Another benefit would be the ability to migrate the client to other platforms, and hedge against forced obsolescence. If you and your friends (or larger group) didn't want to upgrade your gear, you could always fork an earlier version of the engine and host your own game for the FPS challenged - unlike commercial games that evolve away and above marginal users ability to keep up.
As a pure money-making enterprise, PC game development shops focus their dollars on the platform most likely to provide a return on their investment - e.g. MS Windows - with a smaller, lagging amount reserved for Apple (if they even support Apple), and next to nothing for FOSS (Linux/BSD et al). I know game shops that ported early versions of their games 5+ years ago to Linux, but dropped the support due to a lack of return on the investment.
My thought is if FOSS makes it easy for developers to port their current Windows apps as a planned adjunct to their existing development, and also have key tools/engines that are open source, we can attract more of the mainstream game producers to make games for Linux, or at a minimum port their existing Windows games to Linux. By lowering the bar to entry into the FOSS market, they would be more likely to take advantage of it. Many of them probably won't due to percieved threats to their IP and competitive advantage. My hope would be the titles that I enjoy would take advantage of it to get my entertainment dollars; if they did I could remove the last vestage of MS Windows from my network, and gain the benefits of my high-end system for things other than gaming and listening to music.
Failing that or in addition to that - the FOSS world would at least have the tools to build our own stuff on par with mainstream games, without having to reinvent the wheel, and different groups could handle the upgrade path as they desired. So I see it as valuable from that perspective.
...of the hollywood scenes, folks are hacking remote systems; they've viewing code through a terminal emulator, not generally a fancy dev environment.
You know, you can run emacs remotely via X - provided it is loaded on the remote system - which will give you all the syntax highlighting goodness (I do this all the time). It also allows you to use your mouse on the remote instance (particularly useful when you forget what ctrl-x ctrl-b, ctrl-x o or ctrl-x ctrl-s does).
I think that describes it pretty well, and is how it should be explained to the politicians.
That would be the worse thing to tell a Congresstroll. They would be falling all over themselves to write legislation resurrecting the Clipper chip, giving Bill Gates bags full of money to DRM the hell out of it with 'trusted computing', and laws requiring only *approved* software and hardware on any network! All wireless routers, cell phones and laptops would be contraband (except for military/governmental use, of course).
I see this as a larger pattern. Big companies are losing mind-share by outsourcing system development. If taken to an extreme the company has no technical knowledge and does nothing in house, instead it is just a collection of project managers and lawyers - a paper tiger - at great risk of being crushed by savvy competitors.
it becomes so easy and ubiquitous that you would be worse off to do it the old way (an example would be webmail versus desktop client email from an ISP)
Yeah, until the day a SOX auditor comes in and says, "show me all your email from 2003 to the present" - at the same time that the service provider's gateway decides to hickup. So sorry - please pay Uncle Sam $14,000,000 for not securing your email documentation. Or maybe your service provider makes a dumb mistake and allows their servers to be hacked -- goodbye data, or more incidiously goodbye data integrity.
If particular data is your life blood you must control it. It is not good enough to prang your provider after the fact in many cases - unless what you do is so trivial that it does not matter.
I think any business that puts control of all of their non-trivial data into the hands of a service provider is asking for problems, legal, operational, and financial.
It's hardly theft when you never had it in the first place.
This makes the big assumption that what they are patenting is truely a new idea. If you look at their patents you will see that most of them entail ideas that have been prior-art for years. The patent office is not doing due diligence during the investigation process because they do not have the expertise. But it doesn't matter because the MS warchest is so deep they can bankrupt just about any competitor by filing frivilous patent violation lawsuits, keeping people tied up in court for years - and Balmer's recent comments seem to bolster this approach.
On the other hand, this begs the question whether software patents are or should be valid anyway. Software is written text - which is already covered by copyright. Patents should only apply to physical things created by people - not the expression of a language. Software patents put a damper on software development and innovation - because most small developers can not afford a patent, much less a patent search to validate their implimentations. Copyright, on the other hand, has minimal cost in comparison - and it does not limit innovation to the same degree as patents. Software patents raise the burden of entry too high for start-ups and squelch competition.
My concern is that the greed of MS will cause the United States to become a second class country in the areas of computer science and software development. If that happens, we will probably see a brain drain of talent to free areas - such as the EU.
