There is a difference between the x86 architecture (actual silicon), and the x86 ISA (more like an API).
The actually technology behind x86 processors has moved one, and updated. Transmeta is, prehaps, the most extreme example of this, but all modern processors use microcode to 'emulate' x86 (Or, at least, the least commonly used instructions).
So, what has happened is that the 'good' instructions have got faster, but the old cruft, whilst it still works, is slow. Like, I could probably write out some Z80 machine code, and expect it to work on a PIII. But not to maximum efficency.
Out of the four local libraries at home that I go to regularly (hey, I read a lot:), three of them have a collection of software CDROMS. And also Audio CD's.
You mean that for once, the UK is actually ahead in something?
*falls over in shock*
Re:Could there be GPL violations here? - No
on
Gnutella Vs. SPAM
·
· Score: 1
Firstly, you are assuming that they modified an existing client, and did not write thier own. Whilst it is straightforward to write one, I agress that they probably just modified an existing (GPLd) client.
However, doing so does _not_ require them to release thier source. All it does is require them to make the source available to anyone that that distribute the binary to.
I really don't think that most spammers are up there with the GPL and all the legal issues, so all the need to do is sell the binary, and hope no one asks for the source.
Even if someone does get the source, then how likely are they to distribute it, being a spammer dumb enough to by this? I just don't think that it's going to be spread widely.
This is not a 'loophole' in the GPL, nor is it a fixable problem.
The MP3 format is not the most effective for all audio data.
If your really making a point, you don't need music. Music is what MP3 was designed for.
Speech can be encoded in a simple u-law format. This is basically a slightly modified WAV file (requires virtually no processing), at 64 kbps. This is telephone standard, and is barely considered a form of compression.
ADPCM (Adaptive differenial pulse code modulation) gets to 32 kbps, by exploting sample to sample differences.
You can use a CELP (Code Exicted Linear Prediction) algorithm, to get easily recognisable speech in 4800 bps (that's 4.8 kbps).
If you push it, and have plenty of decode time available, some clever acoustic vector prediction, and an LPC algorithm can get you as low as 300 bps, although prehaps 1200 bps is more reasonable.
This extreme compression will also distrocrt the voice, making it more difficult to identify. This is hardly a problem.
At 4.8 kbps, you can get 20 minutes of speech in 100k. Compare that to MP3.
Napster's buycott list is the wrong list to be using. It represents _bands_ that support Napster, not record labels. If you look carefully, you might see something interesting:
Ben Folds Five, one of the bands on Napsters Buycott, is on epic records. This is owned by Sony, a member ofthe RIAA
Remeber that Naspters Buycott is designed to _support_ the RIAA, not send money away from it.
It is not a good list to use if you wish to boycott the RIAA.
RF shielding basically consists of a Farady cage around the compnent. To do this in a transparent case will require the use of a transparent conductive material.
These exist.
For example, the most common in indium tin oxide, although alternatives exists, such as indium gallium nitrate
In this particular case, the requirement for the anneal post-deposition for indium tin oxide could be a problem, assuming the case is made of a polymer. Additionally, you'd need an outer layer, the conductive Faraday cage, and then an inner layer, to protect the conductive layer.
New TLD's will be useless, until, and unless, ICANN enforce some restirctions on who gets what domain names.
It started with how the domain names are adiminsted, (ie, for profit), which ment thaqt it was advantageous for the administrater (NSI) to sell as many as possable. This _must_ change, else the whole squatting, and registering of multiple domains ("Click here to register.net and.org too!", anyone?) will contiue.
ICANN, get some real rules on who gets what in there, else all you've done is up the registrars profits.
Step 1 is pretty easy... in fact you probably already have Napster installed. The laying of eggs will work best if you can install Napster on multiple machines so you have the best chance of letting many users connect.
Uh, right. So the same people who feel strongly enough to make these cukoo's eggs, are the same people who regularly use Napster?
Somehow I doubt it.
And advocating that people break Naspters terms of service (One person, one account, IIRC) is hardly the best way to go about it.
