"Preston: When do you think Linux will take over desktop market from Microsoft?
Linus Torvalds: Oh, I think it's started already, it's just slow."
I think Linus misunderstood the question. Nobody doubts that there are people who have switched from Windows to Linux, the question is if and when Linux will surpass Windows on the desktop.
Usually mass storage refers to something that requires a file system, which a flash ROM does not. Perhaps I should have said file system.
"Video: Maybe not VGA or direct viewing, but depending on the router a serial port needs to be available. Just for text displayed elsewhere granted, but there is a certain amount of I/O programming involved."
Well, that's a serial port, not video.
"The real question is if you were to program a firewall application to run on the router what would you need to include in it? TCP/IP Stack, Device Drivers, Memory Management, Storage support (*PROM/CF/etc), etc."
I think you may have a point about the TCP/IP stack because as I mentioned before, it's difficult to implement directly from the spec. As for the others: Device Drivers are really just hardware interface routines that have extra code to conform to the OS's device model; I guess you mean Memory allocation - often memory can be allocated in a static fashion for dedicated devices since the worst case memory usage is usually the normal case.
"Do you think it is economical for Linksys/Dlink/whoever to build an os/program for something that will sell for so little?"
Well, I don't know, but the point of my original post was that an OS-less device would be more secure since it would require reverse engineering for many exploits. It's possible that customers might be willing to pay more for a more secure device.
So why not work on closing the gap between Linux in AIX instead? It's not that I blame IBM for supporting AIX, I'm just challenging IBM's PR campaign about how pro OSS they are and some Slashdotters' promotion of IBM as the corporate poster boy for OSS. Clearly IBM has a dual Unix strategy for the foreseeable future.
So, if IBM is really embracing Linux, why spend the time and money to certify AIX. They could have spent it on Linux development. Doesn't the certification devalue Linux a bit by comparison?
"So if the even bytes are put into one file and the odd bytes are put into another file the application is no longer monolithic?"
Unless you are using machine code (not even assembly), you don't specify your source code byte-by-byte, so I don't see your point.
"If you put all the source for QNX into one file, that will not make QNX monolithic."
Well, actually it does unless you imagine some strange build process that doesn't exist. How many output files do you expect to generate from a single input file?
In any case, this early definition of monolithic was based on the idea that putting all the source in a single file would lead to a lot of global data and wouldn't support code reuse.
"OK. Remove IE. Boot without a GUI. Change libraries that are currently in use while the system is running."
This is supposed to be your criteria for non-monolithic? Why don't you start with a definition of what you believe monolithic means and then we see if it is the generally accepted definition and how Windows and Linux stack up against it. I'm sure we can come up with a particular definition of monolithic that Linux would be included in.
Is the functionality of a router or firewall so complex or general that it requires an OS? It seems to me that a router without an OS would be potentially the most secure since it would be the hardest to reverse engineer and known vunerabilities of other OS's wouldn't be applicable.
"The renderer adequately makes up for errors in the HTML code, and consequently there are fewer crashes."
This suggests that the solution for preventing crashes is to be liberal about what HTML you allow. You were making excuses for the other browsers implying that they were crashing because of their strict HTML handling, but the fact is that they could have posted an error rather than crashing and still been just as strict.
So your theory is that having access to the source of these browsers has enabled an individual to crash them, but having thousands of people have the source code hasn't resulted in the problems being fixed already.
There's a difference between posting an error message and crashing. So I don't think there's any upside for the other browsers on this issue.
As for allowing sloppy HTML, that's a problem if you think the primary purpose of a browser is to enforce HTML standards rather than display web pages.
"Well, for a comparison, you could ask the folks at Apple this question. I'm sure they would say OSX is an improvement."
Gee, I'm sure Apple would tell us if they had any evidence that OSX was not an improvement. In any case, I don't see how you can draw conclusions about XP to Unix based on Mac OS to OSX.
"However, Clayton Christensen's point is that Microsoft's monopoly is not going to last forever, and that therefore they should learn to love Linux sooner as opposed to later (as Apple did), or pay the price. In that case, the switch would make good business sense."
The problem with this theory is that very few monopolies have succeeded by embracing challenger's ideas. Both AT&T and Xerox tried to reinvent themselves and are currently only shadows of their glory days. Why should MS give up all the big profits of today just to be another commodity Linux company of the future? It's not as if it takes years to become a Linux player anyway - they have plenty of time if Linux actually becomes the dominant OS (which still speculation at this point).
"I reckon that another twenty years should take us to where the automobile industry was in the 50s."
Writing software is never going to be like designing and building cars no matter how long you wait. Perhaps what is really holding us back is failing to recognize the unique characteristics of software and falling back on inappropriate methods from other fields. It's not that our industry or methods are primitive, it's that our goals are much more difficult to achieve than rolling out the next gas-gusler.
