Ah Yes, of course. microsoft is on top because there code is very interoperable, and uses non proprietory,open formats to make it easier for others to make interoperable software. And examples would be Ms version of kerberos, their JVM, their extensions to javascript, html, etc, etc.
As for modularity, they have themselves claimed in court that their code parts are so intertwined that they cannot be isolated. And of course, Your(or mine) boss cares a lot whether MS Word is modular on not, In fact, if it were not, he would stop using it!
The reason that someone is on the top is that it was chosen by IBM as the Os vendor when they launched the PC.
It all depends on the cross section of the said strangelet. If, it is the size of a quark, it will probably just pass through your body without touching anything!
The reason is of course, that an atom is 99% ( maybe more) empty space. All the mass(okay, most) is concentrated at neucleus, and so the possibility of a strangelet just passing through would be very high.
However, if a collison did occur,it would just push the colliding atom out of way and continue on its path. Not much will happen.
So my thinking is, nobody needs to bother about strangelets hitting them.
That said, I feel that todays particle physics has too many particles, and that is sign that something is wrong with our theories. Of course, this has no basis at all.
You americans think always think starwars is the greatest thing in the universe. I am reaaaally bored with this! Sigh. somehow people fighting with tubelights entertains people.
Re:JPEG 2000 looks like the right thing at last.
on
JPEG2000 Coming Soon
·
· Score: 1
JPEG 2000 will have blocking artifacts at lower bit rates, for the same reason that Jpeg has: Subdividing the image into blocks, and treating each block separately.
OK, time to debunk some common misunderstandings
1. DCT compression produces blocking artifacts.
Nope, it dosen't. JPEG standard breaks image into blocks of 8x8, and applies DCT on the blocks separately. Now these blocks are really related to each other (have similar color etc). Jpeg treats them separately. When using higher compression, a lot of information is discarded, which results in blocks looking slightly different. That's what causes blocking artifacts.
2. Wavelet compression is Significantly better
This is true only for very low bit rates. At higher bitrates DCT based methods can acheive efficiencies very close to JPEG 2000 (note that jpeg can't do it, but some newer papers have suggested methods which can).
3. Wavelet compression does not produce blocking artifacts.
Well, it doesn't, but the JPEG 2000 images will have blocking artifacts for the same reasons that JPEG has... dividing images in blocks and treating them separately.
Note that dividing images into blocks and then compressing blocks reduces compression in both DCT and Wavelet transforms . With DCT you can't compress a whole image in One shot (computation intensive), but with wavelets it was possible. I regard this as a missed oppurtunity.
So, to sum up, JPEG2000 isn't as earth shattering it is made out to be, it won't revoulutionize image compression the way jpeg did. Its just another image format. period.
5x reduction in storage space? Not really. Only advantage of Jpeg 2000 over jpeg lies in better quality at lower bit rates, and the ability to encode
ROI etc.
Compression is not a big argument in favour of Jpeg 2000.
Perhaps this is news to you, but US is the only country in the world to have ever used a nuke. It has the world's largest arsenal of nukes, and the capability to deliver it. It is one of the few countries that have a habit of nosing into others affairs, and has shown the tendency to use force at the slighest pretext. It funded the mujahedeen in Afganistan, then funded the taliban, it masscared the vietnamese, it has put a stranglehold on Iraq, leading to shortage of food and medicines. It kills afgan civilians on the sightest suspicion that they are Al-qaeda.That's why the news is dangerous. That's why the US is the greatest threat to world security.Period.
Nukes are weapons and meant to be used. Any body who attacks us must be nuked. we, and only we have the right to exist. So what if generations of the victims would be born wit h horrible genetic defects? So what if complete ecosystem would be destroyed? They dared to attack us, and paid the price. What if the nuclear fallout will cause horrible deaths to their people? So what if there will be babies born without eyes, without livers, mentally retarded? They dared to oppose us, the ones god designated to be above all, to be the most powerful, to rule.
Your military education has completely indoctrinated you to such a task. Remember, you were trained to drop Nukes, and you would have to do that without feeling any remorse at the killing of millions. You will have to be trusted, and therefore were indoctrinated.
Your enemy sir, is also human. Treat him like a human. Kill him maybe, destroy his army, but let his young sons live.Don't use nukes to cover up your cowardice.
