Despite what all the naysayers say, I believe we are seeing a shift in Microsoft's support of open standards towards the better.
Microsoft has always supported open standards well.
They just don't use them in anything purely Microsoft. Any Microsoft user can easily receive things using open standards. Sometimes it is a little hard for them to send things in open standards.
This is a battle royale over distribution of streaming media. In this case, Microsoft's format is already open. Neither format is a standard. Each side wants control over the server side, not the client side. Real wants to stay alive. AOL wants the streaming server marketplace to stay open. Microsoft wants EVERY server that streams audio to pay Microsoft. The same as EVERY web server.
To do this they will support open standards on their client ends. But the servers will support things propriatary to Microsoft. They merely need adequate marketshare before beginning this practice. Then it will be free client upgrades for everyone, and the server pays the piper.
Once the server is under control, leveraging the client can begin. It is all about making a buck, and slowly screwing the crap out of the consumer. The Microsoft way.
If you use shared libraries, each gets loaded into memory ONCE. This is particularly good for something like the GNOME or KDE desktop. Do you really want 13 copies of the KDE libraries loaded into RAM because you used staticly linked binaries ??
Anyway, I wouldn't sweat it. This will not become Windows. Open source apps do not install new versions of existing libraries, and libraries increment version numbers when they break compatibility.
I haven't even thought about this in a long time since apt began taking care of it for me.
It is theoretically impossible to impose a brick wall filter at some frequency without mangling the phase through the transition zone, and even then there will be amplitude effects over a substantial range below the cutoff. This is NOT an appropriate test.
Even if the average person's hearing does have a 15 dB dropoff from 5 KHz to 15 KHz (I'm surprised it's that small), 15 dB isn't much compared to the dynamic range of the ear. Remember that human hearing perceives volume logarithmically, so that 10 times louder (in amplitude) sounds nowhere near that.
Loudness perception is NOT logarithmic. It is a cube root power law. SS Stevens, the foremost 20th century sensory psychologist has written many papers on the topic (and a few books).
Even if the dynamic range of the ear is 130 dB, remember that most environments provide at least 40 dB of background noise. Then, remember that almost no one enjoys sound levels over 80 dB, and you have a dynamic range of 40 dB used for your listening pleasure. Then add cross-channel masking of the higher frequencies, and you have something only an audiophile would claim to hear.
These claims become particularly ridiculous for anyone who has followed audio equipment for a few decades. The high end changes its claims about what aspects of the sound make a difference every 5-10 years as the low end equipment satisfies the previous claims. What you end up with is a market that exists solely to provide some people with satisfaction in thinking they have superior sound reconstruction - whereas in reality they do not.
Even when people add filtering to block out higher frequencies, remember that filtering cannot cut off sharply without mangling phase through the transition zone. A MUCH better test of sound reconstruction would be to sample at 24 kHz with 24 bit sampling (or higher number of bits). The end result would tell you IF you can appreciate aspects of music over 12 kHz.
But believe me, if you generate 24 kHz sound under 200 watts, simply turn slowly your head between the speakers, and you'll find you can hear something.
Believe you me, I play 24 kHz sounds daily, and I know quite well what I can hear. The human thresholds at 24 kHz are not even reliably measurable. There is always some difficulty in creating a VERY loud 24 kHz sound that is reasonably pure in frequency content - you are likely picking up distortion when the sound is turned on or off.
For a baseline, take the threshold at 5 kHz, and then add 5 dB to get the threshold at 10 kHz, and then add another 15 dB to get the 15 dB threshold, and then stop. I can hear pure sustained very loud tones at 18 kHz, but nothing higher.
This underscores the main point. Don't worry about sound reconstruction above 12 kHz in audio equipment. The human sensitivity over that frequency drops off steeply, and no music regularly contains 15 kHz content that is 10 times louder than its 5 kHz content.
I'm very sorry to hear that you have a $10000 microphone that apparently isn't able to measure differences that are discernible to the human ear in simple blind listening tests.
I didn't claim there were not differences. Just that there was not improved sound reconstruction. Most auditory researchers use B & K microphones to measure sound levels and determine frequency:transfer functions. They are expensive because they are sensitive and have broad dynamic range.
When you get into the magnitude of the differences between well constructed $3500 systems and well constructed $40000 systems, the rest of the system plays an important role too. By "rest of the system", I mean the acoustic reflectance and placement of walls and other obstacles, things that might absorb sound, placement of speakers (especially sub woofers) relative to walls...
When an individual puts a system more expensive than that found in a concert hall in his living room without doing a thorough assessment of ALL sources of sound interference, then that person is just spending money to look like he is spending money.