I hope I am wrong about this; watching the posturing it is hard to believe otherwise.
Interesting - we used Tanenbaum's "Modern Operating Systems" back in the day.
I am speaking from years of experience - I'm in my 40s. Before I was in school I was into the BBS scene in the 1980s (my early 20s). I remember when a 2400 baud modem was a big deal - and when they came out with the 9600 baud modems - I snatched one up on sale. To make it work I read the manual, and figured out how to tweak the ATT codes to make it signal and handshake properly with the telephone switch and the BBS's modems. I had to immerse myself in the technology and work it out on my own - at least until I could get that first connection. After that, I was able to download information, or put questions up on message boards to get further along.
I guess the point I am trying to make is to be successful in a career field that is over extended, you have to differentiate yourself. I think a key ingredient is to seek knowledge outside of the cirriculum. Get your hands dirty - build a computer from parts, load Linux on it, then write some low level software to interface with the various I/O ports on the system - get two machines to pass information via RS232 cross-over cable. Buy a modem, and work out how to connect to your ISP (if you don't already have broadband). If you plan on being a developer, examine the different design paradigms - and work out your own philosophy. Be the person with the answers and the ability to build those answers into working software systems that are maintainable and resilient, yet only as complex as minimally required (elegant).
You don't have to love what you do - but it sure helps, particularly during the tough stretches when your reach exceeds your grasp. You should be constantly thinking of new ways to use your knowledge outside of school - and make your personal projects difficult, because working through the problems will educate you more than any book.
I've been messing with computers since I was 16 - and I can't imagine not learning something new every day. As you get older, this also keeps you sharp.
I don't know what crackerjack box they got their CS degree out of, but I had to take an RDBMS (SQL) class, and I also had to impliment a database from scratch in another class. Both of these experiences have served me well in the business world. Of course, as new technologies have emerged (object databases, etc) I've learned all I can about it - and in some cases have used the new technology to gain advantages where traditional methods have failed (every problem is not a nail, and every tool is not a hammer).
CS is useful where ever computers and networks reside - particularly when companies are demanding the most bang for the buck. Today developers can not just be code monkeys. They also must understand architectural issues, network issues, storage issues, and how their system will integrate with hetrogenous networks. I've seen too much wasted money and time when neophytes were put in positions beyond their grasp. I hate being right all the time, and then the company eats an unnecessary $3million or so... What is the price of politics and turf wars in the IT business? I haven't stopped counting yet, and the same people keep phuking up without consequence.
Here is what I have learned that hasn't let me down over the years:
In school: Highschool - Basic (interpreter - not VB), Fortran University - RISC assembler, C++, Java, Perl, awk/sed, sh(shell)
On my own: x86 assembler, C, Python + Zope (after getting frustrated with various Java frameworks)
Today I do integration work for a major Fortune 500 company; statistical agregation from hetrogenous systems, installation automation, backup and recovery automation, content management system configuration and administration, and revision control management for our internally built applications and libraries.
emacs is an IDE - and if you do a little digging you will find that you can load various modes that allow you to do just about anything you want to do in other IDEs - including recognizing syntax errors, syntax highlighting, module/class navigation, block modification, searching and generating diffs between buffers/files. Additionally, you can extend emacs by writing your own extensions - thereby gaining even more efficiency.
The key difference between *nix developers and MS developers is the *nix developers know more about the internals of their tools and how to extend them to be more efficient than their MS counterparts. This, in particular, makes *nix developers question MS's sanity when that company does everything in its power to make their skills irrelevant - particularly given the alternative - a dumbed down black box who's behavior can't be clearly defined.
While the text editor and command line will never die, you can't say that the *nix world has stood stagnant either; it is not just about the CLI and text editors.
"I'd go out and buy myself a hammer that already works in spite of the increased immediate cost to myself..."
Your analogy is flawed. Your expensive hammer will only work on plywood (Windows); you won't be able to nail shingles, pine studs, or hang siding. You can build a fast doghouse out of plywood, but you won't be able to easily translate that into a pine doghouse (Linux). On the other hand, there is a large selection of hand tools in a carpenter's tool box with the name 'FOSS' written on it - master craftsman's tools for cutting, shaping, smoothing, fastening - you just have to take some time to learn these tools - and you can become a master at your craft - able to build cross platform applications and increase your real productivity.
Every problem is not a nail, and every tool is not a hammer. If speed is more important than getting it right, then I would stick with the hammer...