If Nullsoft, as the people who own the copyright to Winamp had realesed Winamp as GPL'd, they would not have to relase any such changes they might make.
The person who owns the copyright on some software can realses that under any liscense they choose. If they were to add proprietry code to Winamp to deal with the copyright issues, then they could easily realse a binary only version, with that support.
Other authors could not, because they would be bound by the GPL, but the owners could, by makeing a non-GPL'd realses of that code.
Remeber that the GPL restricts what other people can do with your code. It does not (and cannot?) restrict what _you_ can do with your own code (cf sendmail and sendmail professional).
It is possable to alter genes at any time. Doing it on a single cell is much easier than on many millions of cells, so pre-natal genetic modification is preffered, from that point of view.
However, there are other aspects too. If you take a fully grown man, and change his genes so that he would grow to 8 foot tall, his height isjust not goin to change. This is because he has already passed the stage where such gene would be used. It's the same for eye colour - your eye are already grown, so they won't chage, even if you change the genes.
However, altering a 5 year old childs gene to make him grow to 8 foot tall would have an effect (although possibly not the _full_ effect), as he has most of his growing still to do. This assumes that you can get the genes to the appropriate cells suitable.
This is a slightly hand-wavy explanation, but I believe it gets the point across.
In summary, gene therepay can only change how you grow, not how you have grown.
For example, in Cystic Fibrosis, the lungs produce mucus that is too sticky. Gene therapy can help, as it can tell the lungs to make mucus with the correct level of stickyness. After a shot time, the mucus in the lungs will be replaced with the corrected mucus, fixing that problem.
With Downs syndrome, however, the brain develops abnormally. Correcting the genetic fault will not fix the symtoms, because the brain is fully grown, and thus not fixable in this way (other solutions, howwever, are possable).
They actualy use scalpels with near monomolecular edged for surgey on eyes (removing catartacts etc), because the scars resulting from those scalpals are so fine they effectivly don't exist.
There is only one problem with these, (from a sword making point of view), and that's the fact that they are made from glasses. Normally, silicate glass, cos it's cheap and hard.
The fine edge results from the way in which the glass fractures, so as you 'chip' small piece off the edge, you form conchcoidial fractures. Repeat ofen enough, and you can get virtually 'monomolecular' [0] edges. If I recall correctly, the Aztec's were aware of this, and made knives from obsidian (volcanic glass) because they were very sharp.
Problem is toughness [1], but there are some somultions to that, such as making them into a composite.
[0] It's a glass. Molecule is an alien term to glasses (and crystals). However, if you take the term 'monomolecular' to mean about 10 atoms across, you can get that sharp with glasses.
[2] Toughness, in a material science context, means essentially fracture resistance. In other words, if you hit it with a hammer, and it doesn't break [3], it's tough, otherwise it's brittle [4].
[3] Bending is allowed. Hence metals are 'tough'
[4] Belive it or not, this is a definition that is actually used (ie Charpy impact test).
First of all, they do have names: Witness potato etc (Okay, that's Debian, I dunno, I don't use distros).
Secondly, with software, the name implies compatability. If you upgrade to the next version, of the same software, you expct it to do the same things, if a little better. Changing the name would be counter intuative.
With (Open Source) software, it is often possable to get bleeding edge downloads (thank you CVS), where a fixed version number implies some form of stability. With hardware, it is fixed, and there is no need to differentate between development and release versions - if the public can get it, it's a release version.
The real beef is not that the name is dull, but that there used to be a proper versioning system (8086, 80286, 80386), albeit one where the version number was in the middle of the 'name' [0], due to historical accident.
Now, we have a name that was ment to replace one incarnation of the 'naming' system that has been stretched to cover multiple versions, to the point where it is now useless to identify each system.
Actually, the speed of theram is less of an issue these days.
Consider the typcial desktop PC:
It will have: an L1 cache right next ot the processor - fastest ram.
L2 Cache somewhere very close, bigger, not quite as fast, but slightly cheaper.