I can listen to my $3500 stereo and hear how it is better than a $500 stereo. I can listen to a $40,000 stereo and tell that it improves on my own stereo. Hear it for yourself, or maybe you can't. Maybe you don't care. But there's something there, and if others value it you needn't deface their hobbies.
When I can measure that there is no improved sound resolution in the more expensive system, I can safely conclude that the audiophile in question is full of crap. Since I do auditory research for a living, I do measure this all the time. I know that the sound production in my systems in the lab does what it is supposed to do.
How many audiophiles have taken a $10000 microphone to test the fidelity of their $40000 system ??? Go ahead, take the test. Record the difference. And tell yourself what the 0.00001% improvement in signal reconstruction is worth the extra $39,500.
The point behind an increase in sample rate past 44.1 KHz is because of the interaction of harmonics on the brick-wall filtering that is taking place.
Audio frequencies above half the sample rate (Nyquist frequency) are filtered, but to avoid a particular type of aliasing the frequencies above the Nyquist frequency are translated to frequencies below, typically in the 5-8 KHz range.
You are making little sense. Almost all sound equipment contains anti-aliasing output filters to prevent aliasing. This creates a substantial problem for a researcher who works with an animal that hears above 20 kHz.
There are quite reasonable digital to analog algorithms that reconstruct the signal with translating frequencies.
Go, do yourself a favor and read some good book on signal physics. If you hear up to 12Khz, you want you equipment to be linear and sampling of the digitization to be up to at least 24Khz. Even then you may get VERY noticable artifacts due to nonlinearity of the system in the area well above 12Khz.. Basically 96Khz/24bit sampling seems to be where you really hit physiological limits.
Actually, the Nyquist sampling limit ASSUMES infinite precision sampling, which is not true. However, if you sample at 16 bits and sample at 4 times the highest frequency content of the signal, you can do quite well. You of course need to match the 16 bit range to the range of the signal.
96 kHz MIGHT be relevant if you could hear 1/4th that rate - or 24 kHz. However, you are human. Bats, monkeys, dogs, and cats all hear 24 kHz very well. Humans do not hear it at all.
When I was younger, I performed a blind test and found I could hear 22KHz. Not sure where it's at now, but it's still better than 12 KHz.
Largely irrelevant. Your hearing threshold at 12 kHz is about 7-8 times greater than it is at 5 kHz, as long as you are human.
Your audio recordings contain VERY little signal at 12 kHz that is 7-8 times louder than signal in the 1-5kHz range. It is a non-issue. You cannot hear it.
If you have a tin ear, consider yourself fortunate then go buy that Mercedes Gull Wing you always wanted, instead of the car you actually did.
This is the thing that amazes me most about audiophiles. The things they claim they can hear. I can easily hook up my Bruel and Kjaer sound level meter microphone, and demonstrate perfect reconstruction from cheap speakers. I can demonstrate to them that they can hardly hear anything over 12 kHz. I can demonstrate that their hearing threshold at 15 kHz is about 50 times higher than it is at 5 kHz.
Yet still, they continue to buy $10k+ speakers. They continue to buy 96 kHz sound cards. It is not a form of addiction - it is a way of peacocking around. Ludicrous. It is as though they were unaware that total harmonic distortion had been a non-issue in even cheap sound amplifiers for over a decade.
Then walk over to the same person's computer and listen to the 50 dB whine from the hard drive...
Well, if you use ctags, emacs/jed, gcc, gdb, and a few xterms, I wonder what additional functionality you want in an IDE.
The reality is that UNIX philosophy leads to small tools that do their jobs well. Hence, emacs, gdb, gcc, a few xterms... (actually, use vi or jed instead of emacs for the UNIX philosophy of small tools).
This philosophy is useless under a Windows environment in which the standard shell is functionally broken by UNIX standards (no pipes, for example). Why would you need instant access to header files when you are using ctags ?
So you might start by asking how you can maintain your Windows programming mentality under linux, or you might more intelligently ask what approach large projects take under linux. As a hint, GNOME and the kernel are mostly programmed with a few consoles, emacs/vi/jed, and SOMETIMES gdb and ctags. KDE has its own C++ builder, but I don't know how many people use it.
The underlying tools take a completely different approach, which makes completely different programming approaches effective.
Come on, don't spew the company line. Talk in plain English.
.NET is several things to Microsoft. First and foremost it is intended to kill Java and any other competition for internet based apps. Secondly, it is designed to interoperate with Microsoft's new revenue model in which each person pays $100 each year for the privilege of using Office, and varying amounts for other Microsoft apps. Thirdly, it is meant to give Microsoft a stranglehold on all internet based communications..NET will require Microsoft name resolution, Microsoft SMTP, and just about anything else from Microsoft as well. The internet will be a COMPLETELY different world if you are running Windows than if you are not.
