Okay - I'm a Software Engineer at Microsoft. And yes, we're listening.
In regards to Microsoft operating as a cluster of separate companies: I have worked in large companies before, and I believe that MS does better than average at working cohesively toward common goals. This is an incredibly hard thing to achieve in such a large organization and it's something we continually strive to improve.
Having said that, I believe it is important to allow our engineers some freedom to take slightly differernt approaches to the problems that they're working on - this encourages innovation. The down-side of this is that some products may not integrate as smoothly as others in the early stages, but seamless integration will come as the products mature. There are heaps of great examples of this - Messenger, DirectX, PowerPoint, PnP, Xbox, Media Center, IE... all of these technologies innovated in a way that may have seemed orthogonal to our other products, and didn't integrate terribly well in the early stages. As these products have matured, they have become more seamless and work better with other technologies.
Bureaucracy? I have heard this comment before, but, to be honest, I don't see it. Microsoft has much less red tape than other companies I have worked for. That's one of the things I love about working here as an engineer - we just do our job and build cool stuff. It's almost like the rest of the company just exists to make that easier.
I know that most of the people who have read this far are thinking "Holy Cow! Check out the Micro$oft fanboy! The PR department has him trained!". I'm the first to admit that we're not perfect - in fact we're a long way from it. But we're self-critical and we're always trying to improve.
You Americans are so paranoid. You think they'll be communist aliens? Or are you just worried that they'll be Muslims? Better launch a pre-emptive strike before it's too late.
I was at the presentation in Melbourne. Wayne made it very clear that they did not intend to open source the Google search engine. That would be commercial suicide (my words, not his).
What he did suggest is that they are keen to give something back to the OSS community that has given them so much. The impression that I got was that they would return much of their operating systems work to the community. They do spend a great deal of effort in Linux kernel hacking and so forth.
I was very pleased to hear this from a senior member of Google's engineering team. We're yet to see the fruits of this decision, but the policy change has to come first. Let's hope it works out.
The difference in quality that is noticable in the shots shown in this article do not represent the capabilities of the codec.
Different encoders are of different qualities (think lame vs. bladeenc vs. l3enc - all mp3 encoders, but all have different output quality). What they are really comparing is specific implementations of encoder and decoder.
The quality of the decoder is critical. Windows Media 9 has substantial post-processing of the decoded content, which reduces blockiness and other artefacts. It is probable that the MPEG-4 decoder is not using these postfilters, so it looks nasty when played back, just as, if you remove the postfilters from WM9, it looks horrible at low bitrates.
The only way to compare the codecs fairly is to use multiple implementations of both encoder and decoder, and deactivate all postfilters. Then take the best implemetation of each codec for comparsion. This will give a much better indication of how good the codecs are. Anything short of this is, at best, misleading.
Another note about the article is the selection of clips. When comparing video codecs, it is worthwhile having different types of content. The action sequences chosen are good examples, but one of them would be enough. The Monsters Inc clip is also worthwhile (to compare colour fidelity and sharpness), but a sport clip should be included (basketball is good because of the polished floors and seas of faces in the crowd) as well as traditional animation, such as The Simpsons. Animation is actually quite hard to encode.
Unfortunately, I am yet to see a particularly good comparison of video codecs outside where I work (Telstra Research Labs). Unfortunately the work done there cannot be shared with Slashdot.
I was recently asked by an anonymous friend to post a review of a beek that he had recently published.
I refused to post a review before reading the book, but I am sure that the review I posted was a little biased.
He also asked several other people to do the same thing, and I am aware of at least two who "reviewed" the book without reading it.
It's hard to deny that this behavior is deceptive, but how can you police it? Sure, you can prevent an author reviewing their own work, but how can you prevent that review being posted by proxy?
But perhaps a low-tech solution might be more sensible? As a victim of automobile hail damage, it would be great to have a preventative. A carport would do the job nicely.
There is currently research in cooling to prevent brain damage in infants that suffer hypoxia (oxygen starvation) during birth. Hypoxia occurs when the baby does not get an adequate oxygen supply to the brain during labour, which is usually as a result of fetal distress or pinching of the umbilical cord.
Usually, an infant in intensive care is put under lights to keep body temperature up, but in one study (Infant Cooling Evaluation at the Royal Womens Hospital in Melbourne, Australia) infants are cooled simply by turning the lights off.
There is not sufficient data to report results of this study at present, but resutls in animals have been very promising.
I'd recommend simplydv.co.uk. Their reviews are detailed an impartial, and they give you a good idea of what to look for in a DV cam.
I recently bought a Canon MVX150i (Optura 20 in US) and it's been really good. I recommend buying from a company that has a good reputation in imaging, like Sony or Canon - you're more likely to get good metering and quality optics. These are the most important things in any camera.