I'm not an expert on Apple systems but I suspect that many applications that ran on the original Mac won't run on OSX.
Windows XP can run applications designed for an 8088, would it really be that easy to write an new Unix based OS that is capable of that level of compatibility? Would the result really be more secure, faster and more reliable than Windows XP? How many years would it take for MS to recover the costs of developing it. At the end of the day this has to make good business sense.
You think just because MS offered a version of Linux that everyone would buy it even though it's not compatible with current Windows applications? MS has made more money on Windows in the last 3 years than they would selling Linux for a decade.
"History is full of companies who fell out of the limelight because they couldn't or wouldn't adapt to new technology."
What new technology has MS failed to adopt? It can't be Linux because Unix is essentially 1970's technology. That's not to say that there hasn't been improvements since then, but there's hardly any cutting-edge technology that's unique to Linux.
Red Hat and others would be better examples of an open source business model if they had invested their own time and money to create Linux and then opened the source. Of course, that's not the way Linux was developed. Linux was essentially functional before any business was involved with it.
Now that open source projects are well known, it will be harder for companies will be able to "brand" them the way Red Hat has Linux. That branding is the key to Red Hat's success (Although I don't know if Red Hat is profitable overall; i.e. do profits over the life of the company exceeded investments over the life of the company.)
The criteria was not that IBM once sold a product, but that the product was profitable.
Let me give you a counter-example:
IBM bought Rational a couple years back. This included not only Rational's main testing software, Rational Robot, but Rational Visual Test, a low cost testing product with a large following.
Shortly after acquiring Rational, IBM stopped selling Visual Test. Not only did they not release the source, but they won't allow anyone to even distribute the binary. Why? Because VT is a bargin that threatens their overpriced testing suite.
So, I return to my previous point. IBM isn't serious about open source not matter how much unprofitable junk they release.
Slashdot Mirror
"Preston: When do you think Linux will take over desktop market from Microsoft?
Linus Torvalds: Oh, I think it's started already, it's just slow."
I think Linus misunderstood the question. Nobody doubts that there are people who have switched from Windows to Linux, the question is if and when Linux will surpass Windows on the desktop.
"Mass storage: what about the flash ROM?"
Usually mass storage refers to something that requires a file system, which a flash ROM does not. Perhaps I should have said file system.
"Video: Maybe not VGA or direct viewing, but depending on the router a serial port needs to be available. Just for text displayed elsewhere granted, but there is a certain amount of I/O programming involved."
Well, that's a serial port, not video.
"The real question is if you were to program a firewall application to run on the router what would you need to include in it? TCP/IP Stack, Device Drivers, Memory Management, Storage support (*PROM/CF/etc), etc."
I think you may have a point about the TCP/IP stack because as I mentioned before, it's difficult to implement directly from the spec. As for the others: Device Drivers are really just hardware interface routines that have extra code to conform to the OS's device model; I guess you mean Memory allocation - often memory can be allocated in a static fashion for dedicated devices since the worst case memory usage is usually the normal case.
"Do you think it is economical for Linksys/Dlink/whoever to build an os/program for something that will sell for so little?"
Well, I don't know, but the point of my original post was that an OS-less device would be more secure since it would require reverse engineering for many exploits. It's possible that customers might be willing to pay more for a more secure device.
So why not work on closing the gap between Linux in AIX instead? It's not that I blame IBM for supporting AIX, I'm just challenging IBM's PR campaign about how pro OSS they are and some Slashdotters' promotion of IBM as the corporate poster boy for OSS. Clearly IBM has a dual Unix strategy for the foreseeable future.
So, if IBM is really embracing Linux, why spend the time and money to certify AIX. They could have spent it on Linux development. Doesn't the certification devalue Linux a bit by comparison?
"So if the even bytes are put into one file and the odd bytes are put into another file the application is no longer monolithic?"
Unless you are using machine code (not even assembly), you don't specify your source code byte-by-byte, so I don't see your point.
"If you put all the source for QNX into one file, that will not make QNX monolithic."
Well, actually it does unless you imagine some strange build process that doesn't exist. How many output files do you expect to generate from a single input file?
In any case, this early definition of monolithic was based on the idea that putting all the source in a single file would lead to a lot of global data and wouldn't support code reuse.
I don't see how the fundamentals require an OS. Routers and Firewalls don't require support for:
Mass storage
Video
Audio
Printing
Multiple Users
Certainly the need for multiple routing algorithms doesn't imply the need for an OS in my view.
Great. Another post without a definition of monolithic.
Let me give you one of the earliest definitions: When an application uses a single file for all of its source code, it's a monolithic application.
"OK. Remove IE. Boot without a GUI. Change libraries that are currently in use while the system is running."