In short, with.NET Microsoft hopes to do to the internet what they did to the desktop in the early and mid 90s - own it and leverage it. It will promise benefits to the consumer but provide none.
The first thing to note is that it is not all the fan. One of the noisier things in a computer is the disk drive. Quantum, for one, has disk drives they have designed for the quiet, and Seagate has followed suit. So, I would aim for one of those.
Faster drives, ceteris paribus, are louder. A nice 5400 RPM drive can be much less noisy than a 10000 RPM drive.
Drive enclosures are available, especially from PC Power and Cooling. And custom cases are as well, and custom low noise power supplies.
Fans are often a culprit, and they can be replaced or have their voltage tweaked down by adding a resistor in series, or changing the power supply to the fan, or replacing the fan.
But putting the entire box in the next room is cheaper and easier. All of these ideas are after the fact add-on hacks. No major computer maker designs for low noise in the US.
AOL is the perfect example of a company that can be successful despite Microsoft. Say what you will about AOL, but the secret to their success was the fact that any idiot could plop a floppy into the drive and it "just worked". In fact, for a long time it was totally self contained -- it found the modem and used it's own communication software. No Internet setup, no TCP/IP setting, no nothing. Boom! You're connected.
You are right. For the nationwide ISP business, getting a user connected easily the first time is incredibly important.
Now think about how important it was to have an AOL icon on the first boot screen for every user of every flavor of Windows. Combine that with their streamlined sign up process, and you have a major reason for their success.
Clearly the market is changing, and having that icon is not worth what it once was. People have name-brand recognition of ISPs, and there are other ways to reach users. But, in marketing, you never want to cut off an advantage, especially one as prominent as being featured on every Windows machine.
And, in exchange, Microsoft is asking AOL to kill REAL. What is that called - when you give something to a competitor in order for them to hurt another competitor ?? Collusion ???
Following from the comment you quoted, the exammples of Sun, HP, and IBM surely carefully check that their own GPL'ed projects are not contaminating their proprietary code (then again, as I'm not employed by any of those companies, I couldn't tell you for sure).
GPL code cannot "infect" other code. This is an error. The license, the GPL, grants the recipient of the software right to redistribute, and the right to redistribute changes. As long as the GPL goes with it.
However, if some other piece of code interacts with a well-described API in the GPL work, then the other piece of code is not governed by the GPL. For example, if the GPl code is a library, and the library API is public in the sense that multiple implementations for that API exist, then code that calls functions in the GPL code are not governed by the GPL license - they do not constitute copyright derivatives, since they match a public API and are not reliant on a specific wording of that API in its source code.
In the case of a unique GPL library, the court is currently deciding on whether linking to that is a copyright violation.
This is a myth - that Microsoft machines require less sys admins than Linux boxen.
First, of all, the software is designed to look and feel just like Microsoft Office. The users should require no further training than they required for MS Office (which is to say none). Secondly, ALL MS network infrastructural strategies are more difficult and require more admins than UNIX infrastructure. Let's face it - UNIX has been doing infrastructure for decades already, and Microsoft STILL is not a presence in the high end server market.
Generally, when comparisons are done, there are 2-3 times more admins for MS boxes than for Unix boxes. Of course, the MS admins are more clueless, in general, and cheaper, so admin time may be a wash.
McNealy says So far the industry has done a pretty good job of regulating itself. Most companies now post formal privacy policies on their Web sites and allow visitors to have a say in how information about them is used.
This is pure unadulterated bull. Amazon sells records on its customers. I've yet to find a single closed source internet app that doesn't track the moves of all its users. The basic finding of the internet era is that companies WILL abuse your privacy for their own financial gain, and you have to go to great lengths to stop them.
I agree with McNealy that selective access to private records is a great concept. However, in its current unregulated form, we can expect NO privacy of any sort in the future. Companies with an interest in you will be able to contact your ISP and get all the web sites you've hit in the last 6 months, the things you've bought... This will happen because people will violate your privacy for money. This I know because it happens every day. Real networks just released a new unzip free (as in beer) software app - the catch is that Real tracks every zip archive you download just as they currently track every media download you make.
Real legislation needs to be crafted that protects the privacy rights of the consumer, and allows it to be unprotected at his request. Currently, it just is not working, and it is rapidly turning into a morass.
Why is right, why the heck would you even think of such crazyness?
* Wine runs apps 75% of the time, and thats being forgiving. What happens when someone wants to run an app with unsupported API's. Say 'Too Bad'?
Why run WINE at all ?
* Just because you have VMWARE doesnt mean you get to run Windows for free, you still need licences. So whats the point? And having to boot that VMware session will go over lovely with users. "You mean I have to boot 2 computers to run Excel!?!!?"
Why run VMWare at all ?