DRAM's: Slower still, but still pretty fast, relativly cheap
Hard drive (Virtual memory): Slowest (realistacally, yes, you _could_ use a floppy as virtual), cheapest.
All this would do would be to give another layer inbetween the main DRAM's and the hard drive. With an appropriate caching and paging algorithm, this could be made _almost_ as useful as ordinary DRAM on a MB for MB ratio, provided only some of it is compressed.
For example, you might have the kernel store a process that is used once every second in the compressed ram, rather than the main DRAM. And so on.
Uh, did you even attempt to read this article? It doesn't appear so, because this is a _class action_ lawsuit. In other words, it is brought by the citizens of your country, not your government.
Why is it there is a "Panic, this is one step closer to 1984" post in every thread these days?
The impression I get this side of the atlantic is that you'dbe better worrying about your corporations, rather than your government.
Sorry if this is a litt rant-y, but every time I see people reffer to the government, it irks me a little. Paranoid scare stories also, and the combination gets silly.
range. Now take the original, subtract the MP3 output and you get the error. Now encode the error using a different format. You now have a (still lossy) compressed version of the sound with less error than the MP3
It's a nice idea, but...
With a psycho acoustic lossy format, such as MP3 (Or Ogg Vorbis), taking the diff, and encoding that is pretty pointless. You've gone to all the trouble of working out what part of the signal you can throw away, as part of the psych acoustic compression, and then you just encode it all back in again?
The _only_ time I can see that being useful is for streaming applications, where, when the data rate drops low, you stop sending the diff, and automatically drop to a lower quality. However, that implies you can get a bandwidth of the order of CD rates. Hardly mass market.
On the point of encoders, if they've added anything to the MP3 file, then it's either an improved encoder engine (compare LAME with early MP3 encoders), or will require a new pair of encoder / decoder. So much for still being MP3.
I've always thought the expression "CD quality" was pretty stupid. Is that equivalent sound quality to a 1992 Saisho CD walkman or 12,000ukp worth of Linn CD12?
The latter one, if you can tell the difference.
When people say "CD quality", they are reffering to the full 22.05 khz frequency response, and the 96 DB dynamic range. A 'proper' CD player will output that. However, due to problems with the digitisation (aliasing etc), most CD's do not use the full range available, and top out at 20 khz (because it's a _lot_ cheaper). Thus a cheaper CD player may not bother doing it all properly, or use crappy analoge amps for thr final stage, because no one can tell.
Also, most CD pressing plants do _not_ press CD's to be good, they press them to be cheap. This means that the error rate on the disk is pushed to the maximum, before people complain, because that means faster pressing, which is cheaper.
Generally, classical CD's are pressed better, because you get people with better ears listening to them, who can tell the difference between partial interpolation, and real sound. [this is one reason the classical CD's are more expensive - they do actually cost (slightly) more to produce].
Acording to this article on Yahoo (but from ZDnet), a poster on Mac Central (although I can't find this actual post), Intel sponsors Harvards graduation (quote from forum posting):
"My wife works at the Harvard Science Center in the media-services department (which is 90 percent Mac-based). She was witness to the harassment and rude attitude of the Intel representatives. They were childishly demanding that the iMacs be removed from the premises or they would have to be covered up.
"Harvard argued that they were in use and wouldn't be removed. This escalated into a threat of Intel not sponsoring Harvard's graduation week (and therby removing their [monetary] contribution). Harvard finally agreed, and Intel also covered up the signs and windows to the nearby Mac multimedia lab. Absolutely pitiful."
The main problem with fractal compression is not patent problems.
It's that it doesn't work.
That is to say, it doesn;t work usefully.
To explain: in order to get efficent compression, you need to use something called an IFS (Iterated Function System). These can be used to represent an image, but only if the IFS used is sufficently "similar" to the target image. And there in lies the problem. IFS's are good for generating natural looking images, but generating an IFS from an image is an inverse process. And has not been solved.