They proposed the CTBT as an instrument to prevent nuclear have nots from becoming haves. For, in world populated with nuclear weapons, superiority in conventional weapons means nothing. Having a nuke does not constitute an advantage unless the other guy does not have it.
And they berated India & Pakistan for seeing throught this strategem and not signing the treaty. Abd you know what? they themselves haven't signed it.
US posseses the largest nuclear arsenal in the world, and is the only country to have dropped one. Unless the US can drop nukes on others, and escape retaliation, they are of no use. Hence the missile defense program.
Well, In India, we can buy CD-R's for Rs 20, which comes to about $0.45, (and if I am not wrong to 50 cents) They are not the best quality(usually locally made), but you get the idea.
Re:I hope MPEG-4 fails
on
More on MPEG4
·
· Score: 1
It's wrong to think of MPEG-4 as "a single standard" in the sense that any implementation needs to handle all of the spec.
This is known. But the point is, all those profiles aren't necessary at all. If what people want is a video codec with higher compression, then that's what you do. Trying to be all for everything is not only futile, but is also stupid to do. MPEG 4 reminds me of OSI protocol standards. Both have unecessary stuff, and both were formulated before their needs were felt. And I predict that MPEG 4 will share the same fate.
As for open source providing a those codecs, they probably will happen, eventually. If that's all you need, then you don't need MPEG-4.
Unfortunately, that's the truth: That's what people need. Obvoiusly nobody is interesed in having an interactive scene were you can move things around, these are applications which dreamt up by the MPEG 4 comittee, and nobody wants them, further, AFAIK, nobody has shown any interest in them. I work in a company known for its expertise in multimedia. And believe me, nobdy has ever inquired with us for any solutions other than simple profile@level 3
But is that open-source codec going to work with new mobile phones from different vendors, over lossy networks?
Implementation over lossy networks is no big deal at all: essentially what you do is embed some resync markers in your bit stream. So I don't think that's going to be too difficult. Yes, mpeg 4 has reversible VLC, too, but that's not too difficult to do, is it?
Re:I hope MPEG-4 fails
on
More on MPEG4
·
· Score: 1
As someone who works with JPEG 2000 I have to call you on your misinformation.
Well, As somebody who knows both Jpg (earlier version) and Jpeg 2000 I stand by my statements. Consider: Divison of an image into components, division of components into tiles, then subbands, preccints, layers. All this complexity was not there in JPG (obviously when I said JPEG 2000 was complex I was not comparing it to MPEG). Then you have ROI coding etc. Then for transmitting the image you have 4 combinations possible. Then you have those passes, etc.This is entirely unecessary. Only advantage of wavelet transforms over DCT transform comes at low bitrates. Instead of using it straightforwardly, the standards committee went ahead and added useless frills.
JPEG 2000 uses wavelets.
I remember mentioning that.
As for reading the standard, one can even read the MPEG 4 standard, if one has enough time. I myself had read all the stuff required for core profile.
(for video only!)
Re:Yet Another Useless Initiative
on
More on MPEG4
·
· Score: 2, Informative
We've seen this with JPG, where as a result a lot of websites are switching to PGN, and now we will see this again with MPEG4.
You probably meant GIF (instead of JPG).
Re:I hope MPEG-4 fails
on
More on MPEG4
·
· Score: 2, Redundant
MPEG-4 is a complete mess.
Completely true. It is too complex, too big, has too many options. It tries to be everything to everybody, and fails. It has dozens of profiles and levels, and except audio-video compression, all others are completely useless.
It has profiles for transmission of facial data, 3D data, shape decoding, Structured Audio etc. All of them are extremely lowlevel and crude specially facial data transmission and 3D data transmission. To say that they will be of use in 3D games & interactive applications is a pipedream. Basically, MPEG 4 is far too complex for these things, and is definitely not enough.
Further, the advances made in MPEG 4 are incremental only, and it should be not too difficult to acheive similar effects by different means.If you look at the H.263 standard, you'd know that much of Mpeg 4 is simply H.263 in disguise (specially the video part), with some frills added.
