You can easly build a dual Athlon system that could trounce an equivilant g5 for less cost. Xeon's arnt the only MP capable x86 chips out there...
Um, put your money where your mouth is. Find a dual-Athlon for sale with at least 512 MB of RAM, a 160 GB serial ATA hard drive, a good video card with 128 MB of RAM, and at least a DVD-R drive. There's no way you could come up with one for under $3000.
Single CPU, yes. But dual-CPU motherboards - even AMD - cost a lot more.
You'll want to be careful with the code the intel compiler spits out. It may be fast but some scientific codes compiled with it give incorrect answers. Compile the same code on any other f90 compiler and it gives the correct answer. Not something to inspire confience in your answers.
I'd have to agree. I tried using Intel's compilers (both C and Fortran) for Linux recently, and I found them to be a joke. Sure, under some circumstances they generate some faster code, but the compiler also crashes or gives weird error messages, doesn't fully implement the C++ spec, has trouble "playing nice" with gcc, and yes, sometimes produces incorrect code.
Compiled languages are damned convenient to the user. "Here's an executable, just run it", versus, "here's a script, go download compile and install the interpreter first, making sure it's the correct version, set up your environment variables correctly, then run the script."
Amen. When I'm trying to solve a problem, I like using high-level languages like Perl or Python. But when I need to write an application that I want ordinary users to be able to download and _just use_, I don't have a choice - I have to write it in C or C++.
Luckily, thanks to faster processors (shorter compile times) and tools like valgrind to detect memory errors, programming in C and C++ isn't nearly as bad as it used to be.
Paul Davis wrote 99% of the Ardour code, he even recently claimed this on the Ardour website. I'm sorry, I've worked on many teams, this does not constitute a team. Ardour is a wonderful piece of philanthropy, but as an example of the open source development model is it a failure.
Audacity 1.0 was released about one year ago. I probably wrote 90% of the code. Since then, dozens of developers have joined the team and contributed huge amounts of code - I'm responsible for maybe only 50% of the code now. That doesn't even count the contributions of the translators, documentation writers, packagers, etc. that have helped make Audacity so successful.
The point is, a stable 1.0 release is a necessary first step before many new developers will join a project and help move it forward. Look at Mozilla for another example - since 1.0 was released, the developer base has grown dramatically.
Also, keep in mind that while Paul Davis did write (almost) all of the Ardour code, and even Jack, Ardour depends on a number of other open-source libraries (libsndfile, libsamplerate, soundtouch, and GTK come to mind), so in that sense it is a success of the open-source model. Paul Davis was able to produce a monumental program in 4 years almost entirely by himself because he was able to build on top of other open-source software.
I downloaded and played with the latest beta [of Audacity]. A few questions:
- How do I access the envelope editor?
Click on the envelope tool - if you have 1.1.3, it looks like an arrow, if you have the latest code from CVS, it looks more like two triangles. Either way, the display changes, and you can click and drag to edit envelopes directly.
Or, click on the multi tool (it looks like an asterisk) which lets you switch between selecting and envelope editing with ease. Try the latest CVS for this...
- Can I use MIDI events (a MIDI mixer, or the knobs on my MIDI synth) to control envelopes? - Is there any way to have VST effects be applied during playback, as opposed to being a processing phase? (For interactive editing) - Can I use envelopes to control the parameters of VST effects? - Do you support mixing for more than two channels?
Sorry, not yet for any of those features. Those are all planned for our next major release.
It's looking better than the last release I tried, but I still can't shake the feeling that everything I do is "offline." If I could find a way to have more interactive access to the controls, I think I'd find it to be a more engaging experience.
One reason we went with this model is because it's incredibly frustrating for people with slower computers to have stuttering audio because their processor can't keep up with the effects they want to apply. Yes, most programs have an option to render all of the effects, but this usually takes an extra step and is non-intuitive. With Audacity, we're starting the other way - effects are always offline for now, so they work on everyone's computer - and we're slowly adding realtime effects.
In the short term, we've added effect previews and a looped-play mode. Actual realtime effects on tracks won't come until after 1.2.0 is released, though.
If you'll read Ardour's documentation, you will see that because Steinberg will not release the headers for the VST files, they are not available to use with Ardour. Steinberg has not come out as hostile to Linux, but they aren't giving OSS developers any of thier cards, either.