State of the are is to use the graduate student theorm to produce an IFS for your image, but this is expensive:
Take a gradutate student, give him a graphics workstation, lock him in a room, and don't let him out until he's come up with an IFS for your image.
This take 100 hours, of input from a grad student, in order to come up with an IFS. Not exactly mass market tech.
There was discovered a way (by one of said gradstudents) to automatically come up with an IFS that is "quite similar" to the image, called Partitioned Integreadted Function Systems (PIFS). Unfortunatle, they're not proper IFS, and the 10 000:1 compression went out of the window, and dropped to 50:1 or so. At 50:1 compression, with patent issues, and a slow algorithm, it just wasn't worth it.
Aside: The technology is patented under US patent 4,941,193. I'd link on IBM, but it's down. Partitioned IFS are US patent 5,065,447. More data in comp.compression FAQ here.
Well, as a gradutate in physics, and chemistry, and currently doing an MSc (one year postgrad course) in materials science, I'm not intending to become a 'guru', in the conventional sense.
I have a strong interest in programming, but no formal training (and not likely to get any), but rather than wanting to write the next killer app, or make improvments to an OS, I use programming to do mathematics.
Most of the code I've written so far has been mathematically orientated stuff, for a specific purpose.
That's not to say I don't use other software, just that it's not the stuff I write. For example, next year, I'll be writing Fortran to run on supercomputers, to calculate electronic band stuctures of solid state materials. Right now, I'm writing code to similate what happens when you shoot an electron beam into a polycrystaline semiconductor.
I suppose, therefore, you could say that I am algorithm programmer.
I reckon it's a often overlooked area. Basically, that's because it's seen more difficult, purely because you need knowledg in more than one field - computing being only one, mathematics being another, and possasbly also a third (if, say, your working with speech, or physical simulation) [see, for example, the bible on algorithms, "The Art of Computer Programmming", by D.E. Knuth, a mathematician.]
It's not as sexy as GUI or OS design (people don't see your stuff, and it's more specialised), but it's just as tricky. Working with algorithms does tend to have fewer logical bugs, as a full formal design is (practically) required, and proven to be mathematically correct.
Actually, for that reason, I'd reccomend people to look at algorithmic work, particularly if you have a strong mathematical bent.
The source code, and all the other IP of Microsoft is just that - property of Microsoft (and shareholders).
To distribute the source code around, the govenment would have to compensate Microsoft for the loss of (legitimate) earnings that would ensue.
In other words, the Govenment would have to buy the code off Microsoft. This is one of the laws desgined to stop overly heavy handed govenment intervention.
Oh, and forking the code would result in multiple, incompatable version of Windows. Just think of all those bugs (sorry features), would be changed.
<ol> <li>Start from <a href="www.microsoft.com">Microsoft</a>, then search for "samba" <li>On the Web workshop page, go to the bottom of the links <li>Last link gets you redirected to <a href="www.littleigloo.com">Littleigloo</a> <li>Click on thier 'featured site' - <a href="http://www.linuxdev.net>Linux Dev</a> <li>Second link on Linux Dev main page is (cutrrently) Slashdot.
I make that 5. Anyone want to try and beat that?
(Yes, I start in the same way. Isn't that part of the point of free software, learn from others, and improve?:)
Slashdot Mirror
Not really.
There is a difference between the x86 architecture (actual silicon), and the x86 ISA (more like an API).
The actually technology behind x86 processors has moved one, and updated. Transmeta is, prehaps, the most extreme example of this, but all modern processors use microcode to 'emulate' x86 (Or, at least, the least commonly used instructions).
So, what has happened is that the 'good' instructions have got faster, but the old cruft, whilst it still works, is slow. Like, I could probably write out some Z80 machine code, and expect it to work on a PIII. But not to maximum efficency.
This is the curse of bakwards compatability.
Ars Technica have a review of this here.
A new architecture could probably do many things better. But would it be sufficently better to make the cost of getting _everything_ rewritten?
Market forces suggest not. (See the Alpha - newer architecture, but not exactly everywhere).
Yep this is relativly common.