This is supposed to be your criteria for non-monolithic? Why don't you start with a definition of what you believe monolithic means and then we see if it is the generally accepted definition and how Windows and Linux stack up against it. I'm sure we can come up with a particular definition of monolithic that Linux would be included in.
Is the functionality of a router or firewall so complex or general that it requires an OS? It seems to me that a router without an OS would be potentially the most secure since it would be the hardest to reverse engineer and known vunerabilities of other OS's wouldn't be applicable.
People who can draw a vivid and exciting mental picture that convinces other people to invest their money in a dubious enterprise.
I assume you meant to reply to metlin instead of me. I agree with you (as my other posts in this thread indicate).
"Would you accept 6mm nuts for 8mm screws?"
A better question is "Would you prefer a product that can accept 6mm nuts or 8mm nuts for an 8mm screw?"
MS isn't selling HTML.
"The renderer adequately makes up for errors in the HTML code, and consequently there are fewer crashes."
This suggests that the solution for preventing crashes is to be liberal about what HTML you allow. You were making excuses for the other browsers implying that they were crashing because of their strict HTML handling, but the fact is that they could have posted an error rather than crashing and still been just as strict.
So your theory is that having access to the source of these browsers has enabled an individual to crash them, but having thousands of people have the source code hasn't resulted in the problems being fixed already.
There's a difference between posting an error message and crashing. So I don't think there's any upside for the other browsers on this issue.
As for allowing sloppy HTML, that's a problem if you think the primary purpose of a browser is to enforce HTML standards rather than display web pages.
The best part is people don't bother to steal it and you don't worry about babysitting it. One less thing to stress over.
"Well, for a comparison, you could ask the folks at Apple this question. I'm sure they would say OSX is an improvement."
Gee, I'm sure Apple would tell us if they had any evidence that OSX was not an improvement. In any case, I don't see how you can draw conclusions about XP to Unix based on Mac OS to OSX.
"However, Clayton Christensen's point is that Microsoft's monopoly is not going to last forever, and that therefore they should learn to love Linux sooner as opposed to later (as Apple did), or pay the price. In that case, the switch would make good business sense."
The problem with this theory is that very few monopolies have succeeded by embracing challenger's ideas. Both AT&T and Xerox tried to reinvent themselves and are currently only shadows of their glory days. Why should MS give up all the big profits of today just to be another commodity Linux company of the future? It's not as if it takes years to become a Linux player anyway - they have plenty of time if Linux actually becomes the dominant OS (which still speculation at this point).
"I reckon that another twenty years should take us to where the automobile industry was in the 50s."
Writing software is never going to be like designing and building cars no matter how long you wait. Perhaps what is really holding us back is failing to recognize the unique characteristics of software and falling back on inappropriate methods from other fields. It's not that our industry or methods are primitive, it's that our goals are much more difficult to achieve than rolling out the next gas-gusler.
I'm not an expert on Apple systems but I suspect that many applications that ran on the original Mac won't run on OSX.
Windows XP can run applications designed for an 8088, would it really be that easy to write an new Unix based OS that is capable of that level of compatibility? Would the result really be more secure, faster and more reliable than Windows XP? How many years would it take for MS to recover the costs of developing it. At the end of the day this has to make good business sense.
You think just because MS offered a version of Linux that everyone would buy it even though it's not compatible with current Windows applications? MS has made more money on Windows in the last 3 years than they would selling Linux for a decade.
"History is full of companies who fell out of the limelight because they couldn't or wouldn't adapt to new technology."
What new technology has MS failed to adopt? It can't be Linux because Unix is essentially 1970's technology. That's not to say that there hasn't been improvements since then, but there's hardly any cutting-edge technology that's unique to Linux.
Red Hat and others would be better examples of an open source business model if they had invested their own time and money to create Linux and then opened the source. Of course, that's not the way Linux was developed. Linux was essentially functional before any business was involved with it.
Now that open source projects are well known, it will be harder for companies will be able to "brand" them the way Red Hat has Linux. That branding is the key to Red Hat's success (Although I don't know if Red Hat is profitable overall; i.e. do profits over the life of the company exceeded investments over the life of the company.)
Eclipse is primarily an anti-Sun initiative wrapped in an open source friendly candy shell.
Sorry, I don't get the Mr. Schwartz reference.
I didn't say that all Free Software was "unprofitable junk. Distortion of my post noted. Thanks for the disinformation.
The criteria was not that IBM once sold a product, but that the product was profitable.
Let me give you a counter-example:
IBM bought Rational a couple years back. This included not only Rational's main testing software, Rational Robot, but Rational Visual Test, a low cost testing product with a large following.
Shortly after acquiring Rational, IBM stopped selling Visual Test. Not only did they not release the source, but they won't allow anyone to even distribute the binary. Why? Because VT is a bargin that threatens their overpriced testing suite.
So, I return to my previous point. IBM isn't serious about open source not matter how much unprofitable junk they release.