I don't think the situation is quite as bad as that. Steinberg releases the full VST SDK for free and gives you pretty reasonable rights to do just about anything with it in your product - however you can't redistribute the VST SDK source code. In Audacity, the source code is VST-ready, but not compiled in by default. All you need to add VST support is to download the VST SDK, put it in the right directory, change one #define, and recompile. All of our binary releases do this already.
If 31 band graphic EQ could be added to Audacity, it would have everything I need and I could switch to Linux full-time.
Note that Audacity does have a FFT filter - not the same thing, but might be able to tide you over.
Programmers: does anyone have the relevant DSP code for a graphic EQ (any number of bands) that they'd be willing to donate to Audacity? I'd be happy to take care of integrating it into our codebase...
How does Ardour compare to Audacity, another free digital audio editing program? I want to know my options before switching from proprietary Cool Edit.
I agree with many of the responses, that Ardour and Audacity are as different as night and day. But I don't think some of the comparisons were quite fair, though, comparing Ardour to Quark Express and Audacity to NotePad! Maybe I'm just biased, though, as the lead developer of Audacity.
First of all, I don't think that Ardour and Audacity are directly competing. There's some feature overlap, but the user interfaces are so different (on purpose) that we're both capturing different markets, and addressing different needs.
Anyway, here are some of the major differences I can think of off the top of my head:
1. Ardour only runs on Linux. Audacity runs (completely natively) on Linux, Mac OS X, and Windows. Ardour requires Jack. Audacity only requires OSS on Linux (supported everywhere) with support for ALSA and Jack in alpha testing now (email the devel list if you want info on how to enable one of these).
2. Ardour supports true multi-track recording and playback. Audacity only records more than 2 tracks on some systems, and always mixes down to 2 output channels (stereo).
3. Both Audacity and Ardour are quite powerful, but in different ways. Ardour supports MIDI control, powerful muting/soloing, and realtime effects. Audacity has only non-realtime effects, but some of these are quite powerful in a different way, like our Noise Removal. Audacity supports plug-in effects in Ladspa (Linux), VST (Win/Mac), and Nyquist (a high-level interpreted language for audio processing).
Lots of features in common, too: both Audacity and Ardour support floating-point samples, high sample rates, resampling, LADSPA plug-ins, unlimited undo, internationalization, etc.
One last thing to dispel the myth that Audacity is "simple" compared to Ardour: I just did a quick wc, and Audacity is 70k+, Ardour is 100k+ lines of code (someone else can feel free to do a sloccount if you want). In both cases, not counting other libraries that aren't directly part of the project. No matter how you look at it, they're within a factor of 2 in terms of size and complexity, just in different areas.
We're nearing a feature freeze for Audacity 1.2.0. If you're comfortable compiling software, please check out our latest code from CVS and help us get the last few bugs out. I'd strongly encourage you to try out both Ardour and Audacity - hopefully you'll find that you use them to complement each other.
The professor's article consistently fails to give credit where credit is due. "Linux" is not an operating system and it never was. Linux is a part of an operating system called a kernel (which acts as a bookkeeper managing the resources of a computer so applications can share those resources without conflict). It's fair to credit the major chunks of an operating system; GNU and Linux are both valuable chunks. It's also less confusing to the reader if one refers to the union of the GNU operating system with the Linux kernel by mentioning both parts (hence the term "GNU/Linux"). For the FSF's take on this, please read their essay which also has a link to a FAQ on this issue.
So far all of SCO's claimed violations are related to code in the Linux kernel. Therefore GNU has nothing to do with this lawsuit.
You might be running a GNU/Linux operating system. More power to you. I'm running a GNU/KDE/Mozilla/Sun/Python/PHP/Apache/Linux operating system, thank you very much. GNU only contributed a tiny portion of what I use. I'm very grateful to GNU for their software contributions, but to say that GNU and Linux are the only important components of the O.S. is the height of silliness.
Text editors, command shells, compiler, linker, debugger, C library, standard Unix tools (grep/awk/diff/etc.), gtk, desktop environment (Gnome)... short of X and the kernel, pretty much everything in a modern "Linux" distribution that I at least consider to be part of the OS comes from GNU. Check the man pages for 'printf', 'tar', and such.