:), three of them have a collection of software CDROMS. And also Audio CD's.
Out of the four local libraries at home that I go to regularly (hey, I read a lot
You mean that for once, the UK is actually ahead in something?
*falls over in shock*
Firstly, you are assuming that they modified an existing client, and did not write thier own. Whilst it is straightforward to write one, I agress that they probably just modified an existing (GPLd) client.
However, doing so does _not_ require them to release thier source. All it does is require them to make the source available to anyone that that distribute the binary to.
I really don't think that most spammers are up there with the GPL and all the legal issues, so all the need to do is sell the binary, and hope no one asks for the source.
Even if someone does get the source, then how likely are they to distribute it, being a spammer dumb enough to by this? I just don't think that it's going to be spread widely.
This is not a 'loophole' in the GPL, nor is it a fixable problem.
(Not currently a lawyer)
The MP3 format is not the most effective for all audio data.
If your really making a point, you don't need music. Music is what MP3 was designed for.
Speech can be encoded in a simple u-law format. This is basically a slightly modified WAV file (requires virtually no processing), at 64 kbps. This is telephone standard, and is barely considered a form of compression.
ADPCM (Adaptive differenial pulse code modulation) gets to 32 kbps, by exploting sample to sample differences.
You can use a CELP (Code Exicted Linear Prediction) algorithm, to get easily recognisable speech in 4800 bps (that's 4.8 kbps).
If you push it, and have plenty of decode time available, some clever acoustic vector prediction, and an LPC algorithm can get you as low as 300 bps, although prehaps 1200 bps is more reasonable.
This extreme compression will also distrocrt the voice, making it more difficult to identify. This is hardly a problem.
At 4.8 kbps, you can get 20 minutes of speech in 100k. Compare that to MP3.
see comp.speech FAQ for more data.
Napster's buycott list is the wrong list to be using. It represents _bands_ that support Napster, not record labels. If you look carefully, you might see something interesting:
Ben Folds Five, one of the bands on Napsters Buycott, is on epic records. This is owned by Sony, a member ofthe RIAA
Remeber that Naspters Buycott is designed to _support_ the RIAA, not send money away from it.
It is not a good list to use if you wish to boycott the RIAA.
is perfectly possable.
RF shielding basically consists of a Farady cage around the compnent. To do this in a transparent case will require the use of a transparent conductive material.
These exist.
For example, the most common in indium tin oxide, although alternatives exists, such as indium gallium nitrate
In this particular case, the requirement for the anneal post-deposition for indium tin oxide could be a problem, assuming the case is made of a polymer. Additionally, you'd need an outer layer, the conductive Faraday cage, and then an inner layer, to protect the conductive layer.
Still, it is 100% possable. And expensive.
New TLD's will be useless, until, and unless, ICANN enforce some restirctions on who gets what domain names.
.net and .org too!", anyone?) will contiue.
It started with how the domain names are adiminsted, (ie, for profit), which ment thaqt it was advantageous for the administrater (NSI) to sell as many as possable. This _must_ change, else the whole squatting, and registering of multiple domains ("Click here to register
ICANN, get some real rules on who gets what in there, else all you've done is up the registrars profits.
Unless, of course, that _was_ the whole idea?
Inside reports "Who's The Boss? Springsteen Imposter Scams Napster", and includes reffernce to the Cockoo thingy.
ZDNet has a comment ary
Enjoy.
How To Lay Cuckoo's Eggs
...
1.Download and install Napster
...
Step 1 is pretty easy... in fact you probably already have Napster installed. The laying of eggs will work best if you can install Napster on multiple machines so you
have the best chance of letting many users connect.
Uh, right. So the same people who feel strongly enough to make these cukoo's eggs, are the same people who regularly use Napster?
Somehow I doubt it.
And advocating that people break Naspters terms of service (One person, one account, IIRC) is hardly the best way to go about it.
If Nullsoft, as the people who own the copyright to Winamp had realesed Winamp as GPL'd, they would not have to relase any such changes they might make.
