Well, I cancelled my Paypal account and gave the reason "WikiLeaks". No need to support a company that enables governments to silence free journalism, even if for the most part, the only "business" they got from me was to boast with one additional customer served...
Chorded keyboards are way more efficient than that! They should integrate a Twiddler with a cell phone. According to Prof. Thad Starner's testimonial on the website, 60 words per minute are entirely possible, and you'll get 10-30 words per minute after only a weekend of practice! I actually recall him telling me about even higher speeds, but I don't want to misquote him...
For comparison, here's some info on normal QWERTY keyboards: "Someone having minor experience with keyboards can reach 20 words per minute, an average typist reaches about 30 to 45 (usually the minimum required for dispatch positions and other typing jobs), while advanced typists work at speeds above 60." (http://en.wikipedia.org/wiki/Words_per_minute)
Ahh, this sounds very familiar. Indeed, every scientist should be aware of the possible flaws in research. If you find someone who takes research to be final, or the truth - it's not really a scientist.
I'm in a graduate level Research Design class right now, so - in a nutshell - here's what we were told science is about (and what it's not about):
Science has the following goals:
description (of phenomena)
understanding (hypotheses about relationships)
prediction (this requires general applicability of hypotheses)
control (use knowledge to influence phenomena)
And the key values of science are:
Empiriscism (objective evidence!)
Skepticism (available knowledge may always contain errors)
Tentativeness (scientific knowledge is never the truth, but always a theory that tries to come as close as currently possible)
Publicness (results and methodology must be published so others can replicate results, thereby establishing validity)
See how some of the basic points we're taught are also what the article tries to point out? This should be common knowledge, at least to scientists. And non-scientists that argue against science usually go against this uncertainty, as if tentativeness were something bad. "It's just a theory" should not be used in a derogatory sense.
As the first post managed to (once again) get this discussion rolling down the Evolution(ism)/Intelligent Design(ism) road, here are some useful indicators of Pseudo-Science, also from my class:
Subjective
Illogical
Unsystematic
Not empirically testable
No predictive power
Knowledge system is fixed or closed
Anachronistic thinking
Seeking mystery or lack of patterns
Appeal to myths
Casual approach to evidence
Irrefutable hypotheses
Explanation by scenario
"Literary" interpretation
Refusal to revise
To make this discussion a little more clear: "Irrefutable hypotheses" in my understanding is similar (if not equal) to a hypothesis not being "falsifiable", a term others have repeatedly used here. So basically: a theory is worthless if its hypotheses can not possibly be proven wrong by later research. This is because from such a hypothesis you can come to any conclusion. As as example to this I especially liked the 6-second universe theory stated by someone earlier in this discussion. As you can't prove him wrong, the theory is worthless to the basic goals of science.
The DRM I'm willing to accept os the DRM that I won't even notice. Like the one on the iTMS seem to be.
Not even iTMS is unnoticable. I found that out yesterday, after I sent my Mac to be repaired. I copied my entire system to another (nearly identical) Powerbook, so I thought I wouldn't run into problems while my machine is gone...
Not only can I no longer play my purchased songs (of course I could authorize my replacement machine), but I just realized that since the motherboard is being replaced on my machine, I've probably permanently wasted one of my 5 allowed "authorized" machines, cause when I get it back, it won't be the "same" machine to iTunes!
Good thing I only made a small test-run with iTMS and purchased 4 songs. That doesn't hurt too much. But others will run into larger problems, that's for sure.
To me, DRM will only be acceptable as a passive measure. Indestructible watermarks are ok, if they don't impede my use of the purchased material. If files appear on the net and the company holding the rights immediately knows who leaked them and comes after that person, that's only fair. Also, that way, it will be far easier to measure the amount of damage inflicted by an individual file sharer.
I know. I use a Mac. And the Windows Media Player sucks. Big time. I mean, I can't even move to different parts of the video in most files. And file errors that aren't even visible on Windows will make the Mac version stutter.
You are also missing a point: this is not a "Not-Only-On-Windows" discussion. It's about formats that are likely to be well supported on the majority of multimedia-capable operating systems.