There's no question that GNU represents a huge chunk of most Linux-based systems, but not as large as RMS would have you believe. Here is a list of most of the software I've used at work this week:
Emacs (GNU) for my text editor
KDE (not GNU) for my desktop environment, Window Manager, terminals, panel, and screensaver
Python and Numpy (not GNU) for programming
Mozilla (not GNU) for web browsing
gcc and make (GNU) for writing fast numerical analysis code
Matlab (not free, alas...) for some scientific programming
wxWindows (mostly not GNU, but uses GTK, which is GNU) for GUI development
OpenSSH (not GNU) for remote command execution
MPICH (not GNU) for parallelization
Linux kernel, drivers for all of my hardware (not GNU)
X11 (not GNU)
I don't have anything against GNU. Obviously their contributions were really important. But of the tools I use on a day to day basis, I'd say that less than 10% of the code is GNU code.
Obviously usage patterns differ from person to person. Since RMS doesn't use GUIs unless he has to (really, not even GNOME!) his experience using a "GNU/Linux" system is quite different than mine.
All, just remember that the definition of spam is fluid. One person's spam is another's direct marketing.
No. The commonly-accepted definion of spam is (1) unsolicited (2) email that is (3) either commercial or bulk in nature. (1), (2), and (3) must all be present for something to be spam.
Why aren't forged headers part of that definition? My impression is that the vast majority of spam has forged headers, and that if we could do away with forged headers, the spam problem would go away because thousands of users would just DOS anyone who attempted to spam them.
Given that BusSpeed * ClockMultiplier = Processor Speed...Means that all three systems have the same multiplier on the chip. Which strongly implies to me that they're all the exact same chip. We'll have to wait and see how easy they are to overclock, but if you could just change the 800Mhz bus system to 1Ghz bus, you'd save yourself $1000 in the process.
You think there's any physical difference between a 1 GHz G4 and a 1.4 GHz G4 either? Or a 2.4 GHz P4 and a 2.8 GHz P4? The processors are the same - they test them and see how fast they can clock each one. The higher they clock them, the fewer chips that can handle that, so they charge more for the ones that can clock faster.
When you overclock, you're taking your chances - and you're definitely shortening the lifespan of your chip.
Besides, the $3000 Power Mac G5 has TWO processors, so you couldn't actually save $1000 by overclocking.
I honestly see this as a good thing. It's evolution. It's moving forward. And, ideally, it could benefit everyone involved. Down with the album (unless you're making a real album, and not a simple compilation of singles - read: most 'albums' released today).
Imagine this scenario. Instead of releasing a new 'album' every year, or every couple of years, or whatnot, artists would instead have the option of releasing each song as they record it.
I totally agree. That certainly doesn't mean that artists wouldn't sometimes come up with an entire album that's meant to be listened to as a whole. But I'd love to see "mini-albums" of 4-5 songs, or larger collections of 18 songs (more than 1 CD worth, but less than 2) - and with download services these could all be priced appropriately.
The fact is that until now there hasn't ever been a mode of distribution that favored any amount of music between a single (or single + bonus track) and an entire album (usually at least an hour, or it doesn't seem worth the money).
They have some of the most loyal fans out there. If Hail to the Thief had been on the iTunes Music Store, I would have bought it there. It's not, so I ended up going to best buy to pick it up.
Radiohead may put out some great albums, but I don't have any of their CDs and I don't really know what they sound like. I'd be much more likely to spend a dollar or two to try out a few songs by a band I don't know, to see it I like those, than I would be to spend $10 to purchase the whole album.
So in my opinion, they're shooting themselves in the foot. Their fans will continue to purchase the albums, but they're losing out on the opportunity to sell a few songs to people like me who haven't discovered them yet.
No troll or flame here, but isn't it about time that Apple offered a Mac for less than $1000 that isn't completely ancient (ie, no 600 MHz iMac please). Why do they still completely forsake the low end?
I don't see how the $999 eMac counts as "ancient", with a 1 GHz G4, monitor included.
Besides, you can get a pretty decent used Mac on eBay for less than $1000. Check out these completed auctions as proof:
The scheme instead computes an MD5 sum for each line (actually each five lines together) and publishes the hashed versions of the files. Then anybody can do the line by line compare without ever seeing a readable version of the source code.
Except that then it's easy to figure out exactly which lines in Linux are the same (since we have access to the Linux source code) and that's exactly what SCO doesn't want us to know!
The issue here isn't that there isn't a way for SCO to tell us what was copied without giving away their IP, it's that they're refusing to play fair at all.
