If there were 2.5 million PC users willing to switch OS, we'd see 2.5 million more Linux users than we do right now, and Be would still be in business.
Plus, if Joe Schmo's expierience is anything like mine, when he tries out these newer Macs at the store, he's not going to be real impressed with the quality and feel of the Apple hardware [...] "So why should I pay 900 bucks for an Emac that's slow (with it's stock 128 or 256 mb of ram) when I can get this HP for 600, or this Emachines for 400?".
You think people are going to dismiss the Apple hardware out of quality concerns, then buy an HP or eMachines system? I hope you're going to share some of that with everyone else.
Proprietary hardware? Don't be stupid. The G5 takes normal RAM, normal PCI cards, normal hard drives. It has regular USB and Firewire. It even has a standard DVI output for a plain old generic monitor. You can buy a G5 and never have to buy another piece of Apple hardware to upgrade it with, ever again.
Mozilla is still open source, even under that (actually incorrect) definition. The trademarked artwork is not required to build it.
Right, but what if it was? That's the point you seem to be diligently avoiding. If Mozilla is legally allowed to do this now with "optional" data files which still let the software work when they're missing, what's to stop companies doing it with data files that are required for the software to function fully?
The issue is that this is the start of a slippery slope. Let's try a thought experiment:
Company X take Mozilla, and make some big improvements to the interface, including adding support for (say) SVG. They release the source code back to the community, as required by the license--but they keep all the artwork and data files proprietary, including all the image and data files necessary for the XUL user interface and the SVG support. Furthermore, they take out trademarks and patents to prevent cloning of those data files.
Don't laugh off the idea--did you know that many methods of representing continuous color images using CMYK ink dots are patented, and that those patents have been upheld? It's not hard to imagine a method of dithering SVG images to textures on a bitmap screen being similarly patentable.
Now in our thought experiment, in spite of the software being "open source", it has effectively been stolen and made closed. Company X can sell their proprietary version of Mozilla, and nobody can use the improvements in the free version. They could even use Palladium-like digital signature technology to make sure that the proprietary binary distribution couldn't be made to work with open data and graphics, even though the source code is available.
As I see it, for a piece of software to be open source, you need to be allowed to redistribute and use all of the sources needed to build the entire thing. If I can't build Mozilla exactly as it is in the official binary distribution and give it to a friend, it's not open source.
That's why when I released one of my screensavers under the GPL, I released the artwork files under the GPL as well.
...in the open source community, we'd see more effort going into GNUstep.
Works with everything we have today? Check, there's compatibility with KDE and GNOME applications as well as Motif, with window style hints too.
High level language support? Check, Objective-C provides Smalltalk-like object orientation, and automatic memory management is available. There are also bindings to Ruby and Java. You can even build Java applications with native quality look and feel.
Compatible with what programmers know today? Yup, Objective-C is a slight superset of C, so almost any programmer can get to grips with it in a weekend. (Speaking as someone who did.)
Good class libraries? Yes, modeled on NeXT's excellent work, the same foundation used to build OS X. I've written Cocoa code, it's the most painless class library I've encountered. (Yes, I write Java too and have written C++.)
Cross platform? Yes again, programs are portable between GNUstep and Cocoa without too much work--see GNUmail for an example. Non-GUI programs even port to Windows without major effort, allegedly.
Good developer tools? Again, yes. Excellent developer tools on OS X. Doubtless the free tools on Linux could use some work, but that shouldn't be too hard. We can even build them using the OS X tools if necessary.
Pretty UI? Well, I think it looks OK. Not as nice as Aqua, but it's functional.
Mature? Well, the Objective-C compiler is GCC, Apple use it for their developer tools and push back improvements, the class library design has been refined over the course of 10+ years.
Think about it, people. We could unify the Linux and Apple developer communities. All work towards one common goal. Get 10%+ desktop market share for OpenStep/OS X/GNUstep in no time.
Hell, get GNUstep up to scratch and you'd probably see developers porting their commercial applications from OS X to Linux. Wouldn't you like to see products from Adobe, Macromedia, maybe even Apple available to run on your Linux desktop?
Think about all the problems that have been solved by NeXT and Apple. Application packaging? Solved, applications are bundles of files that you can just drag-drop wherever you want to keep them, and they work.