Re:Harddrives wear worse from int Amps than 24/7 u
on
Why Vista Took So Long
·
· Score: 1
I replace power supplies on a fairly regular basis (at least once per computer, sometimes twice...
I leave my machines (other than laptops) on 24/7 and have never smoked a power supply in 20 years. I did have to replace a power supply once - HP's engineers decided a 200 watt power supply would be enough for a machine that used close to 300 watts at peak load - the original power supply would trip its circuit breaker when usage would peak (e.g. doing anything CPU intensive, reading/writing HD and CDROM at the same time). Even then, the power supply would come back online after waiting for the circuit breaker to cool down - so it wasn't smoked.
Additionally, of the thousands of servers and routers used by my company running 24/7, I think we might have lost 2 power supplies over the past 5 years - and I would scratch those up to manufacturer's defect.
If you are smoking power supplies 'on a fairly regular basis' - maybe you should look at ways to regulate the power coming into your systems (power spikes etc?), and rethink your policy of powering everything up and down on a daily basis. Another issue might be that the power supply you are using is not rated for the equipment it is powering (as in my HP example) - you could be overloading your powersupply, and just because it came from an OEM doesn't mean it is designed correctly (as in my example - I added up the power requirements of all the components on my system and it was a number quite a bit north of 200 watts - and mind you, the machine was equipped from the factory, I didn't add any new components).
Looking at this from 30,000 feet it occurs to me that the FOSS world has not applied the winning combinations used in other FOSS projects to gaming. Here is what I think we need to restart linux gaming's heart:
1. A selection of FOSS high performance game engines (for small games that need high triangle counts, an engine for high performance and lower triangle counts, and an engine for maximum network performance - very low triangle counts or other tricks to maximize FPS and network throughput) - designed to work on multiple platforms (linux, windows, et al). If these game engines are the best of breed, they will be used by the game development community.
2. A DirectX to SDL translator to lower the cost of entry into the Linux market for the 99% of development shops that develop for Windows - and have a huge investment there.
These two things would do much toward bringing popular Windows games to Linux.
"Therefore, our job has got to be to help our customers get interoperability. And, of course, all vendors secretly are wondering when they do interoperability, did I do something that's going to help me win more of their customers, or something that's going to help them win more of my own customers." - Balmer
Doesn't this exactly describe their long term strategy of embrace, extend, extenguish? As was predicted, as the tide changes (software becomes more commoditized) Microsoft will pull out the big guns to keep the money flowing into their organization - regardless of whether they earned it or not.
"Linux comes from the community -- the fact that that product uses our patented intellectual property is a problem for our shareholders. We spend $7 billion a year on R&D, our shareholders expect us to protect or license or get economic benefit from our patented innovations. So how do we somehow get the appropriate economic return for our patented innovation, and how do we do interoperability... (With Novell) we agreed on a, we call it an IP bridge, essentially an arrangement under which they pay us some money for the right to tell the customer that anybody who uses Suse Linux is appropriately covered. There will be no patent issues. They've appropriately compensated Microsoft for our intellectual property, which is important to us. In a sense you could say anybody who has got Linux in their data center today sort of has an undisclosed balance sheet liability, because it's not just Microsoft patents. Because of the way open-source works, there's nobody who's been able to do patent coverage or patent indemnification behind that." - Balmer
It should be noted, the group he was talking to was composed of CIOs for companies that primarily have Windows in their data centers. His message was to them: if you don't pay me protection money (via using SUSE - or better yet, removing Linux from your data center), then I'll get my cousin Vinnie here (Microsoft Lawyers) to break your legs (sue your pants off over patent violations); on top of that, you won't be sure that you will be able to interoperate with anything else other than SUSE!
As for Python though it's a different story since it's interpreted.
Actually, Python is byte-code compiled at run-time. You can see this in action after you run a Python app one time, you will see a.pyc file - that is the compiled version which gets executed by the Python virtual machine.
Also remember - taking the exact same algorithm between different languages may be misleading. In the case of python, there are more 'pythonic' language constructs that obviate the need to use particular algorithms - and also may provide speed above and beyond what you would find using the traditional algorithm.
