The biggest reason that more diesels aren't sold in the states is that California banned the sale of new ones.
That's really inaccurate and unfair.
California raised their standards on diesel cars to very high levels, which precluded all but VW's top vehicles from being sold. Last year, though, even those weren't clean enough, and there were none to chose from.
The switch to low sulfer diesel will dramatically change that state of affairs.
Why do so many folks go nutty over proving that hybrids are the greatest thing ever
Because people need to justify paying the ~33% higher sticker price... Being "slightly more fuel effecient" just doesn't do it, unless you do ridiculous ammounts of driving, or keep the car as your primary vehicle for something like 10 years.
And in that case, it's quite possible the higher maintenance costs could bite you as well, and move that target back even further.
The primary reason hybrids get good mileage in cities is that they are able to shut off engine at idle and that they are able to recoup energy otherwise lost energy through regenerative braking. This is why hybrids do not get as good mileage as diesels outside of city driving.
That is true. However, there is another reason...
The electric motors used in hybrids have a maximum speed of about 40MPH. After that, you're driving on all-gasoline, no matter what the driving conditions (eg. highways with stop-lights, congested freeways, etc.).
Such conflicts are invariably bad for the consumer in the short run since we have to guess which tech will come out on top and whoever guesses wrong gets his fingers burnt.
No, it's GOOD for the public, because the competition puts pressure for lower prices and more features on the manufacturers. If they had no competition, the players would still cost $1,000 each, possibly more.
Worse for them, sooner or later China is going to be able to foist their version of a next-gen SPD (Shiny Plastic Disc) on the world.
Everyone wants to make China the big bogey man, but it has no basis in reality.
The only notable things China have done involve simply repurposing existing technology. For SVCD, they used very old CD technology, with MPEG-2 for video and audio. For EVD, they are using (now-old) DVD technology, with their home-grown video codec, which is notably inferior to the 15-year old MPEG-2 standard.
10 years down the line, China might come-up with some highdef format. By then, HD-DVD and/or Blu-ray will be dirt cheap, and heavily entrenched. And China's format will probably be entirely based on one of the two existing formats, anyhow.
Personnaly I am sort of happy to see GM get thier lunch eaten. They've been asleep at the switch for too many years.
GM has made many mistakes in the past few years, but it's quite sad that it's Toyota that is taking it's place.
I've seen many GM vehicles, and many Toyotas, and would you like to guess which have the most problems? Toyota seems to be popular because of hype and advertising, rather than cars that are actually (mechanically) good.
The vocal supporters of Toyotas that I've heard from are always comparing their current-generation Toyotas to some GM vehicle from 20 years ago. It's amazing how easily things could have gone the other way, and quite sad to see how incredibly biased the public is in their value judgements.
Second: In your senario, there's no reason not to license the music you want to sample.
Third: There's a world of difference between creating something and recycling something that already exists.
Fourth: The person in question probably wouldn't be putting their effort into making the song, if they didn't think they could have some control over it... usually money (directly or indirectly).
2a. Why can an old guy take a drug to make his dick hard when I can't smoke a joint?
How are these two even remotely related?
Why can a restaurant sell a burger for $10 when it's raining outside?
Get rid of medical lawsuits. A judge and jury have no idea if what a doc did was right or wrong.
A) You could say the same thing about ANY specialized subject. A judge and jury have no idea if cigarettes are addictive or not.
B) Many recent studies have shown that the large majority of doctors and hospitals more or less deserve to be sued for malpractice. They cut corners, hire FAR fewer nurses than required, and other similar cost-saving practices which directly correlate to a higher death-rate.
I think I agree with what you were going for (banning doctors convicted of malpractice) but I can't envision any way this could work in the real world... yet.
All money taken from their budget should be from cruft (how much does DoD spend on office supplies) or from special projects (Do we really need the JSF right after the F-22).
7a. Reduce the funding to every government agency by 2% per year until the customers start complaining.
It sounds good on paper, but in the real world leads to corruption (no-bid contracts, sub-par equipment delivered, accounting failures, etc), more money and more dead soldiers due to maintaining the old, cheap, low-tech equipment longer than it should have been in-service.
Lots should be cut-out of the federal budgets, but it takes tremendous ammounts of work to figure out exactly what and where, without compromising performance.
8. Threaten corporate shareholders with jail for withholding good drugs at low cost.
9. Mandate one special project for major companies.
Very difficult to write such legislation without huge loopholes companies can squeeze through with bare-minimal effort.
And to enforce it??? You're going to need to set-up another government office, with more paid employees.
Slightly-used Psions are still readily available. The included EPOC (now Symbian) Office suite is damn-near as good as full-fledged desktop software, including full formatting, embedding drawings or spreadsheets into documents, etc.