But no, the GNOME crowd will shun a tried and tested, open source, native code solution that works today and is backed by the single biggest vendor of UNIX desktops, in favor of cloning an untested patented bytecode-based Microsoft solution. Gates and co must be laughing.
But hey, go GNOME. Let's reinvent those wheels. Maybe your next set won't be square.
And of course, Gentoo solves all those problems, while keeping the undeniable advantages of compiling from source:
1. Fine-grained automatic dependency resolution. I don't run X on my MP3 server, so when I install vi and CD rippers and so on, they're installed without the optional GUI components. Yet this doesn't require multiple packages for each different combination of features, and it doesn't require me to pick the right package from a list.
2. Optimization for your specific hardware, even if (like me) you have weird hardware (in my case a VIA C3).
3. Faster delivery of updates. If there's a minor point release I need (e.g. to fix a security hole) and nobody's made an ebuild for it yet, chances are I can copy the existing ebuild text file, change the one line containing the version number, and have a working ebuild for the new version. No special knowledge is required. I don't have to wait around for someone to make a binary package. Even better, making my own ebuild doesn't break future auto-upgrades--when version X+2 arrives officially, the one I put together myself will be replaced by it.
4. I always know I'm getting a definitive copy of the source as written by the developer(s), not something patched or built by the distro maintainers in some way I can't easily reproduce. I know exactly what versions of which tools it was build with. (If Gentoo needs to patch the official sources, the patches are supplied with, and applied by, the ebuild--in standard patch format.)
5. Putting the previous point slightly differently, I always know I can reproduce exactly the installed software, with any modification I care to make. Which means I can fix things myself. OK, this isn't a big benefit for end users, but for sysadmins it's a huge plus.
Of course, compilation takes a while. That's why we have a multi-tasking OS. And as people always point out in the Java vs C threads, computers are getting faster all the time.
The other minus is download times; source is often (but not always) significantly bigger than binaries.
I wanted something like GURPS for years for assorted modern-day and SF campaigns. Unfortunately, Steve Jackson games refused to issue the metric version of GURPS in English, and I flat out refuse to run a science fiction game in old-fashioned units. Feet and inches and hogsheads are OK in pseudo-mediaeval games like D&D, for flavor, but I don't want to be working out storage capacities of starships in cubic feet or laser energy in ergs.
Then WotC released d20 under an open license, and now I can get Call of Cthulhu and d20 modern and T20 (Traveller d20) and mix and match as I like, and do all the SF stuff in metric. SJG had their chance, but I think d20 is going to wipe out GURPS.
That said, I may yet pick up a copy of GURPS 4th Edition if they drag it into the 20th Century. (Well, the 19th really, but let's not be too pedantic...)
Well, Linus wanted a UNIX clone with pre-emptive multi-tasking and virtual memory and hackable source code; that's why he was running MINIX, and that's why he started writing Linux rather than sticking with MINIX. Andy Tanenbaum wouldn't add VM and pre-emptive multi-tasking to MINIX because it would have meant it couldn't run on 8086 PCs and Atari STs.
If Linus had been running MINIX on an Amiga instead of a PC, he would still not have had pre-emptive multi-tasking and virtual memory, and would probably still have written Linux. He'd just have targeted it for 68k instead of x86...
I can almost guarantee that this is the usual marketing bullshit from BEA Systems (British Aerospace contactor that inexplicably has an in with U.S. Homeland Security)
Nothing inexplicable about that; British Aerospace is the UK's biggest defense contractor, and one of the world's foremost manufacturers of torture equipment.
The impression I get is that most Windows users warez all their software from work; my cousin specifically told me that was the reason he used Windows rather than a Mac.
As I see it, DRM on Windows PCs will be a great boost for Linux and the Mac, because if all those Windows users actually had to pay for their software, they'd switch.
It's a long time since Windows systems regularly shipped with install CDs. If you're lucky you get a set of crappy "rescue" CDs which wipe the entire hard drive and replace it with the image the machine shipped with.
I'm not sure why the source to Apple's Open Firmware isn't available, but I imagine it's because they licensed it from one of the commercial Open Firmware vendors.
Put together a good web site. Sell MP3 files of your music, encoded with a good encoder (e.g. LAME) from the original master tapes. Make sure you catalog everything in the ID3 tags. Make sure it's really easy to buy an entire album and download all the MP3s together in a zip file, or just buy individual tracks. Charge somewhere from 50 cents to a dollar per track, or comfortably under $10 for an album. Put a good search engine on the front, and let people listen to a preview of the tracks before buying.