Slashdot Mirror
In regards to Microsoft operating as a cluster of separate companies: I have worked in large companies before, and I believe that MS does better than average at working cohesively toward common goals. This is an incredibly hard thing to achieve in such a large organization and it's something we continually strive to improve.
Having said that, I believe it is important to allow our engineers some freedom to take slightly differernt approaches to the problems that they're working on - this encourages innovation. The down-side of this is that some products may not integrate as smoothly as others in the early stages, but seamless integration will come as the products mature. There are heaps of great examples of this - Messenger, DirectX, PowerPoint, PnP, Xbox, Media Center, IE... all of these technologies innovated in a way that may have seemed orthogonal to our other products, and didn't integrate terribly well in the early stages. As these products have matured, they have become more seamless and work better with other technologies.
Bureaucracy? I have heard this comment before, but, to be honest, I don't see it. Microsoft has much less red tape than other companies I have worked for. That's one of the things I love about working here as an engineer - we just do our job and build cool stuff. It's almost like the rest of the company just exists to make that easier.
I know that most of the people who have read this far are thinking "Holy Cow! Check out the Micro$oft fanboy! The PR department has him trained!". I'm the first to admit that we're not perfect - in fact we're a long way from it. But we're self-critical and we're always trying to improve.
Keep the feeback coming. We're listening.
Get the actual video file for the large version of the trailer.
You Americans are so paranoid. You think they'll be communist aliens? Or are you just worried that they'll be Muslims? Better launch a pre-emptive strike before it's too late.
What he did suggest is that they are keen to give something back to the OSS community that has given them so much. The impression that I got was that they would return much of their operating systems work to the community. They do spend a great deal of effort in Linux kernel hacking and so forth.
I was very pleased to hear this from a senior member of Google's engineering team. We're yet to see the fruits of this decision, but the policy change has to come first. Let's hope it works out.
Different encoders are of different qualities (think lame vs. bladeenc vs. l3enc - all mp3 encoders, but all have different output quality). What they are really comparing is specific implementations of encoder and decoder.
The quality of the decoder is critical. Windows Media 9 has substantial post-processing of the decoded content, which reduces blockiness and other artefacts. It is probable that the MPEG-4 decoder is not using these postfilters, so it looks nasty when played back, just as, if you remove the postfilters from WM9, it looks horrible at low bitrates.
The only way to compare the codecs fairly is to use multiple implementations of both encoder and decoder, and deactivate all postfilters. Then take the best implemetation of each codec for comparsion. This will give a much better indication of how good the codecs are. Anything short of this is, at best, misleading.
Another note about the article is the selection of clips. When comparing video codecs, it is worthwhile having different types of content. The action sequences chosen are good examples, but one of them would be enough. The Monsters Inc clip is also worthwhile (to compare colour fidelity and sharpness), but a sport clip should be included (basketball is good because of the polished floors and seas of faces in the crowd) as well as traditional animation, such as The Simpsons. Animation is actually quite hard to encode.
Unfortunately, I am yet to see a particularly good comparison of video codecs outside where I work (Telstra Research Labs). Unfortunately the work done there cannot be shared with Slashdot.
J. R. R. Tolkein
Google can't convert the speed of light to parsecs per jiffy.
Divide by zero error
Try cheque.
He also asked several other people to do the same thing, and I am aware of at least two who "reviewed" the book without reading it.
It's hard to deny that this behavior is deceptive, but how can you police it? Sure, you can prevent an author reviewing their own work, but how can you prevent that review being posted by proxy?
But perhaps a low-tech solution might be more sensible? As a victim of automobile hail damage, it would be great to have a preventative. A carport would do the job nicely.
There is currently research in cooling to prevent brain damage in infants that suffer hypoxia (oxygen starvation) during birth. Hypoxia occurs when the baby does not get an adequate oxygen supply to the brain during labour, which is usually as a result of fetal distress or pinching of the umbilical cord. Usually, an infant in intensive care is put under lights to keep body temperature up, but in one study (Infant Cooling Evaluation at the Royal Womens Hospital in Melbourne, Australia) infants are cooled simply by turning the lights off. There is not sufficient data to report results of this study at present, but resutls in animals have been very promising.
I'd recommend simplydv.co.uk. Their reviews are detailed an impartial, and they give you a good idea of what to look for in a DV cam.
I recently bought a Canon MVX150i (Optura 20 in US) and it's been really good. I recommend buying from a company that has a good reputation in imaging, like Sony or Canon - you're more likely to get good metering and quality optics. These are the most important things in any camera.
Oh, and buy a second battery.
Here's a novel idea - have a different perspective on each screen, hold the unit three inches from your face and go crosseyed! Instant 3D! :-)