* CITRIX? Do you really want to enter the era of the mainframe again? Citrix has its advantages however. Being cheap is not one of those advantages.
Kill the terminal servers for Windows - Microsoft bought them out anyway.
* Staroffice SUCKS ARSE -- Theres no talking anyone out of that. Staroffice better than Microsoft Office? LAUGHABLE! It would be better to use a web-based office like Thinkfree.com.
Either StarOffice, WordPerfect, Abiword, Lyx, or ApplixWord would serve as a word replacement. For spreadsheets, there are even more options. For slide presentation, even more.
The bare fact is that all the functionality that 99.99% of all office users exists in easy to find places in linux, works more consistently, and is free.
You need to lose your politics, and figure out what is best for the user, before you lose your job.
It is not at all clear that you are suggesting anything more than select the software the office wants without respect to the cost, or projected future cost.
For 100 users, Office will run you $100/user if it is OEM installed, and about $300 for upgrades 2-3 years from now. that comes out around $10000/yr for office software.
Whereas there are plenty of offices where linux already functions as the desktop OS (like the studio that made Shrek).
Guess what will happen when they finally make the 1.0 release of Mozilla?
Look - AOL is bankrolling Mozilla. Even if they are continuing to use IE on Windows, the money they are investing is trivial compared to the comfort of knowing that Microsoft has no leverage in removing access to their browser.
AOL has a yearly gross of 4-6 BILLION. The investment in programmers for mozilla is about 5-10 million. And for that, they can make mozilla stay up to date, and ready for use as AOL's browser when and if they should need it.
And that leaves them free to use IE in exchange for being on every desktop on first boot. And that is the true value to AOL.
The GPL cannot be more specific on issues like this - it is beyond its control.
The GPL only applies to copyright issues. If another program interfaces in some way with a GPL program, but is not considered a derivative work, then the interfacing program is not bound by the GPL. Period. There is no re-wording of the GPL that will affect this in any way. GPL is limited by copyright. Copyright can only affect other programs that are derivatives.
The entire question of this lawsuit is whether using a dynamic link library makes something a derivative work. Consider - in many cases you can change the DLL but preserve the API without changing the function of the program. In those cases dynamic linking is clearly NOT making something a derivative work.
In other cases it may be less clear.
Re:Things RMS didn't forsee in 1984
on
GPL FAQ
·
· Score: 3
When the GPL was being written, nobody was using DLLs or object brokers and "linking" was much more straightforward than it is today. This must be better addressed in a future version of the GPL.
It won't matter.
These things are all governed by whether the work is or is not a derivative work in the copyright sense, and it does not matter in the slightest what the GPL or RMS' lawyers say about it.
Recall, the GPL does not extend copyrights by engaging the user in a contract before use, as would be the case in an EULA.
Instead, the GPL removes some rights that are held intrinsically by the author of copyright. In particular, it allows redistribution as long as the source is redistributed and GPLd, and it allows redistribution of derivatives as long as the derivative is GPLd.
But, if it is determined than dynamic linking, or component plug-ins, are NOT derivative works, then the GPL library or component architecture is irrelevant wrt the licensing of a calling program or plug-in.
Copyright protects expression, not ideas, methods, or function. To protect those, you need patent level of protection, or you need to supercede copyright by entering a contract with the user (such as an EULA or NDA). The GPL is not a contract - it is a specific waiver of some rights normally held by authors of copyright.
To re-iterate - the issue is what constitutes a derivative in computer systems in the copyright sense, and this sets bounds on what the GPL can cover.
Just because something links to the kernel doesn't mean it is a derivative work, because the code could be "reasonably considered independent."
Thanks for making my point. Something that requires an API, but not specific expression used to generate that API, is not a derived work, and thus the license of the program providing the API is irrelevant.
In this particular case, Vidomi's work is clearly derivative and an extension of the GPL'd work. Their intent of the DLLs was to circumvent the GPL.
I think they will argue the intent of the DLL was to make a clear separation between the API provided by the GPLd work and the code that they generated in house to provide additional functionality. Anyone may download the GPl'd code - thus it cannot form a substantive part of Vidomi's intellectual property. The IP is only the in house proprietary portion of the code.
Copyright does not protect ideas and methods embodied in the copyrighted work. In that sense, I do not think library APIs are protected by copyright law.
Now, get yourself an EULA, and everything may be different...
The library provides an API. This API can be legally reverse engineered. It is a concept embodied by the source code. If that is true, it is not protected by standard copyright law, and is not protected by the GPL. The API, that is. The source code IS protected, just ideas and concepts embodied in it are not.
That would be a patent level of protection, not a copyright level of protection.
Note that Vidori is distributing a binary that LINKS DYNAMICALLY to a library containing GPLd code. The binary they distribute requires the API, but not the original source, to work.