Many people are saying that "quicktime" runs on Linux. They mean some quicktime files, not all, right? I.E., primitive and uncompressed files run on linux; serious compression Quick Time files can not legally be run on Linux.
They can play complex formats, using a hack that requires Windows.DLLs. This obviously means you can only use QuickTime on Intel compatible Linux machines, as the libraries are in binary format.
So that's not really supporting Linux in its entirety... at all.:-(
You must be careful to differentiate between container formats and codecs:
Containers combine encoded audio and video, and possibly metadata. This usually means interleaving audio and video according to their time in the movie, so during playback your disk doesn't die from constant seeking between the audio and video portions.
Codecs are used to compress the raw audio and video to the desired size, usually reducing the quality (lossy compression).
As a container format, you mainly have the following options:
.AVI (AudioVideoInterleaved): a really old format that just interlaces audio and video data (even mp3 audio is basically hacked into working with this - badbadbad)
.WMV/.ASF: Microsoft stuff. Don't use, if you want compatibility with anything but Windows.
QuickTime.MOV (MooV actually): Apple stuff. Officially supported on Macs and Windows, but still proprietary - you're not being nice to OSS users.
RealMedia.RM: proprietary (see QuickTime)
MPEG-4: New standard by the people who brought us MPEG-1 (crappy low-res by todays standards) and MPEG-2 (DVD video). It's based on the QuickTime container, but it's a public standard (not proprietary). Costs developers to get a license, though.
Ogg: Open/Free container format. Great for OSS people, but less known than MPEG-4.
I'd recommend looking into using MPEG-4 or Ogg containers.
For video compression, whether you use MPEG-4 or Ogg, go with XVID. Theora is still in development, and everything else is a mess by comparison. (flaming ensues;) )
For audio compression, with MPEG-4 you will want to use AAC or MP3 (not sure about the latter), with Ogg containers go with Ogg Vorbis (best quality at low bitrates, IMHO) or MP3.
By sticking to a standard, but non-proprietary combination, such as MPEG-4/XVID/AAC, you might even be able to cater to all platforms without maintaining multiple formats...
Such JPEGs should not exist. The Huffman coding step should remove most of the redundancy, thereby eliminating the possibility of "bad" compression.
If you increase the compression level, what happens is: you loose more information. But the resulting jpeg should still be mostly incompressible, because after this quantization step, further steps (Huffman etc.) ensure that the remaining data is mostly non-redundant. It's kinda like JPEG uses its own compressor by default, and a pretty good one at that.
So this new StuffIt technique really IS quite impressive.
Argh. Yeah, and they keep repeating that mistake, too - really bothersome in an otherwise well-written article.
Apple never mentioned AC-2 anywhere. In fact, they usually expand AAC to Advanced Audio Coding in their explanations. I wonder where the author got that wrong idea.
Let's see how long it takes for this myth to spread across the internet...
Ah, memories... the Tandy 1400 LT was my first home computer. I must have been around 9 at the time. Before that, I always played on friends' computers (C64, Atari, Neckermann etc.) or on the Siemens and VAX systems at my father's lab.
I remember getting to know DOS the hard way, because nobody told me to use "dir" to look for "*.exe" files (once I knew, I executed just about anything I could find...)
Later, I spent weeks "optimizing" boot time by using a RAM disk and playing LHX Attack Chopper on its CGA display. Luckily, I didn't go blind.:-)
Today, its mostly Macs, my Sun and my NextStation running OS X, Linux, OpenBSD and NeXTStep... MS-DOS 6.0 was bloated after all, so I never checked with Microsoft again.;-)
Eeek! Good thing Steve got it removed. I don't want to imagine in what direction Apple would head without his clear vision of future computers! No, wait...
That's what I was thinking. I'd like to own an Apple phone as I'm expecting it to be easier to use, better looking, well integrated with iTunes and nearly bug-free.