So, this, for instance, is a Pythonic way of getting multiple return values from function:
def something():
return "Foo", "Bar"
foo, bar = something()
whereas this is not:
def something(l):
l.append("Foo", "Bar")
l = []
something(l)
foo = l[0]
bar = l[1]
Someone coming from a language like C or C++ might however have designed something like the latter.
The reason most people load the proprietary 'blob' is to use all the capabilities of their card for 3D games (WOW, FPSs etc...). For them '3d operation' are part of their day-to-day desktop experience!
Until high-end capabilities of the latest video cards are opened up to open source driver development (and standardization) I think the linux desktop will not be the solution for people who can only afford one machine.
Don't get me wrong - I have 5 machines, 4 of which run Linux...but that 5th machine has my high-end video and sound cards coupled with a competitive upgradeable motherboard/cpu/ram combo - and it runs Windows 2000. Until the linux desktop makes the whole package - including gaming - easy for the average user, Windows will continue to control the home market imho.
Slashdot Mirror
Mongrel --> Ruby (Rails)
Medusa --> Python (Zope)
They can have my mouse when they pry it from my cold dead fingers...
I could just sell my game box and buy a laptop - write a novel, and retire to the local jazz club on a nightly basis to soak my troubles in scotch?
The scotch is looking mighty good right now...
Another benefit would be the ability to migrate the client to other platforms, and hedge against forced obsolescence. If you and your friends (or larger group) didn't want to upgrade your gear, you could always fork an earlier version of the engine and host your own game for the FPS challenged - unlike commercial games that evolve away and above marginal users ability to keep up.
As a pure money-making enterprise, PC game development shops focus their dollars on the platform most likely to provide a return on their investment - e.g. MS Windows - with a smaller, lagging amount reserved for Apple (if they even support Apple), and next to nothing for FOSS (Linux/BSD et al). I know game shops that ported early versions of their games 5+ years ago to Linux, but dropped the support due to a lack of return on the investment.
My thought is if FOSS makes it easy for developers to port their current Windows apps as a planned adjunct to their existing development, and also have key tools/engines that are open source, we can attract more of the mainstream game producers to make games for Linux, or at a minimum port their existing Windows games to Linux. By lowering the bar to entry into the FOSS market, they would be more likely to take advantage of it. Many of them probably won't due to percieved threats to their IP and competitive advantage. My hope would be the titles that I enjoy would take advantage of it to get my entertainment dollars; if they did I could remove the last vestage of MS Windows from my network, and gain the benefits of my high-end system for things other than gaming and listening to music.
Failing that or in addition to that - the FOSS world would at least have the tools to build our own stuff on par with mainstream games, without having to reinvent the wheel, and different groups could handle the upgrade path as they desired. So I see it as valuable from that perspective.
So, you are the guy I always have to clean up behind when your applications break...
Since when has the 'Real News' not been fake?
You are right - I'm picking a nit.
You know, you can run emacs remotely via X - provided it is loaded on the remote system - which will give you all the syntax highlighting goodness (I do this all the time). It also allows you to use your mouse on the remote instance (particularly useful when you forget what ctrl-x ctrl-b, ctrl-x o or ctrl-x ctrl-s does).
That would be the worse thing to tell a Congresstroll. They would be falling all over themselves to write legislation resurrecting the Clipper chip, giving Bill Gates bags full of money to DRM the hell out of it with 'trusted computing', and laws requiring only *approved* software and hardware on any network! All wireless routers, cell phones and laptops would be contraband (except for military/governmental use, of course).
Alright --- who threw the wombat...?
I see this as a larger pattern. Big companies are losing mind-share by outsourcing system development. If taken to an extreme the company has no technical knowledge and does nothing in house, instead it is just a collection of project managers and lawyers - a paper tiger - at great risk of being crushed by savvy competitors.
Rule No 7: Nobody talks about fight club.
Yeah, until the day a SOX auditor comes in and says, "show me all your email from 2003 to the present" - at the same time that the service provider's gateway decides to hickup. So sorry - please pay Uncle Sam $14,000,000 for not securing your email documentation. Or maybe your service provider makes a dumb mistake and allows their servers to be hacked -- goodbye data, or more incidiously goodbye data integrity.
If particular data is your life blood you must control it. It is not good enough to prang your provider after the fact in many cases - unless what you do is so trivial that it does not matter.
I think any business that puts control of all of their non-trivial data into the hands of a service provider is asking for problems, legal, operational, and financial.
Actually I wouldn't notice - I touch type.