You may be right, but it could be that they'd be perfectly happy with a request like "please send me all GPL software that's used in my Linksys Wireless 4-port Cable/DSL Router". That ought to be perfectly specific enough and that may have been all they wanted. That's very different than "send me all GPL software that you've ever shipped a product with", which is what they might have been trying to avoid.
Bluetooth by itself is not a big deal. But combined with technology like Apple's Rendezvous, it's really cool. Apple seems to be imagining a future where you can walk around with your laptop, and services like printers become available immediately when they come within range, then disappear when you go out of range. To Apple, Bluetooth seems to be all about service discovery.
I think we're just waiting for the killer app. Syncing your cell phone may be fun, and cordless keyboards and mice may be cool, but neither are as big a deal as wireless Internet access on your front porch.
I haven't given up hope, though. Since Bluetooth is cheap and low-power, it's not going anywhere soon.
Slot machines don't have cycles. Whether they pay out or not is entirely random (or at least as random as it is possible to make them). Your chance of winning does not increase even slightly if other people have poured piles of money into the machine already.
SOME slot machines don't have cycles, maybe MOST, but there are plenty that do.
Apparently the 'proof' that sliot machines, fruit machines as those wacky brits choose to call them, is that, if you 'freeze' the state of a fruit machine at some point and then repeat the next step, the machine will generate the same outcome.
No, that's only half of it.
The machine gives you a choice (typically "high" or "low" in the examples they gave), but you will always lose, NO MATTER WHICH ONE YOU CHOOSE.
A deterministic slot machine is one thing: even if it simply paid out exactly 1 in every 10,000 spins, that would be legal.
What's illegal (according to the authors) is that the machine presents a game and says that it's a gamble: if you choose the correct alternative (high or low) you will double your winnings, otherwise you will lose. But the machine has already decided that you will definitely lose no matter what you pick, and that's what they think is illegal under U.K. law.
In other words, playing the game at all is a gamble. If you play, you might win (no matter what you press), or you might lose (no matter what you press). However, the game is presenting choices in the middle of the gameplay as "gambles" when in reality they don't affect the outcome.
Slashdot Mirror
[The most bizarre thing is] that Los Angeles has an apple store not 30 minutes away in decent traffic.
So, given typical L.A. traffic, you'd be lucky to get there in six hours. I'd rather camp out in Santa Monica by the beach.
You can easly build a dual Athlon system that could trounce an equivilant g5 for less cost. Xeon's arnt the only MP capable x86 chips out there...
Um, put your money where your mouth is. Find a dual-Athlon for sale with at least 512 MB of RAM, a 160 GB serial ATA hard drive, a good video card with 128 MB of RAM, and at least a DVD-R drive. There's no way you could come up with one for under $3000.
Single CPU, yes. But dual-CPU motherboards - even AMD - cost a lot more.
You'll want to be careful with the code the intel compiler spits out. It may be fast but some scientific codes compiled with it give incorrect answers. Compile the same code on any other f90 compiler and it gives the correct answer. Not something to inspire confience in your answers.
I'd have to agree. I tried using Intel's compilers (both C and Fortran) for Linux recently, and I found them to be a joke. Sure, under some circumstances they generate some faster code, but the compiler also crashes or gives weird error messages, doesn't fully implement the C++ spec, has trouble "playing nice" with gcc, and yes, sometimes produces incorrect code.
Compiled languages are damned convenient to the user. "Here's an executable, just run it", versus, "here's a script, go download compile and install the interpreter first, making sure it's the correct version, set up your environment variables correctly, then run the script."
Amen. When I'm trying to solve a problem, I like using high-level languages like Perl or Python. But when I need to write an application that I want ordinary users to be able to download and _just use_, I don't have a choice - I have to write it in C or C++.
Luckily, thanks to faster processors (shorter compile times) and tools like valgrind to detect memory errors, programming in C and C++ isn't nearly as bad as it used to be.
Paul Davis wrote 99% of the Ardour code, he even recently claimed this on the Ardour website. I'm sorry, I've worked on many teams, this does not constitute a team. Ardour is a wonderful piece of philanthropy, but as an example of the open source development model is it a failure.
Audacity 1.0 was released about one year ago. I probably wrote 90% of the code. Since then, dozens of developers have joined the team and contributed huge amounts of code - I'm responsible for maybe only 50% of the code now. That doesn't even count the contributions of the translators, documentation writers, packagers, etc. that have helped make Audacity so successful.