That's it. That's all you need to do. People like me will rush to purchase music from you, even though we could be pirating it. Why? Convenience. You are providing a convenient high quality product, and that's worth some money to me.
Not convinced? Look at Starbucks. I can drink free coffee in the office any time I want, but I walk across the street (in the cold!) to buy coffee at Starbucks because the quality is a bit better, the ambiance is more pleasant, and the selection is better.
Look at suicidegirls.com. OK, you can stop looking. No, really, come back... The site owner was initially angry when he found copies of images from the site on P2P systems. Then after a few weeks, he noticed that a big chunk of people signing up for the site had heard about it because they'd downloaded a pirate image or ten, liked them, and decided it was worth paying money for convenience and quality.
I'll throw in my obligatory plug for bleep.com, who have worked out the whole "online music sales" thing, as described above.
The point is, you don't need elaborate DRM, monitoring systems, or taxation, if you're prepared to settle for making a reasonable profit in spite of piracy. Just view piracy as a business overhead, a fact of life, something that cannot be eliminated--and then spend your time working out how to entice customers into giving you money anyway. Use the carrot, not the stick.
Most of the linear algebra libraries rely in some way on the basic linear algebra subprograms, which are of course FORTRAN subroutines.
They may have been originally, but on modern CPUs they aren't. For example, on OS X on a G4 or G5, the BLAS routines are implemented as Altivec code which runs directly on the CPU's vector processor. It's many times faster than the FORTRAN implementations, and you can call it from C++ or Objective-C; I haven't tried Java. So, it's very easy to write OO code which will top FORTRAN on a G5.
On a G3, I believe the fallback implementation of BLAS is in assembler, but I confess that I haven't checked. If you're doing any kind of linear algebra on a G3 you're using the wrong hardware anyway.
Many directors get their needle drops so engrained in their heads that they often chuck the whole score some poor sod has written in favor of the temp music.
Slashdot Mirror
If there were 2.5 million PC users willing to switch OS, we'd see 2.5 million more Linux users than we do right now, and Be would still be in business.
You think people are going to dismiss the Apple hardware out of quality concerns, then buy an HP or eMachines system? I hope you're going to share some of that with everyone else.
Proprietary hardware? Don't be stupid. The G5 takes normal RAM, normal PCI cards, normal hard drives. It has regular USB and Firewire. It even has a standard DVI output for a plain old generic monitor. You can buy a G5 and never have to buy another piece of Apple hardware to upgrade it with, ever again.
Right, but what if it was? That's the point you seem to be diligently avoiding. If Mozilla is legally allowed to do this now with "optional" data files which still let the software work when they're missing, what's to stop companies doing it with data files that are required for the software to function fully?
...the fabulous portrayal of gay people in Grand Theft Auto.
The issue is that this is the start of a slippery slope. Let's try a thought experiment:
Company X take Mozilla, and make some big improvements to the interface, including adding support for (say) SVG. They release the source code back to the community, as required by the license--but they keep all the artwork and data files proprietary, including all the image and data files necessary for the XUL user interface and the SVG support. Furthermore, they take out trademarks and patents to prevent cloning of those data files.
Don't laugh off the idea--did you know that many methods of representing continuous color images using CMYK ink dots are patented, and that those patents have been upheld? It's not hard to imagine a method of dithering SVG images to textures on a bitmap screen being similarly patentable.
Now in our thought experiment, in spite of the software being "open source", it has effectively been stolen and made closed. Company X can sell their proprietary version of Mozilla, and nobody can use the improvements in the free version. They could even use Palladium-like digital signature technology to make sure that the proprietary binary distribution couldn't be made to work with open data and graphics, even though the source code is available.
As I see it, for a piece of software to be open source, you need to be allowed to redistribute and use all of the sources needed to build the entire thing. If I can't build Mozilla exactly as it is in the official binary distribution and give it to a friend, it's not open source.
That's why when I released one of my screensavers under the GPL, I released the artwork files under the GPL as well.
...in the open source community, we'd see more effort going into GNUstep.
Works with everything we have today? Check, there's compatibility with KDE and GNOME applications as well as Motif, with window style hints too.