But right now, I've got a Sony Ericsson P800 with a 64 MB Memory Stick. It plays OGG Vorbis using OggPlay, and I can convert any song or entire playlists to OGG directly from iTunes using an AppleScript I wrote for that purpose. It internally uses mplayer and oggenc for conversion.
For mobile use, oggs are the BEST! I figure with all the background noise while commuting, 32kHz @quality "-1" work just fine. This setting results in bitrates around 32 - 48 kbps, or song files between 1 and 2 MB in size!!! No problem getting two albums onto my phone...
Re:The metric system is the tool of the devil!
on
Our Friend, The Meter
·
· Score: 1
My car gets forty rods to the hogshead and that's the way I likes it.
201.168 meters / 238.48 liters...
That's 118547.6815 l / 100km - I don't think you'd be allowed to use your car in any country using the metric system anyway. I suppose it's some sort of SUVs?;-)
I've read about this in Germany: the law (in most states of the federation, I guess), allows a university to recall a doctor's degree, if the person proves to be unworthy (regarding science) at a later time.
Faking data is not to be taken lightly - scientists rely on the quality of previous work. If several other scientists have wasted years of their time because of this, that's a lot of damage done.
I realize 250 Mtriangles/sec aren't quite the 380 stated by ATI for their current generation GPU (Radeon 9800 Pro), but the paper I linked to is from 2001.
The hardware raytracing site has a nice video of their FPGA-based system rendering about 187 million triangles at about 15 - 40 fps (512x384, 90MHz FPGA).
I'm not sure that's gonna happen. The fact of the matter is that current graphics hardware is fast approaching the point where raytracing will be irrelevant.
Actually, AFAIK the opposite is true.
Raytracers scale very nicely with geometric complexity: O(log n). So as the virtual environments continue to grow, raytracing should gain popularity over scan conversion. Have a look at this - that's 50 million triangles raytraced at 4-5 fps!
Most of the current interactive raytracing is still done on parallel computers or PC clusters, but there are a lot of optimizations that can be combined to achieve interactivity even on a single CPU. And hardware architectures are underway as well...
Slashdot Mirror
Well, I cancelled my Paypal account and gave the reason "WikiLeaks". No need to support a company that enables governments to silence free journalism, even if for the most part, the only "business" they got from me was to boast with one additional customer served...
Chorded keyboards are way more efficient than that! They should integrate a Twiddler with a cell phone. According to Prof. Thad Starner's testimonial on the website, 60 words per minute are entirely possible, and you'll get 10-30 words per minute after only a weekend of practice! I actually recall him telling me about even higher speeds, but I don't want to misquote him...
For comparison, here's some info on normal QWERTY keyboards: "Someone having minor experience with keyboards can reach 20 words per minute, an average typist reaches about 30 to 45 (usually the minimum required for dispatch positions and other typing jobs), while advanced typists work at speeds above 60." (http://en.wikipedia.org/wiki/Words_per_minute)
I, for one, welcome our new police state overlords.
- description (of phenomena)
- understanding (hypotheses about relationships)
- prediction (this requires general applicability of hypotheses)
- control (use knowledge to influence phenomena)
And the key values of science are:- Empiriscism (objective evidence!)
- Skepticism (available knowledge may always contain errors)
- Tentativeness (scientific knowledge is never the truth, but always a theory that tries to come as close as currently possible)
- Publicness (results and methodology must be published so others can replicate results, thereby establishing validity)
See how some of the basic points we're taught are also what the article tries to point out? This should be common knowledge, at least to scientists. And non-scientists that argue against science usually go against this uncertainty, as if tentativeness were something bad. "It's just a theory" should not be used in a derogatory sense. As the first post managed to (once again) get this discussion rolling down the Evolution(ism)/Intelligent Design(ism) road, here are some useful indicators of Pseudo-Science, also from my class:- Subjective
- Illogical
- Unsystematic
- Not empirically testable
- No predictive power
- Knowledge system is fixed or closed
- Anachronistic thinking
- Seeking mystery or lack of patterns
- Appeal to myths
- Casual approach to evidence
- Irrefutable hypotheses
- Explanation by scenario
- "Literary" interpretation
- Refusal to revise
To make this discussion a little more clear: "Irrefutable hypotheses" in my understanding is similar (if not equal) to a hypothesis not being "falsifiable", a term others have repeatedly used here. So basically: a theory is worthless if its hypotheses can not possibly be proven wrong by later research. This is because from such a hypothesis you can come to any conclusion. As as example to this I especially liked the 6-second universe theory stated by someone earlier in this discussion. As you can't prove him wrong, the theory is worthless to the basic goals of science.The DRM I'm willing to accept os the DRM that I won't even notice. Like the one on the iTMS seem to be.
