The good news is that the "winner" in the computer world is hardly ever the guy with the best product. Instead the folks that win generally end up being the folks with the cheapest product that is "good enough." If usability or innovation mattered then the early Macintoshes would have destroyed their DOS based PC competitors. The problem is that most people aren't really willing to pay extra for innovation or usability. They simply want something that will get the job done at the lowest price.
Don't get me wrong. I am not saying that Free Software doesn't innovate or anything like that. In the long run the ability to take existing software and innovate on top of it (instead of starting from scratch) is going to be a huge win for hackers everywhere. Once OpenOffice.org (or whatever) becomes wide spread then all sorts of cool things are going to be possible just because anyone with a crazy idea as to how office suites should work will actually be able to try those ideas out. Most of these ideas will be crap, but the wider range of ideas will still almost certainly be a net win. The fact of the matter is that Microsoft has been lifting ideas from other software developers and implementing then in Windows and Office for years. You would be hard pressed to point to a single major feature that Microsoft actually pioneered.
Microsoft has made a living by being "good enough" and less expensive, and for years the dominated the desktop despite the fact that Windows (and DOS before that) were pathetic knockoffs of other people's innovation. Now Windows is finally getting to be pretty good, but Linux is cheaper, and for an increasingly large group of people it is becoming "good enough."
The BSD TCP/IP stack was meant to be reusable. The folks working on it were part of a U.S. government project to create a standard networking stack that would be useable on a wide variety of OSes.
The problem is that companies are finding that they do have a choice, and the choice is to replace their expensive Sun gear with a pile of inexpensive Linux boxes. And before you say "impossible," you should talk to your Oracle dealer. They will happily tell you that 16 4-way Linux boxes will outperform a 64-way Sun box, and be more reliable (in the aggregate) to boot.
Sun's high profit margins low volume strategy has left them vulnerable to companies that can get away with lower profit margins. Red Hat has basically gone into the "convert from Sun and save a bundle" business.
The difference is that you need a web browser anyhow. No one needs Java on their desktop.
Besides which, it is far easier to create a GUI in XUL + Javascript (and coming soon Python) than it is to create a Java GUI. Basically the idea is to lower the bar for application development so that the folks currently doing HTML + Javascript can create rich client front ends.
If anything this just goes to show how much the average consumer cares about usability. Most consumers don't really care how usable their software is. Usability and $0.50 will get you a Snickers bar.
Don't get me wrong. I think that Apple really does have the edge when it comes to making usable systems. Especially if you don't have to share documents and files with Windows users. However, when push comes to shove, consumers want "usable enough" at the lowest price, and that's not Apple.
Not to mention the fact software compiled with gcj or linked with libgcj don't fall under the GPL. You can write proprietary software and compile with gcj and not be "infected" by the GPL. So this part of Gosling's anti-RMS rant is pure FUD.
The real problem is that after all of the work that Sun has put into making Java a platform in real life Java is currently splitting into a million different directions. gcj and GNU Classpath are picking up steam, IBM is pushing platform dependent SWT and Eclipse instead of Swing, etc. With Sun losing the hardware war to Intel and AMD, and the UNIX war to Linux, that leaves Sun with Java as its best hope for a recovery. However, it's a pretty slim hope. Java application servers are basically a comodity as are Java development tools.
Yes, people still run stuff on NT4, but new projects aren't being rolled out on NT4. Even worse, folks that are currently running NT4 almost certainly are working on migration plans.
With programming languages (like operating systems) marketshare is not the critical statistic. No one cares if your application contains 1 hojillion lines of COBOL if all of your new development is in Python. The important factor is how much *new* development is being done in the language.
Yes both Python and Perl have modules that only run on one platform or another. Of course, there are Java libraries that likewise only run on one platform or another (like the JavaCOM libraries). For the most part, however, "porting" Python or Perl software consists of installing Python or Perl on the machine in question and copying over the files.
As an added bonus Python and Perl run on a lot more platforms than Sun's JVM. Python and Perl, for instance, are well supported on all of the BSDs, and Java isn't. So much for Java's touted WORA abilities.