Standard committes in recent times have gone completely overboard and created mammoth, unusable standards. MPEG 4, and more recently JPEG 2000 are a case in point. JPEG 2000 uses 4 times the CPU time of JPEG!. And this is despite the fact that wavelet is a O(n) transform and DCT is O(n2).And upcoming H.26l is going to take decoder complexities to completely new heights!
Slashdot Mirror
What about interoperability? Modularity?
Ah Yes, of course. microsoft is on top because there code is very interoperable, and uses non proprietory,open formats to make it easier for others to make interoperable software. And examples would be Ms version of kerberos, their JVM, their extensions to javascript, html, etc, etc.
As for modularity, they have themselves claimed in court that their code parts are so intertwined that they cannot be isolated. And of course, Your(or mine) boss cares a lot whether MS Word is modular on not, In fact, if it were not, he would stop using it!
The reason that someone is on the top is that it was chosen by IBM as the Os vendor when they launched the PC.
It all depends on the cross section of the said strangelet. If, it is the size of a quark, it will probably just pass through your body without touching anything!
The reason is of course, that an atom is 99% ( maybe more) empty space. All the mass(okay, most) is concentrated at neucleus, and so the possibility of a strangelet just passing through would be very high.
However, if a collison did occur,it would just push the colliding atom out of way and continue on its path. Not much will happen.
So my thinking is, nobody needs to bother about strangelets hitting them.
That said, I feel that todays particle physics has too many particles, and that is sign that something is wrong with our theories. Of course, this has no basis at all.
You americans think always think starwars is the greatest thing in the universe. I am reaaaally bored with this! Sigh. somehow people fighting with tubelights entertains people.
JPEG 2000 will have blocking artifacts at lower bit rates, for the same reason that Jpeg has: Subdividing the image into blocks, and treating each block separately.
OK, time to debunk some common misunderstandings 1. DCT compression produces blocking artifacts. Nope, it dosen't. JPEG standard breaks image into blocks of 8x8, and applies DCT on the blocks separately. Now these blocks are really related to each other (have similar color etc). Jpeg treats them separately. When using higher compression, a lot of information is discarded, which results in blocks looking slightly different. That's what causes blocking artifacts. 2. Wavelet compression is Significantly better This is true only for very low bit rates. At higher bitrates DCT based methods can acheive efficiencies very close to JPEG 2000 (note that jpeg can't do it, but some newer papers have suggested methods which can). 3. Wavelet compression does not produce blocking artifacts. Well, it doesn't, but the JPEG 2000 images will have blocking artifacts for the same reasons that JPEG has... dividing images in blocks and treating them separately. Note that dividing images into blocks and then compressing blocks reduces compression in both DCT and Wavelet transforms . With DCT you can't compress a whole image in One shot (computation intensive), but with wavelets it was possible. I regard this as a missed oppurtunity. So, to sum up, JPEG2000 isn't as earth shattering it is made out to be, it won't revoulutionize image compression the way jpeg did. Its just another image format. period.
5x reduction in storage space? Not really. Only advantage of Jpeg 2000 over jpeg lies in better quality at lower bit rates, and the ability to encode ROI etc. Compression is not a big argument in favour of Jpeg 2000.
Perhaps this is news to you, but US is the only country in the world to have ever used a nuke. It has the world's largest arsenal of nukes, and the capability to deliver it. It is one of the few countries that have a habit of nosing into others affairs, and has shown the tendency to use force at the slighest pretext. It funded the mujahedeen in Afganistan, then funded the taliban, it masscared the vietnamese, it has put a stranglehold on Iraq, leading to shortage of food and medicines. It kills afgan civilians on the sightest suspicion that they are Al-qaeda.That's why the news is dangerous. That's why the US is the greatest threat to world security.Period.
Ah, the classical war monger.
Nukes are weapons and meant to be used. Any body who attacks us must be nuked. we, and only we have the right to exist. So what if generations of the victims would be born wit h horrible genetic defects? So what if complete ecosystem would be destroyed? They dared to attack us, and paid the price. What if the nuclear fallout will cause horrible deaths to their people? So what if there will be babies born without eyes, without livers, mentally retarded? They dared to oppose us, the ones god designated to be above all, to be the most powerful, to rule.