High level language support? Check, Objective-C provides Smalltalk-like object orientation, and automatic memory management is available. There are also bindings to Ruby and Java. You can even build Java applications with native quality look and feel.
Compatible with what programmers know today? Yup, Objective-C is a slight superset of C, so almost any programmer can get to grips with it in a weekend. (Speaking as someone who did.)
Good class libraries? Yes, modeled on NeXT's excellent work, the same foundation used to build OS X. I've written Cocoa code, it's the most painless class library I've encountered. (Yes, I write Java too and have written C++.)
Cross platform? Yes again, programs are portable between GNUstep and Cocoa without too much work--see GNUmail for an example. Non-GUI programs even port to Windows without major effort, allegedly.
Good developer tools? Again, yes. Excellent developer tools on OS X. Doubtless the free tools on Linux could use some work, but that shouldn't be too hard. We can even build them using the OS X tools if necessary.
Pretty UI? Well, I think it looks OK. Not as nice as Aqua, but it's functional.
Mature? Well, the Objective-C compiler is GCC, Apple use it for their developer tools and push back improvements, the class library design has been refined over the course of 10+ years.
Think about it, people. We could unify the Linux and Apple developer communities. All work towards one common goal. Get 10%+ desktop market share for OpenStep/OS X/GNUstep in no time.
Hell, get GNUstep up to scratch and you'd probably see developers porting their commercial applications from OS X to Linux. Wouldn't you like to see products from Adobe, Macromedia, maybe even Apple available to run on your Linux desktop?
Think about all the problems that have been solved by NeXT and Apple. Application packaging? Solved, applications are bundles of files that you can just drag-drop wherever you want to keep them, and they work.
But no, the GNOME crowd will shun a tried and tested, open source, native code solution that works today and is backed by the single biggest vendor of UNIX desktops, in favor of cloning an untested patented bytecode-based Microsoft solution. Gates and co must be laughing.
But hey, go GNOME. Let's reinvent those wheels. Maybe your next set won't be square.
And of course, Gentoo solves all those problems, while keeping the undeniable advantages of compiling from source:
1. Fine-grained automatic dependency resolution. I don't run X on my MP3 server, so when I install vi and CD rippers and so on, they're installed without the optional GUI components. Yet this doesn't require multiple packages for each different combination of features, and it doesn't require me to pick the right package from a list.
2. Optimization for your specific hardware, even if (like me) you have weird hardware (in my case a VIA C3).
3. Faster delivery of updates. If there's a minor point release I need (e.g. to fix a security hole) and nobody's made an ebuild for it yet, chances are I can copy the existing ebuild text file, change the one line containing the version number, and have a working ebuild for the new version. No special knowledge is required. I don't have to wait around for someone to make a binary package. Even better, making my own ebuild doesn't break future auto-upgrades--when version X+2 arrives officially, the one I put together myself will be replaced by it.
4. I always know I'm getting a definitive copy of the source as written by the developer(s), not something patched or built by the distro maintainers in some way I can't easily reproduce. I know exactly what versions of which tools it was build with. (If Gentoo needs to patch the official sources, the patches are supplied with, and applied by, the ebuild--in standard patch format.)
5. Putting the previous point slightly differently, I always know I can reproduce exactly the installed software, with any modification I care to make. Which means I can fix things myself. OK, this isn't a big benefit for end users, but for sysadmins it's a huge plus.
Of course, compilation takes a while. That's why we have a multi-tasking OS. And as people always point out in the Java vs C threads, computers are getting faster all the time.
The other minus is download times; source is often (but not always) significantly bigger than binaries.
I wanted something like GURPS for years for assorted modern-day and SF campaigns. Unfortunately, Steve Jackson games refused to issue the metric version of GURPS in English, and I flat out refuse to run a science fiction game in old-fashioned units. Feet and inches and hogsheads are OK in pseudo-mediaeval games like D&D, for flavor, but I don't want to be working out storage capacities of starships in cubic feet or laser energy in ergs.
Then WotC released d20 under an open license, and now I can get Call of Cthulhu and d20 modern and T20 (Traveller d20) and mix and match as I like, and do all the SF stuff in metric. SJG had their chance, but I think d20 is going to wipe out GURPS.
That said, I may yet pick up a copy of GURPS 4th Edition if they drag it into the 20th Century. (Well, the 19th really, but let's not be too pedantic...)