Not even iTMS is unnoticable. I found that out yesterday, after I sent my Mac to be repaired. I copied my entire system to another (nearly identical) Powerbook, so I thought I wouldn't run into problems while my machine is gone...
Not only can I no longer play my purchased songs (of course I could authorize my replacement machine), but I just realized that since the motherboard is being replaced on my machine, I've probably permanently wasted one of my 5 allowed "authorized" machines, cause when I get it back, it won't be the "same" machine to iTunes!
Good thing I only made a small test-run with iTMS and purchased 4 songs. That doesn't hurt too much. But others will run into larger problems, that's for sure.
To me, DRM will only be acceptable as a passive measure. Indestructible watermarks are ok, if they don't impede my use of the purchased material. If files appear on the net and the company holding the rights immediately knows who leaked them and comes after that person, that's only fair. Also, that way, it will be far easier to measure the amount of damage inflicted by an individual file sharer.
Sounds more left than right to me.
I just love it when people assume politics to be one-dimensional...
...WMV files aren't Windows only.
I know. I use a Mac. And the Windows Media Player sucks. Big time. I mean, I can't even move to different parts of the video in most files. And file errors that aren't even visible on Windows will make the Mac version stutter.
You are also missing a point: this is not a "Not-Only-On-Windows" discussion. It's about formats that are likely to be well supported on the majority of multimedia-capable operating systems.
Many people are saying that "quicktime" runs on Linux. They mean some quicktime files, not all, right? I.E., primitive and uncompressed files run on linux; serious compression Quick Time files can not legally be run on Linux.
.DLLs. This obviously means you can only use QuickTime on Intel compatible Linux machines, as the libraries are in binary format.
:-(
They can play complex formats, using a hack that requires Windows
So that's not really supporting Linux in its entirety... at all.
Containers combine encoded audio and video, and possibly metadata. This usually means interleaving audio and video according to their time in the movie, so during playback your disk doesn't die from constant seeking between the audio and video portions.
Codecs are used to compress the raw audio and video to the desired size, usually reducing the quality (lossy compression).
As a container format, you mainly have the following options:
- .AVI (AudioVideoInterleaved): a really old format that just interlaces audio and video data (even mp3 audio is basically hacked into working with this - badbadbad)
- .WMV/.ASF: Microsoft stuff. Don't use, if you want compatibility with anything but Windows.
- QuickTime
.MOV (MooV actually): Apple stuff. Officially supported on Macs and Windows, but still proprietary - you're not being nice to OSS users.
- RealMedia
.RM: proprietary (see QuickTime)
- MPEG-4: New standard by the people who brought us MPEG-1 (crappy low-res by todays standards) and MPEG-2 (DVD video). It's based on the QuickTime container, but it's a public standard (not proprietary). Costs developers to get a license, though.
- Ogg: Open/Free container format. Great for OSS people, but less known than MPEG-4.
I'd recommend looking into using MPEG-4 or Ogg containers.For video compression, whether you use MPEG-4 or Ogg, go with XVID. Theora is still in development, and everything else is a mess by comparison. (flaming ensues
For audio compression, with MPEG-4 you will want to use AAC or MP3 (not sure about the latter), with Ogg containers go with Ogg Vorbis (best quality at low bitrates, IMHO) or MP3.
By sticking to a standard, but non-proprietary combination, such as MPEG-4/XVID/AAC, you might even be able to cater to all platforms without maintaining multiple formats...