The point is, a stable 1.0 release is a necessary first step before many new developers will join a project and help move it forward. Look at Mozilla for another example - since 1.0 was released, the developer base has grown dramatically.
Also, keep in mind that while Paul Davis did write (almost) all of the Ardour code, and even Jack, Ardour depends on a number of other open-source libraries (libsndfile, libsamplerate, soundtouch, and GTK come to mind), so in that sense it is a success of the open-source model. Paul Davis was able to produce a monumental program in 4 years almost entirely by himself because he was able to build on top of other open-source software.
I downloaded and played with the latest beta [of Audacity]. A few questions:
- How do I access the envelope editor?
Click on the envelope tool - if you have 1.1.3, it looks like an arrow, if you have the latest code from CVS, it looks more like two triangles. Either way, the display changes, and you can click and drag to edit envelopes directly.
Or, click on the multi tool (it looks like an asterisk) which lets you switch between selecting and envelope editing with ease. Try the latest CVS for this...
- Can I use MIDI events (a MIDI mixer, or the knobs on my MIDI synth) to control envelopes?
- Is there any way to have VST effects be applied during playback, as opposed to being a processing phase? (For interactive editing)
- Can I use envelopes to control the parameters of VST effects?
- Do you support mixing for more than two channels?
Sorry, not yet for any of those features. Those are all planned for our next major release.
It's looking better than the last release I tried, but I still can't shake the feeling that everything I do is "offline." If I could find a way to have more interactive access to the controls, I think I'd find it to be a more engaging experience.
One reason we went with this model is because it's incredibly frustrating for people with slower computers to have stuttering audio because their processor can't keep up with the effects they want to apply. Yes, most programs have an option to render all of the effects, but this usually takes an extra step and is non-intuitive. With Audacity, we're starting the other way - effects are always offline for now, so they work on everyone's computer - and we're slowly adding realtime effects.
In the short term, we've added effect previews and a looped-play mode. Actual realtime effects on tracks won't come until after 1.2.0 is released, though.
Big difference between theoretical discussion and debugged, tuned code.
If you'll read Ardour's documentation, you will see that because Steinberg will not release the headers for the VST files, they are not available to use with Ardour. Steinberg has not come out as hostile to Linux, but they aren't giving OSS developers any of thier cards, either.
I don't think the situation is quite as bad as that. Steinberg releases the full VST SDK for free and gives you pretty reasonable rights to do just about anything with it in your product - however you can't redistribute the VST SDK source code. In Audacity, the source code is VST-ready, but not compiled in by default. All you need to add VST support is to download the VST SDK, put it in the right directory, change one #define, and recompile. All of our binary releases do this already.
If 31 band graphic EQ could be added to Audacity, it would have everything I need and I could switch to Linux full-time.
Note that Audacity does have a FFT filter - not the same thing, but might be able to tide you over.
Programmers: does anyone have the relevant DSP code for a graphic EQ (any number of bands) that they'd be willing to donate to Audacity? I'd be happy to take care of integrating it into our codebase...
How does Ardour compare to Audacity, another free digital audio editing program? I want to know my options before switching from proprietary Cool Edit.
I agree with many of the responses, that Ardour and Audacity are as different as night and day. But I don't think some of the comparisons were quite fair, though, comparing Ardour to Quark Express and Audacity to NotePad! Maybe I'm just biased, though, as the lead developer of Audacity.
First of all, I don't think that Ardour and Audacity are directly competing. There's some feature overlap, but the user interfaces are so different (on purpose) that we're both capturing different markets, and addressing different needs.
Anyway, here are some of the major differences I can think of off the top of my head:
1. Ardour only runs on Linux. Audacity runs (completely natively) on Linux, Mac OS X, and Windows. Ardour requires Jack. Audacity only requires OSS on Linux (supported everywhere) with support for ALSA and Jack in alpha testing now (email the devel list if you want info on how to enable one of these).
2. Ardour supports true multi-track recording and playback. Audacity only records more than 2 tracks on some systems, and always mixes down to 2 output channels (stereo).
3. Both Audacity and Ardour are quite powerful, but in different ways. Ardour supports MIDI control, powerful muting/soloing, and realtime effects. Audacity has only non-realtime effects, but some of these are quite powerful in a different way, like our Noise Removal. Audacity supports plug-in effects in Ladspa (Linux), VST (Win/Mac), and Nyquist (a high-level interpreted language for audio processing).