Slashdot Mirror
Despite what all the naysayers say, I believe we are seeing a shift in Microsoft's support of open standards towards the better.
Microsoft has always supported open standards well.
They just don't use them in anything purely Microsoft. Any Microsoft user can easily receive things using open standards. Sometimes it is a little hard for them to send things in open standards.
This is a battle royale over distribution of streaming media. In this case, Microsoft's format is already open. Neither format is a standard. Each side wants control over the server side, not the client side. Real wants to stay alive. AOL wants the streaming server marketplace to stay open. Microsoft wants EVERY server that streams audio to pay Microsoft. The same as EVERY web server.
To do this they will support open standards on their client ends. But the servers will support things propriatary to Microsoft. They merely need adequate marketshare before beginning this practice. Then it will be free client upgrades for everyone, and the server pays the piper.
Once the server is under control, leveraging the client can begin. It is all about making a buck, and slowly screwing the crap out of the consumer. The Microsoft way.
RAM usage.
If you use shared libraries, each gets loaded into memory ONCE. This is particularly good for something like the GNOME or KDE desktop. Do you really want 13 copies of the KDE libraries loaded into RAM because you used staticly linked binaries ??
Anyway, I wouldn't sweat it. This will not become Windows. Open source apps do not install new versions of existing libraries, and libraries increment version numbers when they break compatibility.
I haven't even thought about this in a long time since apt began taking care of it for me.
No you moron.
It is theoretically impossible to impose a brick wall filter at some frequency without mangling the phase through the transition zone, and even then there will be amplitude effects over a substantial range below the cutoff. This is NOT an appropriate test.
Even if the average person's hearing does have a 15 dB dropoff from 5 KHz to 15 KHz (I'm surprised it's that small), 15 dB isn't much compared to the dynamic range of the ear. Remember that human hearing perceives volume logarithmically, so that 10 times louder (in amplitude) sounds nowhere near that.
Loudness perception is NOT logarithmic. It is a cube root power law. SS Stevens, the foremost 20th century sensory psychologist has written many papers on the topic (and a few books).
Even if the dynamic range of the ear is 130 dB, remember that most environments provide at least 40 dB of background noise. Then, remember that almost no one enjoys sound levels over 80 dB, and you have a dynamic range of 40 dB used for your listening pleasure. Then add cross-channel masking of the higher frequencies, and you have something only an audiophile would claim to hear.
These claims become particularly ridiculous for anyone who has followed audio equipment for a few decades. The high end changes its claims about what aspects of the sound make a difference every 5-10 years as the low end equipment satisfies the previous claims. What you end up with is a market that exists solely to provide some people with satisfaction in thinking they have superior sound reconstruction - whereas in reality they do not.
Even when people add filtering to block out higher frequencies, remember that filtering cannot cut off sharply without mangling phase through the transition zone. A MUCH better test of sound reconstruction would be to sample at 24 kHz with 24 bit sampling (or higher number of bits). The end result would tell you IF you can appreciate aspects of music over 12 kHz.
But believe me, if you generate 24 kHz sound under 200 watts, simply turn slowly your head between the speakers, and you'll find you can hear something.
Believe you me, I play 24 kHz sounds daily, and I know quite well what I can hear. The human thresholds at 24 kHz are not even reliably measurable. There is always some difficulty in creating a VERY loud 24 kHz sound that is reasonably pure in frequency content - you are likely picking up distortion when the sound is turned on or off.
For a baseline, take the threshold at 5 kHz, and then add 5 dB to get the threshold at 10 kHz, and then add another 15 dB to get the 15 dB threshold, and then stop. I can hear pure sustained very loud tones at 18 kHz, but nothing higher.
This underscores the main point. Don't worry about sound reconstruction above 12 kHz in audio equipment. The human sensitivity over that frequency drops off steeply, and no music regularly contains 15 kHz content that is 10 times louder than its 5 kHz content.
I'm very sorry to hear that you have a $10000 microphone that apparently isn't able to measure differences that are discernible to the human ear in simple blind listening tests.
I didn't claim there were not differences. Just that there was not improved sound reconstruction. Most auditory researchers use B & K microphones to measure sound levels and determine frequency:transfer functions. They are expensive because they are sensitive and have broad dynamic range.
When you get into the magnitude of the differences between well constructed $3500 systems and well constructed $40000 systems, the rest of the system plays an important role too. By "rest of the system", I mean the acoustic reflectance and placement of walls and other obstacles, things that might absorb sound, placement of speakers (especially sub woofers) relative to walls...
When an individual puts a system more expensive than that found in a concert hall in his living room without doing a thorough assessment of ALL sources of sound interference, then that person is just spending money to look like he is spending money.