This makes the big assumption that what they are patenting is truely a new idea. If you look at their patents you will see that most of them entail ideas that have been prior-art for years. The patent office is not doing due diligence during the investigation process because they do not have the expertise. But it doesn't matter because the MS warchest is so deep they can bankrupt just about any competitor by filing frivilous patent violation lawsuits, keeping people tied up in court for years - and Balmer's recent comments seem to bolster this approach.
On the other hand, this begs the question whether software patents are or should be valid anyway. Software is written text - which is already covered by copyright. Patents should only apply to physical things created by people - not the expression of a language. Software patents put a damper on software development and innovation - because most small developers can not afford a patent, much less a patent search to validate their implimentations. Copyright, on the other hand, has minimal cost in comparison - and it does not limit innovation to the same degree as patents. Software patents raise the burden of entry too high for start-ups and squelch competition.
My concern is that the greed of MS will cause the United States to become a second class country in the areas of computer science and software development. If that happens, we will probably see a brain drain of talent to free areas - such as the EU.
I hope I am wrong about this; watching the posturing it is hard to believe otherwise.
Interesting - we used Tanenbaum's "Modern Operating Systems" back in the day.
I am speaking from years of experience - I'm in my 40s. Before I was in school I was into the BBS scene in the 1980s (my early 20s). I remember when a 2400 baud modem was a big deal - and when they came out with the 9600 baud modems - I snatched one up on sale. To make it work I read the manual, and figured out how to tweak the ATT codes to make it signal and handshake properly with the telephone switch and the BBS's modems. I had to immerse myself in the technology and work it out on my own - at least until I could get that first connection. After that, I was able to download information, or put questions up on message boards to get further along.
I guess the point I am trying to make is to be successful in a career field that is over extended, you have to differentiate yourself. I think a key ingredient is to seek knowledge outside of the cirriculum. Get your hands dirty - build a computer from parts, load Linux on it, then write some low level software to interface with the various I/O ports on the system - get two machines to pass information via RS232 cross-over cable. Buy a modem, and work out how to connect to your ISP (if you don't already have broadband). If you plan on being a developer, examine the different design paradigms - and work out your own philosophy. Be the person with the answers and the ability to build those answers into working software systems that are maintainable and resilient, yet only as complex as minimally required (elegant).
You don't have to love what you do - but it sure helps, particularly during the tough stretches when your reach exceeds your grasp. You should be constantly thinking of new ways to use your knowledge outside of school - and make your personal projects difficult, because working through the problems will educate you more than any book.
I've been messing with computers since I was 16 - and I can't imagine not learning something new every day. As you get older, this also keeps you sharp.
That is my advise to you. Good luck!
I don't know what crackerjack box they got their CS degree out of, but I had to take an RDBMS (SQL) class, and I also had to impliment a database from scratch in another class. Both of these experiences have served me well in the business world. Of course, as new technologies have emerged (object databases, etc) I've learned all I can about it - and in some cases have used the new technology to gain advantages where traditional methods have failed (every problem is not a nail, and every tool is not a hammer).
CS is useful where ever computers and networks reside - particularly when companies are demanding the most bang for the buck. Today developers can not just be code monkeys. They also must understand architectural issues, network issues, storage issues, and how their system will integrate with hetrogenous networks. I've seen too much wasted money and time when neophytes were put in positions beyond their grasp. I hate being right all the time, and then the company eats an unnecessary $3million or so... What is the price of politics and turf wars in the IT business? I haven't stopped counting yet, and the same people keep phuking up without consequence.
Here is what I have learned that hasn't let me down over the years:
In school:
Highschool - Basic (interpreter - not VB), Fortran
University - RISC assembler, C++, Java, Perl, awk/sed, sh(shell)
On my own:
x86 assembler, C, Python + Zope (after getting frustrated with various Java frameworks)
Today I do integration work for a major Fortune 500 company; statistical agregation from hetrogenous systems, installation automation, backup and recovery automation, content management system configuration and administration, and revision control management for our internally built applications and libraries.
emacs is an IDE - and if you do a little digging you will find that you can load various modes that allow you to do just about anything you want to do in other IDEs - including recognizing syntax errors, syntax highlighting, module/class navigation, block modification, searching and generating diffs between buffers/files. Additionally, you can extend emacs by writing your own extensions - thereby gaining even more efficiency.