The person who owns the copyright on some software can realses that under any liscense they choose. If they were to add proprietry code to Winamp to deal with the copyright issues, then they could easily realse a binary only version, with that support.
Other authors could not, because they would be bound by the GPL, but the owners could, by makeing a non-GPL'd realses of that code.
Remeber that the GPL restricts what other people can do with your code. It does not (and cannot?) restrict what _you_ can do with your own code (cf sendmail and sendmail professional).
It is possable to alter genes at any time. Doing it on a single cell is much easier than on many millions of cells, so pre-natal genetic modification is preffered, from that point of view.
However, there are other aspects too. If you take a fully grown man, and change his genes so that he would grow to 8 foot tall, his height isjust not goin to change. This is because he has already passed the stage where such gene would be used. It's the same for eye colour - your eye are already grown, so they won't chage, even if you change the genes.
However, altering a 5 year old childs gene to make him grow to 8 foot tall would have an effect (although possibly not the _full_ effect), as he has most of his growing still to do. This assumes that you can get the genes to the appropriate cells suitable.
This is a slightly hand-wavy explanation, but I believe it gets the point across.
In summary, gene therepay can only change how you grow, not how you have grown.
For example, in Cystic Fibrosis, the lungs produce mucus that is too sticky. Gene therapy can help, as it can tell the lungs to make mucus with the correct level of stickyness. After a shot time, the mucus in the lungs will be replaced with the corrected mucus, fixing that problem.
With Downs syndrome, however, the brain develops abnormally. Correcting the genetic fault will not fix the symtoms, because the brain is fully grown, and thus not fixable in this way (other solutions, howwever, are possable).
... and I can make them.
Mind you, it's not hard.
They actualy use scalpels with near monomolecular edged for surgey on eyes (removing catartacts etc), because the scars resulting from those scalpals are so fine they effectivly don't exist.
There is only one problem with these, (from a sword making point of view), and that's the fact that they are made from glasses. Normally, silicate glass, cos it's cheap and hard.
The fine edge results from the way in which the glass fractures, so as you 'chip' small piece off the edge, you form conchcoidial fractures. Repeat ofen enough, and you can get virtually 'monomolecular' [0] edges. If I recall correctly, the Aztec's were aware of this, and made knives from obsidian (volcanic glass) because they were very sharp.
Problem is toughness [1], but there are some somultions to that, such as making them into a composite.
[0] It's a glass. Molecule is an alien term to glasses (and crystals). However, if you take the term 'monomolecular' to mean about 10 atoms across, you can get that sharp with glasses.
[2] Toughness, in a material science context, means essentially fracture resistance. In other words, if you hit it with a hammer, and it doesn't break [3], it's tough, otherwise it's brittle [4].
[3] Bending is allowed. Hence metals are 'tough'
[4] Belive it or not, this is a definition that is actually used (ie Charpy impact test).
Software versioning is subtly different.
.eq. unique identifier.
First of all, they do have names: Witness potato etc (Okay, that's Debian, I dunno, I don't use distros).
Secondly, with software, the name implies compatability. If you upgrade to the next version, of the same software, you expct it to do the same things, if a little better. Changing the name would be counter intuative.
With (Open Source) software, it is often possable to get bleeding edge downloads (thank you CVS), where a fixed version number implies some form of stability. With hardware, it is fixed, and there is no need to differentate between development and release versions - if the public can get it, it's a release version.
The real beef is not that the name is dull, but that there used to be a proper versioning system (8086, 80286, 80386), albeit one where the version number was in the middle of the 'name' [0], due to historical accident.
Now, we have a name that was ment to replace one incarnation of the 'naming' system that has been stretched to cover multiple versions, to the point where it is now useless to identify each system.
And that is why I think it's stupid.
[0] In this context name
Actually, the speed of theram is less of an issue these days.
Consider the typcial desktop PC:
It will have: an L1 cache right next ot the processor - fastest ram.
L2 Cache somewhere very close, bigger, not quite as fast, but slightly cheaper.