I can listen to my $3500 stereo and hear how it is better than a $500 stereo. I can listen to a $40,000 stereo and tell that it improves on my own stereo. Hear it for yourself, or maybe you can't. Maybe you don't care. But there's something there, and if others value it you needn't deface their hobbies.
When I can measure that there is no improved sound resolution in the more expensive system, I can safely conclude that the audiophile in question is full of crap. Since I do auditory research for a living, I do measure this all the time. I know that the sound production in my systems in the lab does what it is supposed to do.
How many audiophiles have taken a $10000 microphone to test the fidelity of their $40000 system ??? Go ahead, take the test. Record the difference. And tell yourself what the 0.00001% improvement in signal reconstruction is worth the extra $39,500.
The point behind an increase in sample rate past 44.1 KHz is because of the interaction of harmonics on the brick-wall filtering that is taking place.
Audio frequencies above half the sample rate (Nyquist frequency) are filtered, but to avoid a particular type of aliasing the frequencies above the Nyquist frequency are translated to frequencies below, typically in the 5-8 KHz range.
You are making little sense. Almost all sound equipment contains anti-aliasing output filters to prevent aliasing. This creates a substantial problem for a researcher who works with an animal that hears above 20 kHz.
There are quite reasonable digital to analog algorithms that reconstruct the signal with translating frequencies.
Go, do yourself a favor and read some good book on signal physics. If you hear up to 12Khz, you want you equipment to be linear and sampling of the digitization to be up to at least 24Khz. Even then you may get VERY noticable artifacts due to nonlinearity of the system in the area well above 12Khz.. Basically 96Khz/24bit sampling seems to be where you really hit physiological limits.
Actually, the Nyquist sampling limit ASSUMES infinite precision sampling, which is not true. However, if you sample at 16 bits and sample at 4 times the highest frequency content of the signal, you can do quite well. You of course need to match the 16 bit range to the range of the signal.
96 kHz MIGHT be relevant if you could hear 1/4th that rate - or 24 kHz. However, you are human. Bats, monkeys, dogs, and cats all hear 24 kHz very well. Humans do not hear it at all.
When I was younger, I performed a blind test and found I could hear 22KHz. Not sure where it's at now, but it's still better than 12 KHz.
Largely irrelevant. Your hearing threshold at 12 kHz is about 7-8 times greater than it is at 5 kHz, as long as you are human.
Your audio recordings contain VERY little signal at 12 kHz that is 7-8 times louder than signal in the 1-5kHz range. It is a non-issue. You cannot hear it.
If you have a tin ear, consider yourself fortunate then go buy that Mercedes Gull Wing you always wanted, instead of the car you actually did.
This is the thing that amazes me most about audiophiles. The things they claim they can hear. I can easily hook up my Bruel and Kjaer sound level meter microphone, and demonstrate perfect reconstruction from cheap speakers. I can demonstrate to them that they can hardly hear anything over 12 kHz. I can demonstrate that their hearing threshold at 15 kHz is about 50 times higher than it is at 5 kHz.
Yet still, they continue to buy $10k+ speakers. They continue to buy 96 kHz sound cards. It is not a form of addiction - it is a way of peacocking around. Ludicrous. It is as though they were unaware that total harmonic distortion had been a non-issue in even cheap sound amplifiers for over a decade.
Then walk over to the same person's computer and listen to the 50 dB whine from the hard drive...
Well, if you use ctags, emacs/jed, gcc, gdb, and a few xterms, I wonder what additional functionality you want in an IDE.
The reality is that UNIX philosophy leads to small tools that do their jobs well. Hence, emacs, gdb, gcc, a few xterms... (actually, use vi or jed instead of emacs for the UNIX philosophy of small tools).
This philosophy is useless under a Windows environment in which the standard shell is functionally broken by UNIX standards (no pipes, for example). Why would you need instant access to header files when you are using ctags ?
So you might start by asking how you can maintain your Windows programming mentality under linux, or you might more intelligently ask what approach large projects take under linux. As a hint, GNOME and the kernel are mostly programmed with a few consoles, emacs/vi/jed, and SOMETIMES gdb and ctags. KDE has its own C++ builder, but I don't know how many people use it.
The underlying tools take a completely different approach, which makes completely different programming approaches effective.
Come on, don't spew the company line. Talk in plain English.
.NET will require Microsoft name resolution, Microsoft SMTP, and just about anything else from Microsoft as well. The internet will be a COMPLETELY different world if you are running Windows than if you are not.
.NET Microsoft hopes to do to the internet what they did to the desktop in the early and mid 90s - own it and leverage it. It will promise benefits to the consumer but provide none.
.NET is several things to Microsoft. First and foremost it is intended to kill Java and any other competition for internet based apps. Secondly, it is designed to interoperate with Microsoft's new revenue model in which each person pays $100 each year for the privilege of using Office, and varying amounts for other Microsoft apps. Thirdly, it is meant to give Microsoft a stranglehold on all internet based communications.