The key difference between *nix developers and MS developers is the *nix developers know more about the internals of their tools and how to extend them to be more efficient than their MS counterparts. This, in particular, makes *nix developers question MS's sanity when that company does everything in its power to make their skills irrelevant - particularly given the alternative - a dumbed down black box who's behavior can't be clearly defined.
While the text editor and command line will never die, you can't say that the *nix world has stood stagnant either; it is not just about the CLI and text editors.
I leave my machines (other than laptops) on 24/7 and have never smoked a power supply in 20 years. I did have to replace a power supply once - HP's engineers decided a 200 watt power supply would be enough for a machine that used close to 300 watts at peak load - the original power supply would trip its circuit breaker when usage would peak (e.g. doing anything CPU intensive, reading/writing HD and CDROM at the same time). Even then, the power supply would come back online after waiting for the circuit breaker to cool down - so it wasn't smoked.
Additionally, of the thousands of servers and routers used by my company running 24/7, I think we might have lost 2 power supplies over the past 5 years - and I would scratch those up to manufacturer's defect.
If you are smoking power supplies 'on a fairly regular basis' - maybe you should look at ways to regulate the power coming into your systems (power spikes etc?), and rethink your policy of powering everything up and down on a daily basis. Another issue might be that the power supply you are using is not rated for the equipment it is powering (as in my HP example) - you could be overloading your powersupply, and just because it came from an OEM doesn't mean it is designed correctly (as in my example - I added up the power requirements of all the components on my system and it was a number quite a bit north of 200 watts - and mind you, the machine was equipped from the factory, I didn't add any new components).
Looking at this from 30,000 feet it occurs to me that the FOSS world has not applied the winning combinations used in other FOSS projects to gaming. Here is what I think we need to restart linux gaming's heart:
1. A selection of FOSS high performance game engines (for small games that need high triangle counts, an engine for high performance and lower triangle counts, and an engine for maximum network performance - very low triangle counts or other tricks to maximize FPS and network throughput) - designed to work on multiple platforms (linux, windows, et al). If these game engines are the best of breed, they will be used by the game development community.
2. A DirectX to SDL translator to lower the cost of entry into the Linux market for the 99% of development shops that develop for Windows - and have a huge investment there.
These two things would do much toward bringing popular Windows games to Linux.
Doesn't this exactly describe their long term strategy of embrace, extend, extenguish? As was predicted, as the tide changes (software becomes more commoditized) Microsoft will pull out the big guns to keep the money flowing into their organization - regardless of whether they earned it or not.
It should be noted, the group he was talking to was composed of CIOs for companies that primarily have Windows in their data centers. His message was to them: if you don't pay me protection money (via using SUSE - or better yet, removing Linux from your data center), then I'll get my cousin Vinnie here (Microsoft Lawyers) to break your legs (sue your pants off over patent violations); on top of that, you won't be sure that you will be able to interoperate with anything else other than SUSE!
Actually, Python is byte-code compiled at run-time. You can see this in action after you run a Python app one time, you will see a
http://wiki.python.org/moin/PythonSpeed
Also remember - taking the exact same algorithm between different languages may be misleading. In the case of python, there are more 'pythonic' language constructs that obviate the need to use particular algorithms - and also may provide speed above and beyond what you would find using the traditional algorithm.
So, this, for instance, is a Pythonic way of getting multiple return values from function:
def something():
return "Foo", "Bar"
foo, bar = something()
whereas this is not:
def something(l):
l.append("Foo", "Bar")
l = []
something(l)
foo = l[0]
bar = l[1]
Someone coming from a language like C or C++ might however have designed something like the latter.
Here is a link to comparisons of python with other languages for your edification: http://wiki.python.org/moin/LanguageComparisons
The reason most people load the proprietary 'blob' is to use all the capabilities of their card for 3D games (WOW, FPSs etc...). For them '3d operation' are part of their day-to-day desktop experience!
Until high-end capabilities of the latest video cards are opened up to open source driver development (and standardization) I think the linux desktop will not be the solution for people who can only afford one machine.
Don't get me wrong - I have 5 machines, 4 of which run Linux...but that 5th machine has my high-end video and sound cards coupled with a competitive upgradeable motherboard/cpu/ram combo - and it runs Windows 2000. Until the linux desktop makes the whole package - including gaming - easy for the average user, Windows will continue to control the home market imho.