Lots of features in common, too: both Audacity and Ardour support floating-point samples, high sample rates, resampling, LADSPA plug-ins, unlimited undo, internationalization, etc.
One last thing to dispel the myth that Audacity is "simple" compared to Ardour: I just did a quick wc, and Audacity is 70k+, Ardour is 100k+ lines of code (someone else can feel free to do a sloccount if you want). In both cases, not counting other libraries that aren't directly part of the project. No matter how you look at it, they're within a factor of 2 in terms of size and complexity, just in different areas.
We're nearing a feature freeze for Audacity 1.2.0. If you're comfortable compiling software, please check out our latest code from CVS and help us get the last few bugs out. I'd strongly encourage you to try out both Ardour and Audacity - hopefully you'll find that you use them to complement each other.
- Dominic
The professor's article consistently fails to give credit where credit is due. "Linux" is not an operating system and it never was. Linux is a part of an operating system called a kernel (which acts as a bookkeeper managing the resources of a computer so applications can share those resources without conflict). It's fair to credit the major chunks of an operating system; GNU and Linux are both valuable chunks. It's also less confusing to the reader if one refers to the union of the GNU operating system with the Linux kernel by mentioning both parts (hence the term "GNU/Linux"). For the FSF's take on this, please read their essay which also has a link to a FAQ on this issue.
So far all of SCO's claimed violations are related to code in the Linux kernel. Therefore GNU has nothing to do with this lawsuit.
You might be running a GNU/Linux operating system. More power to you. I'm running a GNU/KDE/Mozilla/Sun/Python/PHP/Apache/Linux operating system, thank you very much. GNU only contributed a tiny portion of what I use. I'm very grateful to GNU for their software contributions, but to say that GNU and Linux are the only important components of the O.S. is the height of silliness.
There's no question that GNU represents a huge chunk of most Linux-based systems, but not as large as RMS would have you believe. Here is a list of most of the software I've used at work this week:
I don't have anything against GNU. Obviously their contributions were really important. But of the tools I use on a day to day basis, I'd say that less than 10% of the code is GNU code.
Obviously usage patterns differ from person to person. Since RMS doesn't use GUIs unless he has to (really, not even GNOME!) his experience using a "GNU/Linux" system is quite different than mine.
That's why I don't call my O.S. "GNU/Linux".
All, just remember that the definition of spam is fluid. One person's spam is another's direct marketing.
No. The commonly-accepted definion of spam is (1) unsolicited (2) email that is (3) either commercial or bulk in nature. (1), (2), and (3) must all be present for something to be spam.
Why aren't forged headers part of that definition? My impression is that the vast majority of spam has forged headers, and that if we could do away with forged headers, the spam problem would go away because thousands of users would just DOS anyone who attempted to spam them.
Given that BusSpeed * ClockMultiplier = Processor Speed...Means that all three systems have the same multiplier on the chip. Which strongly implies to me that they're all the exact same chip. We'll have to wait and see how easy they are to overclock, but if you could just change the 800Mhz bus system to 1Ghz bus, you'd save yourself $1000 in the process.
You think there's any physical difference between a 1 GHz G4 and a 1.4 GHz G4 either? Or a 2.4 GHz P4 and a 2.8 GHz P4? The processors are the same - they test them and see how fast they can clock each one. The higher they clock them, the fewer chips that can handle that, so they charge more for the ones that can clock faster.
When you overclock, you're taking your chances - and you're definitely shortening the lifespan of your chip.
Besides, the $3000 Power Mac G5 has TWO processors, so you couldn't actually save $1000 by overclocking.
I agree completely. They should offer a small discount to Jaguar owners.
I honestly see this as a good thing. It's evolution. It's moving forward. And, ideally, it could benefit everyone involved. Down with the album (unless you're making a real album, and not a simple compilation of singles - read: most 'albums' released today).
Imagine this scenario. Instead of releasing a new 'album' every year, or every couple of years, or whatnot, artists would instead have the option of releasing each song as they record it.
I totally agree. That certainly doesn't mean that artists wouldn't sometimes come up with an entire album that's meant to be listened to as a whole. But I'd love to see "mini-albums" of 4-5 songs, or larger collections of 18 songs (more than 1 CD worth, but less than 2) - and with download services these could all be priced appropriately.