In short, with
The first thing to note is that it is not all the fan. One of the noisier things in a computer is the disk drive. Quantum, for one, has disk drives they have designed for the quiet, and Seagate has followed suit. So, I would aim for one of those.
Faster drives, ceteris paribus, are louder. A nice 5400 RPM drive can be much less noisy than a 10000 RPM drive.
Drive enclosures are available, especially from PC Power and Cooling. And custom cases are as well, and custom low noise power supplies.
Fans are often a culprit, and they can be replaced or have their voltage tweaked down by adding a resistor in series, or changing the power supply to the fan, or replacing the fan.
But putting the entire box in the next room is cheaper and easier. All of these ideas are after the fact add-on hacks. No major computer maker designs for low noise in the US.
AOL is the perfect example of a company that can be successful despite Microsoft. Say what you will about AOL, but the secret to their success was the fact that any idiot could plop a floppy into the drive and it "just worked". In fact, for a long time it was totally self contained -- it found the modem and used it's own communication software. No Internet setup, no TCP/IP setting, no nothing. Boom! You're connected.
You are right. For the nationwide ISP business, getting a user connected easily the first time is incredibly important.
Now think about how important it was to have an AOL icon on the first boot screen for every user of every flavor of Windows. Combine that with their streamlined sign up process, and you have a major reason for their success.
Clearly the market is changing, and having that icon is not worth what it once was. People have name-brand recognition of ISPs, and there are other ways to reach users. But, in marketing, you never want to cut off an advantage, especially one as prominent as being featured on every Windows machine.
And, in exchange, Microsoft is asking AOL to kill REAL. What is that called - when you give something to a competitor in order for them to hurt another competitor ?? Collusion ???
Following from the comment you quoted, the exammples of Sun, HP, and IBM surely carefully check that their own GPL'ed projects are not contaminating their proprietary code (then again, as I'm not employed by any of those companies, I couldn't tell you for sure).
GPL code cannot "infect" other code. This is an error. The license, the GPL, grants the recipient of the software right to redistribute, and the right to redistribute changes. As long as the GPL goes with it.
However, if some other piece of code interacts with a well-described API in the GPL work, then the other piece of code is not governed by the GPL. For example, if the GPl code is a library, and the library API is public in the sense that multiple implementations for that API exist, then code that calls functions in the GPL code are not governed by the GPL license - they do not constitute copyright derivatives, since they match a public API and are not reliant on a specific wording of that API in its source code.
In the case of a unique GPL library, the court is currently deciding on whether linking to that is a copyright violation.
This is a myth - that Microsoft machines require less sys admins than Linux boxen.
First, of all, the software is designed to look and feel just like Microsoft Office. The users should require no further training than they required for MS Office (which is to say none). Secondly, ALL MS network infrastructural strategies are more difficult and require more admins than UNIX infrastructure. Let's face it - UNIX has been doing infrastructure for decades already, and Microsoft STILL is not a presence in the high end server market.
Generally, when comparisons are done, there are 2-3 times more admins for MS boxes than for Unix boxes. Of course, the MS admins are more clueless, in general, and cheaper, so admin time may be a wash.
McNealy says So far the industry has done a pretty good job of regulating itself. Most companies now post formal privacy policies on their Web sites and allow visitors to have a say in how information about them is used.
This is pure unadulterated bull. Amazon sells records on its customers. I've yet to find a single closed source internet app that doesn't track the moves of all its users. The basic finding of the internet era is that companies WILL abuse your privacy for their own financial gain, and you have to go to great lengths to stop them.
I agree with McNealy that selective access to private records is a great concept. However, in its current unregulated form, we can expect NO privacy of any sort in the future. Companies with an interest in you will be able to contact your ISP and get all the web sites you've hit in the last 6 months, the things you've bought... This will happen because people will violate your privacy for money. This I know because it happens every day. Real networks just released a new unzip free (as in beer) software app - the catch is that Real tracks every zip archive you download just as they currently track every media download you make.
Real legislation needs to be crafted that protects the privacy rights of the consumer, and allows it to be unprotected at his request. Currently, it just is not working, and it is rapidly turning into a morass.
Why is right, why the heck would you even think of such crazyness?
* Wine runs apps 75% of the time, and thats being forgiving. What happens when someone wants to run an app with unsupported API's. Say 'Too Bad'?
Why run WINE at all ?
* Just because you have VMWARE doesnt mean you get to run Windows for free, you still need licences. So whats the point? And having to boot that VMware session will go over lovely with users. "You mean I have to boot 2 computers to run Excel!?!!?"
Why run VMWare at all ?