DRAM's: Slower still, but still pretty fast, relativly cheap
Hard drive (Virtual memory): Slowest (realistacally, yes, you _could_ use a floppy as virtual), cheapest.
All this would do would be to give another layer inbetween the main DRAM's and the hard drive. With an appropriate caching and paging algorithm, this could be made _almost_ as useful as ordinary DRAM on a MB for MB ratio, provided only some of it is compressed.
For example, you might have the kernel store a process that is used once every second in the compressed ram, rather than the main DRAM. And so on.
Uh, did you even attempt to read this article? It doesn't appear so, because this is a _class action_ lawsuit. In other words, it is brought by the citizens of your country, not your government.
Why is it there is a "Panic, this is one step closer to 1984" post in every thread these days?
The impression I get this side of the atlantic is that you'dbe better worrying about your corporations, rather than your government.
Sorry if this is a litt rant-y, but every time I see people reffer to the government, it irks me a little. Paranoid scare stories also, and the combination gets silly.
range. Now take the original, subtract the MP3 output and you get the error. Now encode the error using a different format. You now have a (still lossy) compressed version of the sound with less error than the MP3
It's a nice idea, but...
With a psycho acoustic lossy format, such as MP3 (Or Ogg Vorbis), taking the diff, and encoding that is pretty pointless. You've gone to all the trouble of working out what part of the signal you can throw away, as part of the psych acoustic compression, and then you just encode it all back in again?
The _only_ time I can see that being useful is for streaming applications, where, when the data rate drops low, you stop sending the diff, and automatically drop to a lower quality. However, that implies you can get a bandwidth of the order of CD rates. Hardly mass market.
On the point of encoders, if they've added anything to the MP3 file, then it's either an improved encoder engine (compare LAME with early MP3 encoders), or will require a new pair of encoder / decoder. So much for still being MP3.
I've always thought the expression "CD quality" was pretty stupid. Is that equivalent sound quality to a 1992 Saisho CD walkman or 12,000ukp worth of Linn CD12?
The latter one, if you can tell the difference.
When people say "CD quality", they are reffering to the full 22.05 khz frequency response, and the 96 DB dynamic range. A 'proper' CD player will output that. However, due to problems with the digitisation (aliasing etc), most CD's do not use the full range available, and top out at 20 khz (because it's a _lot_ cheaper). Thus a cheaper CD player may not bother doing it all properly, or use crappy analoge amps for thr final stage, because no one can tell.
Also, most CD pressing plants do _not_ press CD's to be good, they press them to be cheap. This means that the error rate on the disk is pushed to the maximum, before people complain, because that means faster pressing, which is cheaper.
Generally, classical CD's are pressed better, because you get people with better ears listening to them, who can tell the difference between partial interpolation, and real sound. [this is one reason the classical CD's are more expensive - they do actually cost (slightly) more to produce].
The way the DNS system works currently makes that not really an option (without compleatly re-working it).
.com's are.
.coms where microsoft.com's DNS is.
Currently, if you look for www.microsoft.com, a request is sent to your local DNS, and asks it if it knows the number for that name.
Assuming it doesn't, then the DNS ask's one of the root servers where the central list of all
It then asks the central list of all
Then is asks microsoft.com's DNS for the number for www.microsoft.com
It's a hierachical structure.
So, no, it's not technologicaly straightfrowerd to remove TLD's a the moment.
Acording to this article on Yahoo (but from ZDnet), a poster on Mac Central (although I can't find this actual post), Intel sponsors Harvards graduation (quote from forum posting):
"My wife works at the Harvard Science Center in the media-services department (which is 90 percent Mac-based). She was witness to the harassment and rude attitude of the Intel representatives. They were childishly demanding that the iMacs be removed from the premises or they would have to be covered up.
"Harvard argued that they were in use and wouldn't be removed. This escalated into a threat of Intel not sponsoring Harvard's graduation week (and therby removing their [monetary] contribution). Harvard finally agreed, and Intel also covered up the signs and windows to the nearby Mac multimedia lab. Absolutely pitiful."