The fact is that until now there hasn't ever been a mode of distribution that favored any amount of music between a single (or single + bonus track) and an entire album (usually at least an hour, or it doesn't seem worth the money).
I wouldn't call Radiohead a "big crap artist".
They have some of the most loyal fans out there. If Hail to the Thief had been on the iTunes Music Store, I would have bought it there. It's not, so I ended up going to best buy to pick it up.
Radiohead may put out some great albums, but I don't have any of their CDs and I don't really know what they sound like. I'd be much more likely to spend a dollar or two to try out a few songs by a band I don't know, to see it I like those, than I would be to spend $10 to purchase the whole album.
So in my opinion, they're shooting themselves in the foot. Their fans will continue to purchase the albums, but they're losing out on the opportunity to sell a few songs to people like me who haven't discovered them yet.
No troll or flame here, but isn't it about time that Apple offered a Mac for less than $1000 that isn't completely ancient (ie, no 600 MHz iMac please). Why do they still completely forsake the low end?
I don't see how the $999 eMac counts as "ancient", with a 1 GHz G4, monitor included.
Besides, you can get a pretty decent used Mac on eBay for less than $1000. Check out these completed auctions as proof:
G4 733/512M/60G - $951
G4 733/1.5G/40G - $999
iMac 800/256M/60G - $999 (15" flat panel)
But I think the best quote I ever heard on the subject was: "Macs are only more expensive if your time has no value."
The scheme instead computes an MD5 sum for each line (actually each five lines together) and publishes the hashed versions of the files. Then anybody can do the line by line compare without ever seeing a readable version of the source code.
Except that then it's easy to figure out exactly which lines in Linux are the same (since we have access to the Linux source code) and that's exactly what SCO doesn't want us to know!
The issue here isn't that there isn't a way for SCO to tell us what was copied without giving away their IP, it's that they're refusing to play fair at all.
I've tried to install it myself but can't get past the point of needing libmoney.a
Finally, as for Napster, they deserved to lose. It was a stupid case for a great toy but a lousy business. Nobody could have won it.
Which just proves that Boies is incompetent - why would he take such a high profile case unless he thought he could win?
Clearly Boies et. al haven't done their research on the SCO case either...
You may be right, but it could be that they'd be perfectly happy with a request like "please send me all GPL software that's used in my Linksys Wireless 4-port Cable/DSL Router". That ought to be perfectly specific enough and that may have been all they wanted. That's very different than "send me all GPL software that you've ever shipped a product with", which is what they might have been trying to avoid.
Bluetooth by itself is not a big deal. But combined with technology like Apple's Rendezvous, it's really cool. Apple seems to be imagining a future where you can walk around with your laptop, and services like printers become available immediately when they come within range, then disappear when you go out of range. To Apple, Bluetooth seems to be all about service discovery.
I think we're just waiting for the killer app. Syncing your cell phone may be fun, and cordless keyboards and mice may be cool, but neither are as big a deal as wireless Internet access on your front porch.
I haven't given up hope, though. Since Bluetooth is cheap and low-power, it's not going anywhere soon.
Slot machines don't have cycles. Whether they pay out or not is entirely random (or at least as random as it is possible to make them). Your chance of winning does not increase even slightly if other people have poured piles of money into the machine already.
SOME slot machines don't have cycles, maybe MOST, but there are plenty that do.
Apparently the 'proof' that sliot machines, fruit machines as those wacky brits choose to call them, is that, if you 'freeze' the state of a fruit machine at some point and then repeat the next step, the machine will generate the same outcome.
No, that's only half of it.
The machine gives you a choice (typically "high" or "low" in the examples they gave), but you will always lose, NO MATTER WHICH ONE YOU CHOOSE.
A deterministic slot machine is one thing: even if it simply paid out exactly 1 in every 10,000 spins, that would be legal.
What's illegal (according to the authors) is that the machine presents a game and says that it's a gamble: if you choose the correct alternative (high or low) you will double your winnings, otherwise you will lose. But the machine has already decided that you will definitely lose no matter what you pick, and that's what they think is illegal under U.K. law.
In other words, playing the game at all is a gamble. If you play, you might win (no matter what you press), or you might lose (no matter what you press). However, the game is presenting choices in the middle of the gameplay as "gambles" when in reality they don't affect the outcome.