This is precisely why I am not interested in eBook readers. Not only do you end up with a single purpose device, but you probably get stuck with DRM and Windows-only software as well.
Personally, I have switched almost 100% to reading on my Visor Handspring (the B&W Clie my last job provided was better, but the Visor is good enough). I only purchase material that is available in open formats (fictionwise.com and baen.com are good sources). Throw in stuff from Project Gutenberg and I have read nearly 100 ebooks to date (and for less than $100 too).
The benefits of using a PDA are various. I can read in the dark, I always have my reading material with me, and I can read "discreetly." More importantly, 10 years from now I will still have access to all of these books in electronic format. I like reading on my Visor so much that I won't read a new author unless I can get his or her work in an open ebook format first.
I am not holding my breath. There are several good reasons why those who have been singled out have settled. The most important reason that everyone has settled has to do with the fact that so far the RIAA has had a good case against everyone they have sued. Basically the RIAA compiles an extensive set of evidence that the person targetted is responsible for distributing quadzillions of dollars worth of music. Then they threaten to go after the person using laws that were designed to keep mobsters from going into the bootleg music business. When faced with the prospect of going to court and facing hundreds of thousands of dollars in fines (and possibly felony distribution of copyrighted material charges) or paying $3000 the rational choice is obvious. Even rich kids aren't stupid.
So it's not a question of waiting for a rich kid that is willing to fight. It's more about waiting for the RIAA to finger the wrong person. The RIAA cases are high profile enough that someone wrongfully accused could almost certainly get some help paying for legal expenses. The catch is that the RIAA would have to pick on someone who isn't guilty.
Personally, I am glad to see the RIAA using existing laws to curb the sharing of their copyrighted material. I would much rather see them use existing laws than to see them push for new laws requiring universal DRM.
Novell probably will be a little more successful than Red Hat simply because they now employ both the folks at Ximian and the bulk of the KDE hackers (who used to work for SuSE). Red Hat, on the other hand, employed very few KDE hackers (and the one outspoken KDE hacker they did employ quit:).
My guess is that the folks at Ximian and SuSE are likely to see more eye to eye seeing as how their paycheck will depend on them getting along.
This isn't about freedom of speech or freedom of religion. "This is f*cking brilliant" is not protected speech, never has been, and never will be, and you don't have to be religious to find certain phrases vulgar or offensive.
You can rant and rave all you want about this, but there is plenty of legal precedence for what can be considered protected speech. To use an example that/. readers can comprehend SCO can not claim that "All your Linux base belong to us," and then pretend they are merely exercising their right to free speech when it is shown that in fact they don't own all of your Linux bases.
You might disagree with this, and you might even feel that the use of vulgar language should be encouraged over public broadcasts. However, the vast majority of folks disagree with you.
The problem with this sort of tactic is that it is easily abused. If all it takes to get fat cash from the RIAA is to set up a website that sells music from independent artists then you can bet that there will soon be a landrush to create websites that sell music by independent artists.
If sites thut cut out the RIAA middleman become popular there is very little that the recording industry can do besides lower their profit margins and try to compete on price. There's room for markup in the music promotion business, but there's not as much room as there used to be. Eventually this is going to catch up to the RIAA no matter what they do. The price of promoting and distributing music has gone way down, and if the RIAA doesn't adjust their prices to match, then their competitors will start to see success.
Exactly. Novell has plenty of good reasons to give away YaST. First of all they are seeing that RedHat's GPLed installer is already being integrated into other distributions (like Debian). SuSE and Caldera had better distributions than RedHat for years, but RedHat won because every time that someone wanted to borrow code they borrowed it from RedHat. If RedHat's installer becomes ubiquitous then it will be the one folks use, it will also receive the lion's share of debugging efforts. IMHO SuSE/Novell is already late to the party on this one.
The money in Linux rollouts lies in support contracts and value-added software (think eDirectory, Groupwise, etc.). Distributions are nothing more than inexpensive marketing. The further YaST spreads the better chances Novell has of getting sysadmins hooked on their tools.
What management really wants to do is automate the docks (like the East Coast) fire half of the workers and break the union. The Union did what it normally does, which is to fight for the status quo.