Nuff said.
The main problem with fractal compression is not patent problems.
It's that it doesn't work.
That is to say, it doesn;t work usefully.
To explain: in order to get efficent compression, you need to use something called an IFS (Iterated Function System). These can be used to represent an image, but only if the IFS used is sufficently "similar" to the target image. And there in lies the problem. IFS's are good for generating natural looking images, but generating an IFS from an image is an inverse process. And has not been solved.
State of the are is to use the graduate student theorm to produce an IFS for your image, but this is expensive:
Take a gradutate student, give him a graphics workstation, lock him in a room, and don't let him out until he's come up with an IFS for your image.
This take 100 hours, of input from a grad student, in order to come up with an IFS. Not exactly mass market tech.
There was discovered a way (by one of said gradstudents) to automatically come up with an IFS that is "quite similar" to the image, called Partitioned Integreadted Function Systems (PIFS). Unfortunatle, they're not proper IFS, and the 10 000:1 compression went out of the window, and dropped to 50:1 or so. At 50:1 compression, with patent issues, and a slow algorithm, it just wasn't worth it.
Aside: The technology is patented under US patent 4,941,193. I'd link on IBM, but it's down. Partitioned IFS are US patent 5,065,447. More data in comp.compression FAQ here.
Well, as a gradutate in physics, and chemistry, and currently doing an MSc (one year postgrad course) in materials science, I'm not intending to become a 'guru', in the conventional sense.
I have a strong interest in programming, but no formal training (and not likely to get any), but rather than wanting to write the next killer app, or make improvments to an OS, I use programming to do mathematics.
Most of the code I've written so far has been mathematically orientated stuff, for a specific purpose.
That's not to say I don't use other software, just that it's not the stuff I write. For example, next year, I'll be writing Fortran to run on supercomputers, to calculate electronic band stuctures of solid state materials. Right now, I'm writing code to similate what happens when you shoot an electron beam into a polycrystaline semiconductor.
I suppose, therefore, you could say that I am algorithm programmer.
I reckon it's a often overlooked area. Basically, that's because it's seen more difficult, purely because you need knowledg in more than one field - computing being only one, mathematics being another, and possasbly also a third (if, say, your working with speech, or physical simulation) [see, for example, the bible on algorithms, "The Art of Computer Programmming", by D.E. Knuth, a mathematician.]
It's not as sexy as GUI or OS design (people don't see your stuff, and it's more specialised), but it's just as tricky. Working with algorithms does tend to have fewer logical bugs, as a full formal design is (practically) required, and proven to be mathematically correct.
Actually, for that reason, I'd reccomend people to look at algorithmic work, particularly if you have a strong mathematical bent.
There is a very nive DVD of the Bolshi ballet available, in the UK.
It's not region enconded. And it's a top performance too.
Windows 98
Three days without a remote hole in any install!
Uh, localhost hole?
The govenment can't do this.
The source code, and all the other IP of Microsoft is just that - property of Microsoft (and shareholders).
To distribute the source code around, the govenment would have to compensate Microsoft for the loss of (legitimate) earnings that would ensue.
In other words, the Govenment would have to buy the code off Microsoft. This is one of the laws desgined to stop overly heavy handed govenment intervention.
Oh, and forking the code would result in multiple, incompatable version of Windows. Just think of all those bugs (sorry features), would be changed.
You took eight jumps. Here's
:)
<ol>
<li>Start from <a href="www.microsoft.com">Microsoft</a>, then search for "samba"
<li>On the Web workshop page, go to the bottom of the links
<li>Last link gets you redirected to <a href="www.littleigloo.com">Littleigloo</a>
<li>Click on thier 'featured site' - <a href="http://www.linuxdev.net>Linux Dev</a>
<li>Second link on Linux Dev main page is (cutrrently) Slashdot.
I make that 5. Anyone want to try and beat that?
(Yes, I start in the same way. Isn't that part of the point of free software, learn from others, and improve?