It simply works on badly compressed jpeg files.
Such JPEGs should not exist. The Huffman coding step should remove most of the redundancy, thereby eliminating the possibility of "bad" compression.
If you increase the compression level, what happens is: you loose more information. But the resulting jpeg should still be mostly incompressible, because after this quantization step, further steps (Huffman etc.) ensure that the remaining data is mostly non-redundant. It's kinda like JPEG uses its own compressor by default, and a pretty good one at that.
So this new StuffIt technique really IS quite impressive.
Argh. Yeah, and they keep repeating that mistake, too - really bothersome in an otherwise well-written article.
Apple never mentioned AC-2 anywhere. In fact, they usually expand AAC to Advanced Audio Coding in their explanations. I wonder where the author got that wrong idea.
Let's see how long it takes for this myth to spread across the internet...
<eyes watering>
Ah, memories... the Tandy 1400 LT was my first home computer. I must have been around 9 at the time. Before that, I always played on friends' computers (C64, Atari, Neckermann etc.) or on the Siemens and VAX systems at my father's lab.
I remember getting to know DOS the hard way, because nobody told me to use "dir" to look for "*.exe" files (once I knew, I executed just about anything I could find...)
Later, I spent weeks "optimizing" boot time by using a RAM disk and playing LHX Attack Chopper on its CGA display. Luckily, I didn't go blind.
Today, its mostly Macs, my Sun and my NextStation running OS X, Linux, OpenBSD and NeXTStep... MS-DOS 6.0 was bloated after all, so I never checked with Microsoft again.
Symptoms:
[...]
* Clouding of vision
Eeek! Good thing Steve got it removed. I don't want to imagine in what direction Apple would head without his clear vision of future computers! No, wait...
they have that. it's called a Sony Ericsson P900.
That's what I was thinking. I'd like to own an Apple phone as I'm expecting it to be easier to use, better looking, well integrated with iTunes and nearly bug-free.
But right now, I've got a Sony Ericsson P800 with a 64 MB Memory Stick. It plays OGG Vorbis using OggPlay, and I can convert any song or entire playlists to OGG directly from iTunes using an AppleScript I wrote for that purpose. It internally uses mplayer and oggenc for conversion.
For mobile use, oggs are the BEST! I figure with all the background noise while commuting, 32kHz @quality "-1" work just fine. This setting results in bitrates around 32 - 48 kbps, or song files between 1 and 2 MB in size!!! No problem getting two albums onto my phone...
My car gets forty rods to the hogshead and that's the way I likes it.
;-)
201.168 meters / 238.48 liters...
That's 118547.6815 l / 100km - I don't think you'd be allowed to use your car in any country using the metric system anyway. I suppose it's some sort of SUVs?
Indeed? I should know... ;-)
I've read about this in Germany: the law (in most states of the federation, I guess), allows a university to recall a doctor's degree, if the person proves to be unworthy (regarding science) at a later time.
Faking data is not to be taken lightly - scientists rely on the quality of previous work. If several other scientists have wasted years of their time because of this, that's a lot of damage done.
Some links: The article in german and Google's attempt at translating it
Shark still looks fake... ;-)
I realize 250 Mtriangles/sec aren't quite the 380 stated by ATI for their current generation GPU (Radeon 9800 Pro), but the paper I linked to is from 2001.
The hardware raytracing site has a nice video of their FPGA-based system rendering about 187 million triangles at about 15 - 40 fps (512x384, 90MHz FPGA).
I'm not sure that's gonna happen. The fact of the matter is that current graphics hardware is fast approaching the point where raytracing will be irrelevant.
Actually, AFAIK the opposite is true.
Raytracers scale very nicely with geometric complexity: O(log n). So as the virtual environments continue to grow, raytracing should gain popularity over scan conversion. Have a look at this - that's 50 million triangles raytraced at 4-5 fps!
Most of the current interactive raytracing is still done on parallel computers or PC clusters, but there are a lot of optimizations that can be combined to achieve interactivity even on a single CPU. And hardware architectures are underway as well...