Re:It's the little things....
on
GTK 2.4.0 Released
·
· Score: 2, Informative
Code reuse is the "good" sort of laziness. It saves time, memory (shared libs are good), and it allows for easier centralized testing. Basically code reuse is the sort of laziness that saves time and energy for everyone involved.
Not using a packaging manager is the "bad" sort of laziness.
That's just another major design difference between Windows and UNIX. The folks at Microsoft are concerned about binary compatibility, and the UNIX folks (well, at least the Linux and BSD folks) are more concerned about source compatibility. Who cares if your old libc5 application doesn't work under a new version of Linux if you can simply type "make all" and have a shiny new version that works with the glibc.
The good news is that the worst case scenario for Linux is that you would have to compile libc5 yourself (and the other libraries the application needs) and make sure that the application finds the right libraries. That might be harder than simply installing the application on Windows and having it "just work," but it is a heck of a lot easier to troubleshoot than depending on Microsoft's shifting morass of libraries.
Microsoft pretty much has to provide backwards binary compatibility because of their business model, and so they are stuck with a whole pile of work that (as we have all seen) sometimes still doesn't work. Linux, on the other hand, can take the technical high road and leave it to the systems administrators to sort out any problems. Window's advantage is that I can take almost any wacky Windows program and have some chance of having it run on the version of Windows that I am actually using, but that advantage carries a definite pricetag.
ESR has a long history of releasing internal memos from Microsoft (and others) that are verified to be genuine. Heck, this is his tenth "Halloween" memo with no hoaxes yet.
It might be time for you to rethink who the zealot is. Just because you don't like ESR's politics doesn't mean that he isn't an excellent source of information.
Yes, it would have made more sense to sue someone like you if SCO was actually trying to win the lawsuit. However, you can't trick investors into thinking that you can extract $3 billion from "Skyshadow," but it is at least theoretically possible to extract $3 billion from IBM. With a little help from Microsoft and Sun (to add a bit of legitimacy to the claims), SCO had all the tools it needed to extract millions of dollars from the stock market.
The trick, of course, is to promise investors the moon to drive up the stock price, and then use the high stock price to either cash out or to purchase (at inflated prices) other companies that SCO backers have an interest in (like Vultus).
The reason that SCO is going to sue someone tomorrow is that on the 3rd they are supposed to post their quarterly earnings. My guess is that the financial reports are going to be very very bad and the new law suit is designed to draw attention away from the bottom line.
I can name any number of commercial software products that no longer exist. In fact, just the list of commercial word processors that have gone the way of all the world would fill a small book. Many of these word processor's sole legacy is an obscure Emacs-mode that tries to emulate the keybindings.
At least with Free Software you can maintain the project yourself.
As someone that just walked my in-laws through the purchase of a "low-end" Dell, this is how it works. By default the machine comes with WordPerfect (not PerfectOffice). For $11 you could upgrade to MS Works (which is basically MS Word and some crap). The fact that the low-end machine starts at $499 and you get a $25 discount for spending over $500 means that the MS Word "upgrade" is essentially $14 cheaper than the WordPerfect default.
Not to mention the fact that, as you stated, many people opt for the inexpensive MS Office Small business edition.
Basically WordPerfect and PerfectOffice are included as a reminder to Microsoft that Dell is the one making the sale. Dell has no problems going along with Microsoft, but they don't want MS to forget that, when push comes to shove, Dell has other options.
The problem is that most of these Asian countries that are heavily into piracy have to show the WTO that they are "making an effort" to curb software piracy. The easiest way to do this is to cut down on the prevalence of software piracy by government agencies. If the governments themselves can show that they no longer pirate software then that goes a long way towards keeping the company out of trouble with the WTO.
Microsoft thought that these government agencies would cut down on software piracy by purchasing Microsoft software. However, these countries can't really afford to license their pirated software, and so they are looking at using Free Software instead.
Since the governments of these countries own the bulk of the computer resources this is a very big deal. As these economies grow and computers spread computer users will find that they will need to use Linux and Free Software to be compatible with the government software.
Slashdot Mirror
As a buddy of mine used to say, "all cats are black in the dark."
