! Of course there are occasionally tight deadlines and the hours are a wee bit on the long side.
I was LMAO when I read that.
What Boot Camp does is remove the barrier to adoption. There are a number of Windows users who would like to switch, but need access to Software X or don't want to give up Game Y
I agree. But I also think there's an additional barrier. Business that would otherwise be willing to buy a Mac to allow an individual to work according to their preference often won't do so because they are worried that if the individual leaves, they will be stuck with a Mac that nobody wants to use. This way, the business can simply be reassured that they machine can be converted to run Windows. Sure there are many other significant barriers to business adoption, but that's one fewer.
The reason Apple doesn't want to sell OS X to PC users (aside from the obvious ties with their more lucrative hardware business) is that OS X simply wouldn't be as stable or bulletproof in the PC world as it has been in the Mac world.
Nonsense. The reason Apple doesn't want to sell OS X to PC users is because they make much more money selling hardware than software. Apple is in business to make a profit (and fortunately they seem to believe that producing a great product is the right way to do that). They aren't going to intentionally do things that reduce their profits.
Don't forget that Apple has already been down the road of licensing the OS. It nearly killed them. People starting buying Power Computing machines because they ran Mac OS as well or better than Apple hardware, and significantly cut into their sales -- sure they were getting OS license fees, but at the same time it was causing Apple's market share to plumet. Even though between Apple and Power Computing the Mac OS market share was growing, the press saw Apple's market share going down and started sounding the death knells. This helped convince developers and consumers that Apple was irrelevant and (combined with many other factors, including increasing quality of Windows) they were in real trouble. It finally took Microsoft making a deal with Apple to keep producing Office for the next 5 years to reduce the hemoraging enough for a turnaround (which started with killing the licensing and bring out the iMac).
There are some differences today, Apple is hip because of the iPod and OS X, but it would still be a really tough battle to get to the point where OS X licenses replaced the lost revenue from hardware sales. That not to say that this could never happen, but I would say Apple would have to have 10%+ market share and growing before it would be worth the risk.