I know. The key phrase was "UNIX-like".
Well, Linus wanted a UNIX clone with pre-emptive multi-tasking and virtual memory and hackable source code; that's why he was running MINIX, and that's why he started writing Linux rather than sticking with MINIX. Andy Tanenbaum wouldn't add VM and pre-emptive multi-tasking to MINIX because it would have meant it couldn't run on 8086 PCs and Atari STs.
If Linus had been running MINIX on an Amiga instead of a PC, he would still not have had pre-emptive multi-tasking and virtual memory, and would probably still have written Linux. He'd just have targeted it for 68k instead of x86...
Nothing inexplicable about that; British Aerospace is the UK's biggest defense contractor, and one of the world's foremost manufacturers of torture equipment.
The impression I get is that most Windows users warez all their software from work; my cousin specifically told me that was the reason he used Windows rather than a Mac.
As I see it, DRM on Windows PCs will be a great boost for Linux and the Mac, because if all those Windows users actually had to pay for their software, they'd switch.
It's a long time since Windows systems regularly shipped with install CDs. If you're lucky you get a set of crappy "rescue" CDs which wipe the entire hard drive and replace it with the image the machine shipped with.
OpenBIOS is a project to develop an open source implementation of the Open Firmware specification.
If you want to look at the OS X source code, you can get it from http://developer.apple.com/darwin/.
I'm not sure why the source to Apple's Open Firmware isn't available, but I imagine it's because they licensed it from one of the commercial Open Firmware vendors.
Oh boy, I thought that was one of those "*Every* Kubrick fan knows *that*" facts.
You can even buy the Alex North soundtrack that was going to be used for "2001" on CD now, so it's not exactly obscure.
Anyway, here is a link and here is another one.
I'm not allowed to vote. I just live here and pay taxes. So much for the American Revolution.
Put together a good web site. Sell MP3 files of your music, encoded with a good encoder (e.g. LAME) from the original master tapes. Make sure you catalog everything in the ID3 tags. Make sure it's really easy to buy an entire album and download all the MP3s together in a zip file, or just buy individual tracks. Charge somewhere from 50 cents to a dollar per track, or comfortably under $10 for an album. Put a good search engine on the front, and let people listen to a preview of the tracks before buying.
That's it. That's all you need to do. People like me will rush to purchase music from you, even though we could be pirating it. Why? Convenience. You are providing a convenient high quality product, and that's worth some money to me.
Not convinced? Look at Starbucks. I can drink free coffee in the office any time I want, but I walk across the street (in the cold!) to buy coffee at Starbucks because the quality is a bit better, the ambiance is more pleasant, and the selection is better.
Look at suicidegirls.com. OK, you can stop looking. No, really, come back... The site owner was initially angry when he found copies of images from the site on P2P systems. Then after a few weeks, he noticed that a big chunk of people signing up for the site had heard about it because they'd downloaded a pirate image or ten, liked them, and decided it was worth paying money for convenience and quality.
I'll throw in my obligatory plug for bleep.com, who have worked out the whole "online music sales" thing, as described above.
The point is, you don't need elaborate DRM, monitoring systems, or taxation, if you're prepared to settle for making a reasonable profit in spite of piracy. Just view piracy as a business overhead, a fact of life, something that cannot be eliminated--and then spend your time working out how to entice customers into giving you money anyway. Use the carrot, not the stick.
They may have been originally, but on modern CPUs they aren't. For example, on OS X on a G4 or G5, the BLAS routines are implemented as Altivec code which runs directly on the CPU's vector processor. It's many times faster than the FORTRAN implementations, and you can call it from C++ or Objective-C; I haven't tried Java. So, it's very easy to write OO code which will top FORTRAN on a G5.
On a G3, I believe the fallback implementation of BLAS is in assembler, but I confess that I haven't checked. If you're doing any kind of linear algebra on a G3 you're using the wrong hardware anyway.
I'm still waiting for someone to sell a 10" ultraportable with WiFi, no Windows license, and hardware that has open source drivers.
The definitive example, of course, being 2001.
Well, GNOME self-destructed on my Fedora system, and when I went back to KDE to get some work done I discovered that KDE was much, much faster.
Yeah, now all it needs is some documentation.
Dude, your page is missing a link to bleep.com.
At $15 a CD plus shipping they're no less overpriced than the RIAA.