The good news is that the "winner" in the computer world is hardly ever the guy with the best product. Instead the folks that win generally end up being the folks with the cheapest product that is "good enough." If usability or innovation mattered then the early Macintoshes would have destroyed their DOS based PC competitors. The problem is that most people aren't really willing to pay extra for innovation or usability. They simply want something that will get the job done at the lowest price.
Don't get me wrong. I am not saying that Free Software doesn't innovate or anything like that. In the long run the ability to take existing software and innovate on top of it (instead of starting from scratch) is going to be a huge win for hackers everywhere. Once OpenOffice.org (or whatever) becomes wide spread then all sorts of cool things are going to be possible just because anyone with a crazy idea as to how office suites should work will actually be able to try those ideas out. Most of these ideas will be crap, but the wider range of ideas will still almost certainly be a net win. The fact of the matter is that Microsoft has been lifting ideas from other software developers and implementing then in Windows and Office for years. You would be hard pressed to point to a single major feature that Microsoft actually pioneered.
Microsoft has made a living by being "good enough" and less expensive, and for years the dominated the desktop despite the fact that Windows (and DOS before that) were pathetic knockoffs of other people's innovation. Now Windows is finally getting to be pretty good, but Linux is cheaper, and for an increasingly large group of people it is becoming "good enough."
The BSD TCP/IP stack was meant to be reusable. The folks working on it were part of a U.S. government project to create a standard networking stack that would be useable on a wide variety of OSes.
Microsoft has a lot of "punch" all right. They punch you in the gut, steal your cookies and light your hair on fire.
Microsoft has a long history of double crossing their partners. Sun's just the next sucker on the list.
The problem is that companies are finding that they do have a choice, and the choice is to replace their expensive Sun gear with a pile of inexpensive Linux boxes. And before you say "impossible," you should talk to your Oracle dealer. They will happily tell you that 16 4-way Linux boxes will outperform a 64-way Sun box, and be more reliable (in the aggregate) to boot.
Sun's high profit margins low volume strategy has left them vulnerable to companies that can get away with lower profit margins. Red Hat has basically gone into the "convert from Sun and save a bundle" business.
The difference is that you need a web browser anyhow. No one needs Java on their desktop.
Besides which, it is far easier to create a GUI in XUL + Javascript (and coming soon Python) than it is to create a Java GUI. Basically the idea is to lower the bar for application development so that the folks currently doing HTML + Javascript can create rich client front ends.
If anything this just goes to show how much the average consumer cares about usability. Most consumers don't really care how usable their software is. Usability and $0.50 will get you a Snickers bar. Don't get me wrong. I think that Apple really does have the edge when it comes to making usable systems. Especially if you don't have to share documents and files with Windows users. However, when push comes to shove, consumers want "usable enough" at the lowest price, and that's not Apple.
Not to mention the fact software compiled with gcj or linked with libgcj don't fall under the GPL. You can write proprietary software and compile with gcj and not be "infected" by the GPL. So this part of Gosling's anti-RMS rant is pure FUD.
The real problem is that after all of the work that Sun has put into making Java a platform in real life Java is currently splitting into a million different directions. gcj and GNU Classpath are picking up steam, IBM is pushing platform dependent SWT and Eclipse instead of Swing, etc. With Sun losing the hardware war to Intel and AMD, and the UNIX war to Linux, that leaves Sun with Java as its best hope for a recovery. However, it's a pretty slim hope. Java application servers are basically a comodity as are Java development tools.
Yes, people still run stuff on NT4, but new projects aren't being rolled out on NT4. Even worse, folks that are currently running NT4 almost certainly are working on migration plans.
With programming languages (like operating systems) marketshare is not the critical statistic. No one cares if your application contains 1 hojillion lines of COBOL if all of your new development is in Python. The important factor is how much *new* development is being done in the language.
Yes both Python and Perl have modules that only run on one platform or another. Of course, there are Java libraries that likewise only run on one platform or another (like the JavaCOM libraries). For the most part, however, "porting" Python or Perl software consists of installing Python or Perl on the machine in question and copying over the files.
As an added bonus Python and Perl run on a lot more platforms than Sun's JVM. Python and Perl, for instance, are well supported on all of the BSDs, and Java isn't. So much for Java's touted WORA abilities.