You can print ANYTHING directly from the unit to infrared or serial printers. There is even a 3rd party PDF printer (shareware), as well as a GPLd PDF viewer.
As for durability, the (spring) hinges on the 5s have an unfortunate tendancy to break, but the unit will continue to function, with just a bit more hassle when opening/closing the unit. Never heard of any problems on other units.
ESR does make some good points, despite everyone here trying to say he's wrong because of [insert personal pet peave]. He is correct that Linux (and FreeBSD) is far ahead of the competitors on x86-64 hardware. Unfortunately, he doesn't really provide any solutions, just little bits and pieces here and there.
Open Source software really is right there... it can do 99% of everything people need. The problem is: "some assembly required." ALL distros of both Linux and BSD, that I've seen (and I've seen just about all of them), cop-out on actually putting together an operating system, rather than just a random collection of various software. The current problems are numerous, but they could *all* be worked-out by a small team of perhaps 5 people who know what they're doing:
Configuration: -- First and foremost is GUI configuration. There should be a handful of applications in the menu, which allow you to configure every possible aspect of the system. From your network card profile, aliases, etc., to GRUB/Lilo boot-up options. Many projects/distros have done tiny bits and pieces, but they usually don't work properly, and certainly don't cover all cases. And those that do exist are too often just a glorified text editor, requiring you to already know the option names, and type them into some dialog yourself. EVERYTHING needs to be there. With this small step, configuration could be easier than Windows/Mac. -- Second is command-line configuration. BSD has it all over Linux on this one. Where on Linux you're editing a dozen files, FreeBSD has this down to effectively 1 (2 other are very rarely, if EVER, needed), and has example files with every possible option listed. The only mistake FreeBSD makes, IMHO, is listing the defaults in a seperate file, rather than including the common ones (commented-out) in the main config file itself. In other words, Linux distros really need to have configuration in a standard location, and everything should be configurable in 2 or 3 files, rather than requiring people to edit several of the their rc.XYZs. Ironically, Slackware was closest to BSD simplicity around v9, but has started moving torwards more and more complexity...
Default Configuration: -- The default configuration all too often completely sucks. Options that 99.999% of users are going to want, for some reason aren't enabled by default. The distros just pass the buck, and leave it to the users to every time configure things from scratch. For example, take Firefox... Why don't distros set a minimum font size, correct DPI, Download locations, etc. -- Why must everyone be an expert on modules.conf just to get their hardware working? A nice long default modules.conf packed with EVERYTHING, and sane default options, should be included. -- Why do included filemanagers so often have no associations? I realize GNOME and KDE's methods are somewhat iffy and not commonly used, but even with the most brain-dead filemanager, you could easily pre-associate file types with all possible applications, so the user need only look at what apps could open the file, and install one of them. You could go slightly further and launch the package manager, opened to the proper section, if none of them are yet installed.
Desktop: -- Select a toolkit in the beginning, and use it. That means if you select QT, and no QT versions of an app are available, don't install one, nor make an available package. Mixing different toolkits, different desktop environments, different button layouts, etc., just makes for a confusing mess, which can be easily and benefitally avoided. Even if the toolkit you select isn't the one the user may like, the simplicity and other benefits will likely make them happy to accept it anyhow. eg. I may prefer driving on the left side of the road, but I would rather have _everyone_ driving on the right, rather than people chosing for themselves, and therefore having to constantly switch from one to another... -- The above means you pick your desktop environment, and s
If you decide to stream some songs with the intention of trying to "stay legit" (if you don't care, that is another story) and paying royalties, the costs of doing so are prohibitive except for those who have some serious $$$ behind them.
The internet remain free for anyone to use, which is vastly unlike the airwaves...
This "government interference" can be completely circumvented by playing music that is not owned by RIAA/ASCAP. There isn't any similar way to circumvent the restrictions on setting-up your own radio station.
Prohibitive fees, whether for a spectrum license, minimum royalties or any other arbitrary basis that keeps publishing in the hands of "the club", is the barrier that favors "official publishers".
You can avoid those "minimum royalties" if you don't insist on publishing OTHER PEOPLE'S WORKS.
Your willful ignorance of this simple fact is overwhelming.
On a good night, 600 miles away, I can still listen to them.
That's nothing...
Every night 850KHz becomes an extremely strong signal around here... Absolutely no fading to speak of.
850 is KOA in Denver, Colorado, and I'm outside of Los Angeles, CA. That's around 1000 miles, and it's still a stronger signal than any of the stations from Los Angeles, around 50 miles away.
Broadcasting from the top of a mountain automatically gives you the tallest antenna, anywhere.