Your military education has completely indoctrinated you to such a task. Remember, you were trained to drop Nukes, and you would have to do that without feeling any remorse at the killing of millions. You will have to be trusted, and therefore were indoctrinated. Your enemy sir, is also human. Treat him like a human. Kill him maybe, destroy his army, but let his young sons live.Don't use nukes to cover up your cowardice.
They proposed the CTBT as an instrument to prevent nuclear have nots from becoming haves. For, in world populated with nuclear weapons, superiority in conventional weapons means nothing. Having a nuke does not constitute an advantage unless the other guy does not have it.
And they berated India & Pakistan for seeing throught this strategem and not signing the treaty. Abd you know what? they themselves haven't signed it.
US posseses the largest nuclear arsenal in the world, and is the only country to have dropped one. Unless the US can drop nukes on others, and escape retaliation, they are of no use. Hence the missile defense program.
Well, In India, we can buy CD-R's for Rs 20, which comes to about $0.45, (and if I am not wrong to 50 cents) They are not the best quality(usually locally made), but you get the idea.
It's wrong to think of MPEG-4 as "a single standard" in the sense that any implementation needs to handle all of the spec.
This is known. But the point is, all those profiles aren't necessary at all. If what people want is a video codec with higher compression, then that's what you do. Trying to be all for everything is not only futile, but is also stupid to do. MPEG 4 reminds me of OSI protocol standards. Both have unecessary stuff, and both were formulated before their needs were felt. And I predict that MPEG 4 will share the same fate. As for open source providing a those codecs, they probably will happen, eventually. If that's all you need, then you don't need MPEG-4.
Unfortunately, that's the truth: That's what people need. Obvoiusly nobody is interesed in having an interactive scene were you can move things around, these are applications which dreamt up by the MPEG 4 comittee, and nobody wants them, further, AFAIK, nobody has shown any interest in them. I work in a company known for its expertise in multimedia. And believe me, nobdy has ever inquired with us for any solutions other than simple profile@level 3 But is that open-source codec going to work with new mobile phones from different vendors, over lossy networks?
Implementation over lossy networks is no big deal at all: essentially what you do is embed some resync markers in your bit stream. So I don't think that's going to be too difficult. Yes, mpeg 4 has reversible VLC, too, but that's not too difficult to do, is it?
As someone who works with JPEG 2000 I have to call you on your misinformation.
Well, As somebody who knows both Jpg (earlier version) and Jpeg 2000 I stand by my statements. Consider: Divison of an image into components, division of components into tiles, then subbands, preccints, layers. All this complexity was not there in JPG (obviously when I said JPEG 2000 was complex I was not comparing it to MPEG). Then you have ROI coding etc. Then for transmitting the image you have 4 combinations possible. Then you have those passes, etc.This is entirely unecessary. Only advantage of wavelet transforms over DCT transform comes at low bitrates. Instead of using it straightforwardly, the standards committee went ahead and added useless frills. JPEG 2000 uses wavelets.
I remember mentioning that. As for reading the standard, one can even read the MPEG 4 standard, if one has enough time. I myself had read all the stuff required for core profile. (for video only!)
We've seen this with JPG, where as a result a lot of websites are switching to PGN, and now we will see this again with MPEG4.
You probably meant GIF (instead of JPG).
MPEG-4 is a complete mess. Completely true. It is too complex, too big, has too many options. It tries to be everything to everybody, and fails. It has dozens of profiles and levels, and except audio-video compression, all others are completely useless. It has profiles for transmission of facial data, 3D data, shape decoding, Structured Audio etc. All of them are extremely lowlevel and crude specially facial data transmission and 3D data transmission. To say that they will be of use in 3D games & interactive applications is a pipedream. Basically, MPEG 4 is far too complex for these things, and is definitely not enough. Further, the advances made in MPEG 4 are incremental only, and it should be not too difficult to acheive similar effects by different means.If you look at the H.263 standard, you'd know that much of Mpeg 4 is simply H.263 in disguise (specially the video part), with some frills added. Standard committes in recent times have gone completely overboard and created mammoth, unusable standards. MPEG 4, and more recently JPEG 2000 are a case in point. JPEG 2000 uses 4 times the CPU time of JPEG!. And this is despite the fact that wavelet is a O(n) transform and DCT is O(n2).And upcoming H.26l is going to take decoder complexities to completely new heights!