This is precisely why I am not interested in eBook readers. Not only do you end up with a single purpose device, but you probably get stuck with DRM and Windows-only software as well.
Personally, I have switched almost 100% to reading on my Visor Handspring (the B&W Clie my last job provided was better, but the Visor is good enough). I only purchase material that is available in open formats (fictionwise.com and baen.com are good sources). Throw in stuff from Project Gutenberg and I have read nearly 100 ebooks to date (and for less than $100 too).
The benefits of using a PDA are various. I can read in the dark, I always have my reading material with me, and I can read "discreetly." More importantly, 10 years from now I will still have access to all of these books in electronic format. I like reading on my Visor so much that I won't read a new author unless I can get his or her work in an open ebook format first.
I am not holding my breath. There are several good reasons why those who have been singled out have settled. The most important reason that everyone has settled has to do with the fact that so far the RIAA has had a good case against everyone they have sued. Basically the RIAA compiles an extensive set of evidence that the person targetted is responsible for distributing quadzillions of dollars worth of music. Then they threaten to go after the person using laws that were designed to keep mobsters from going into the bootleg music business. When faced with the prospect of going to court and facing hundreds of thousands of dollars in fines (and possibly felony distribution of copyrighted material charges) or paying $3000 the rational choice is obvious. Even rich kids aren't stupid.
So it's not a question of waiting for a rich kid that is willing to fight. It's more about waiting for the RIAA to finger the wrong person. The RIAA cases are high profile enough that someone wrongfully accused could almost certainly get some help paying for legal expenses. The catch is that the RIAA would have to pick on someone who isn't guilty.
Personally, I am glad to see the RIAA using existing laws to curb the sharing of their copyrighted material. I would much rather see them use existing laws than to see them push for new laws requiring universal DRM.
Novell probably will be a little more successful than Red Hat simply because they now employ both the folks at Ximian and the bulk of the KDE hackers (who used to work for SuSE). Red Hat, on the other hand, employed very few KDE hackers (and the one outspoken KDE hacker they did employ quit :).
My guess is that the folks at Ximian and SuSE are likely to see more eye to eye seeing as how their paycheck will depend on them getting along.
This isn't about freedom of speech or freedom of religion. "This is f*cking brilliant" is not protected speech, never has been, and never will be, and you don't have to be religious to find certain phrases vulgar or offensive.
You can rant and rave all you want about this, but there is plenty of legal precedence for what can be considered protected speech. To use an example that /. readers can comprehend SCO can not claim that "All your Linux base belong to us," and then pretend they are merely exercising their right to free speech when it is shown that in fact they don't own all of your Linux bases.
You might disagree with this, and you might even feel that the use of vulgar language should be encouraged over public broadcasts. However, the vast majority of folks disagree with you.
The problem with this sort of tactic is that it is easily abused. If all it takes to get fat cash from the RIAA is to set up a website that sells music from independent artists then you can bet that there will soon be a landrush to create websites that sell music by independent artists.
If sites thut cut out the RIAA middleman become popular there is very little that the recording industry can do besides lower their profit margins and try to compete on price. There's room for markup in the music promotion business, but there's not as much room as there used to be. Eventually this is going to catch up to the RIAA no matter what they do. The price of promoting and distributing music has gone way down, and if the RIAA doesn't adjust their prices to match, then their competitors will start to see success.
Exactly. Novell has plenty of good reasons to give away YaST. First of all they are seeing that RedHat's GPLed installer is already being integrated into other distributions (like Debian). SuSE and Caldera had better distributions than RedHat for years, but RedHat won because every time that someone wanted to borrow code they borrowed it from RedHat. If RedHat's installer becomes ubiquitous then it will be the one folks use, it will also receive the lion's share of debugging efforts. IMHO SuSE/Novell is already late to the party on this one. The money in Linux rollouts lies in support contracts and value-added software (think eDirectory, Groupwise, etc.). Distributions are nothing more than inexpensive marketing. The further YaST spreads the better chances Novell has of getting sysadmins hooked on their tools.