* CITRIX? Do you really want to enter the era of the mainframe again? Citrix has its advantages however. Being cheap is not one of those advantages.
Kill the terminal servers for Windows - Microsoft bought them out anyway.
* Staroffice SUCKS ARSE -- Theres no talking anyone out of that. Staroffice better than Microsoft Office? LAUGHABLE! It would be better to use a web-based office like Thinkfree.com.
Either StarOffice, WordPerfect, Abiword, Lyx, or ApplixWord would serve as a word replacement. For spreadsheets, there are even more options. For slide presentation, even more.
The bare fact is that all the functionality that 99.99% of all office users exists in easy to find places in linux, works more consistently, and is free.
You need to lose your politics, and figure out what is best for the user, before you lose your job.
It is not at all clear that you are suggesting anything more than select the software the office wants without respect to the cost, or projected future cost.
For 100 users, Office will run you $100/user if it is OEM installed, and about $300 for upgrades 2-3 years from now. that comes out around $10000/yr for office software.
Whereas there are plenty of offices where linux already functions as the desktop OS (like the studio that made Shrek).
Kung Fu Fighting in Flash
Guess what will happen when they finally make the 1.0 release of Mozilla?
Look - AOL is bankrolling Mozilla. Even if they are continuing to use IE on Windows, the money they are investing is trivial compared to the comfort of knowing that Microsoft has no leverage in removing access to their browser.
AOL has a yearly gross of 4-6 BILLION. The investment in programmers for mozilla is about 5-10 million. And for that, they can make mozilla stay up to date, and ready for use as AOL's browser when and if they should need it.
And that leaves them free to use IE in exchange for being on every desktop on first boot. And that is the true value to AOL.
The GPL cannot be more specific on issues like this - it is beyond its control.
The GPL only applies to copyright issues. If another program interfaces in some way with a GPL program, but is not considered a derivative work, then the interfacing program is not bound by the GPL. Period. There is no re-wording of the GPL that will affect this in any way. GPL is limited by copyright. Copyright can only affect other programs that are derivatives.
The entire question of this lawsuit is whether using a dynamic link library makes something a derivative work. Consider - in many cases you can change the DLL but preserve the API without changing the function of the program. In those cases dynamic linking is clearly NOT making something a derivative work.
In other cases it may be less clear.
When the GPL was being written, nobody was using DLLs or object brokers and "linking" was much more straightforward than it is today. This must be better addressed in a future version of the GPL.
It won't matter.
These things are all governed by whether the work is or is not a derivative work in the copyright sense, and it does not matter in the slightest what the GPL or RMS' lawyers say about it.
Recall, the GPL does not extend copyrights by engaging the user in a contract before use, as would be the case in an EULA.
Instead, the GPL removes some rights that are held intrinsically by the author of copyright. In particular, it allows redistribution as long as the source is redistributed and GPLd, and it allows redistribution of derivatives as long as the derivative is GPLd.
But, if it is determined than dynamic linking, or component plug-ins, are NOT derivative works, then the GPL library or component architecture is irrelevant wrt the licensing of a calling program or plug-in.
Copyright protects expression, not ideas, methods, or function. To protect those, you need patent level of protection, or you need to supercede copyright by entering a contract with the user (such as an EULA or NDA). The GPL is not a contract - it is a specific waiver of some rights normally held by authors of copyright.
To re-iterate - the issue is what constitutes a derivative in computer systems in the copyright sense, and this sets bounds on what the GPL can cover.
Just because something links to the kernel doesn't mean it is a derivative work, because the code could be "reasonably considered independent."
Thanks for making my point. Something that requires an API, but not specific expression used to generate that API, is not a derived work, and thus the license of the program providing the API is irrelevant.
In this particular case, Vidomi's work is clearly derivative and an extension of the GPL'd work. Their intent of the DLLs was to circumvent the GPL.
I think they will argue the intent of the DLL was to make a clear separation between the API provided by the GPLd work and the code that they generated in house to provide additional functionality. Anyone may download the GPl'd code - thus it cannot form a substantive part of Vidomi's intellectual property. The IP is only the in house proprietary portion of the code.
Copyright does not protect ideas and methods embodied in the copyrighted work. In that sense, I do not think library APIs are protected by copyright law.
Now, get yourself an EULA, and everything may be different...
The library provides an API. This API can be legally reverse engineered. It is a concept embodied by the source code. If that is true, it is not protected by standard copyright law, and is not protected by the GPL. The API, that is. The source code IS protected, just ideas and concepts embodied in it are not.
That would be a patent level of protection, not a copyright level of protection.
Note that Vidori is distributing a binary that LINKS DYNAMICALLY to a library containing GPLd code. The binary they distribute requires the API, but not the original source, to work.