What management really wants to do is automate the docks (like the East Coast) fire half of the workers and break the union. The Union did what it normally does, which is to fight for the status quo.
Code reuse is the "good" sort of laziness. It saves time, memory (shared libs are good), and it allows for easier centralized testing. Basically code reuse is the sort of laziness that saves time and energy for everyone involved.
Not using a packaging manager is the "bad" sort of laziness.
That's just another major design difference between Windows and UNIX. The folks at Microsoft are concerned about binary compatibility, and the UNIX folks (well, at least the Linux and BSD folks) are more concerned about source compatibility. Who cares if your old libc5 application doesn't work under a new version of Linux if you can simply type "make all" and have a shiny new version that works with the glibc.
The good news is that the worst case scenario for Linux is that you would have to compile libc5 yourself (and the other libraries the application needs) and make sure that the application finds the right libraries. That might be harder than simply installing the application on Windows and having it "just work," but it is a heck of a lot easier to troubleshoot than depending on Microsoft's shifting morass of libraries.
Microsoft pretty much has to provide backwards binary compatibility because of their business model, and so they are stuck with a whole pile of work that (as we have all seen) sometimes still doesn't work. Linux, on the other hand, can take the technical high road and leave it to the systems administrators to sort out any problems. Window's advantage is that I can take almost any wacky Windows program and have some chance of having it run on the version of Windows that I am actually using, but that advantage carries a definite pricetag.
ESR has a long history of releasing internal memos from Microsoft (and others) that are verified to be genuine. Heck, this is his tenth "Halloween" memo with no hoaxes yet.
It might be time for you to rethink who the zealot is. Just because you don't like ESR's politics doesn't mean that he isn't an excellent source of information.
Yes, it would have made more sense to sue someone like you if SCO was actually trying to win the lawsuit. However, you can't trick investors into thinking that you can extract $3 billion from "Skyshadow," but it is at least theoretically possible to extract $3 billion from IBM. With a little help from Microsoft and Sun (to add a bit of legitimacy to the claims), SCO had all the tools it needed to extract millions of dollars from the stock market.
The trick, of course, is to promise investors the moon to drive up the stock price, and then use the high stock price to either cash out or to purchase (at inflated prices) other companies that SCO backers have an interest in (like Vultus).
The reason that SCO is going to sue someone tomorrow is that on the 3rd they are supposed to post their quarterly earnings. My guess is that the financial reports are going to be very very bad and the new law suit is designed to draw attention away from the bottom line.
That might explain why ogg files are so much thinner than MP3 files. They get more exercise.
I can name any number of commercial software products that no longer exist. In fact, just the list of commercial word processors that have gone the way of all the world would fill a small book. Many of these word processor's sole legacy is an obscure Emacs-mode that tries to emulate the keybindings.
At least with Free Software you can maintain the project yourself.
As someone that just walked my in-laws through the purchase of a "low-end" Dell, this is how it works. By default the machine comes with WordPerfect (not PerfectOffice). For $11 you could upgrade to MS Works (which is basically MS Word and some crap). The fact that the low-end machine starts at $499 and you get a $25 discount for spending over $500 means that the MS Word "upgrade" is essentially $14 cheaper than the WordPerfect default.
Not to mention the fact that, as you stated, many people opt for the inexpensive MS Office Small business edition.
Basically WordPerfect and PerfectOffice are included as a reminder to Microsoft that Dell is the one making the sale. Dell has no problems going along with Microsoft, but they don't want MS to forget that, when push comes to shove, Dell has other options.
The problem is that most of these Asian countries that are heavily into piracy have to show the WTO that they are "making an effort" to curb software piracy. The easiest way to do this is to cut down on the prevalence of software piracy by government agencies. If the governments themselves can show that they no longer pirate software then that goes a long way towards keeping the company out of trouble with the WTO.
Microsoft thought that these government agencies would cut down on software piracy by purchasing Microsoft software. However, these countries can't really afford to license their pirated software, and so they are looking at using Free Software instead.
Since the governments of these countries own the bulk of the computer resources this is a very big deal. As these economies grow and computers spread computer users will find that they will need to use Linux and Free Software to be compatible with the government software.