That's like saying that the radar wasn't really invented until there were planes for it to track, or the TV wasn't invented until there were a million households from which to gather ratings.
Horrible analogies.
In fact, YOU are the one (essentially) saying that radar and TV were invented by Tesla...
Tesla did many things with radio waves, but he never transmitted audio, never set-up a public-broadcast radio station (which is the anniversy in question) etc.
Those who discovered electricity aren't responsible for every subsuquent device which happened to require/use electricity to operate. Ditto for radio waves.
Publishing streams of copyrighted material on the Net costs not only a ridiculous $0.0007 per "song" per listener (therefore 10K listeners costs $7, thousands of times more than broadcast, though the tech is cheaper). But the license requires a minimum $500 per year.
I can understand frustration with the fees for internet streaming, but equating copyright fees to FCC licenses is patently ridiculous.
It doesn't matter what the media... YOU CAN'T JUST FREELY COPY OTHER PEOPLE'S WORKS.
No matter what the fees have been established at, the internet pipes remain free and clear, completely contrary to your comparison.
You can have an internet radio station playing only your own music. Or you can privately negotiate lower fees with copyright holders (eg. non-RIAA artists). You don't pay a large fee to "broadcast" on the internet... Anybody can do it. (And Net non-Neutrality can't stop it either.) Just look at the overwhelming number of podcasts available.
Yes, but you can typically say with certainly whether a particular codec will be able to run with reasonable parameters on a single core or not.
"Reasonable" is entirely in the eye of the beholder.
The fact of the matter remains, just because most systems can't play back high resolution video on a single core, does not mean those writing it are going to make it threaded.
Examples of which include three common codecs, all of which are known to work fine on a reasonably modern single-core system.
WHAT THE HELL ARE YOU TALKING ABOUT?
He lists three programs which play videos. He does not mention specific codecs.
I have no idea where you get "known to work fine on a reasonably modern single-core system" from. How did you possibly determine that? How did you reach that conclusion?
It's a bit like saying "bicycles are know to be able to exceed the speed limit"...
Hulk was a Shakespearian, father-son conflict, tragedy shot comic book panel style.
No, it was a case of someone taking an action movie, and try to pretent it was serious. The end result was to watch everyone walk around for 2 hours, utterly depressed about everyone and everything, for no particular reason. As well as everyone making blindly stupid, and obviously horrible decisions, with no consistency, just to move the plot along.
It was esentially a really crappy Batman movie, trying to also be something akin to (a really crappy) American Beauty, and failing miserably on all counts.
There is basically no difference between transmitting in 1080i vs 1080p when viewing content at or below 30 frames/second.
That is just utterly wrong. No matter what the framerate, with an interlaced display you have artifacts like spacial and temporal aliasing, twitter, etc.
With 24fps material on a 60fps interlaced display, you have to put up with things like judder as well, which can look really terrible in panning scenes.
When talking about high def tv's, you're mostly talking about progressive displays
The vast majority of HDTVs are interlaced. Plasmas, LCD, etc., are in the minority.
Direct-view and projection CRTs are still the standard, because of the higher resolution, response-time, contrast, and price.
1080i sends half the image on cycle 1 and half the image on cycle 2, your tv deinterlaces the image fields and shows you a progressive image for 2 frames. 1080p on the other hand sends the whole image on cycle 1, and nothing on cycle 2, and shows the progressive image for 2 frames as well.
This shows a complete and utter lack of understand of any relevant video concepts.
Film is converted to interlaced with 3:2 pulldown (aka telecine), and you need significant processing power to reverse it (IVTC) to return it to the original progressive image. Even with unlimited processing power, there has never been an IVTC process divised which does a perfect job. You will always get some artifacts. Plus, I'm willing to bet that the under-powered processors in most HDTVs aren't good enough to even keep those to a minimum. .
For some reason though, ignorant idiots like yourself (Toshiba shills, perhaps?) repeatedly spout-off on stuff like this whenever HDTV comes up on slashdot, and manage to trick some unsuspecting mods into giving them points for their misinformation.
Perhaps now you can jump to the next one, and say that people can't possibly see any improvement of HDTV over standard definition, or that the DRM on HD-DVD/Blu-ray is so much worse than CSS on DVDs...
Slashdot Mirror
That's really inaccurate and unfair.
California raised their standards on diesel cars to very high levels, which precluded all but VW's top vehicles from being sold. Last year, though, even those weren't clean enough, and there were none to chose from.
The switch to low sulfer diesel will dramatically change that state of affairs.
Because people need to justify paying the ~33% higher sticker price... Being "slightly more fuel effecient" just doesn't do it, unless you do ridiculous ammounts of driving, or keep the car as your primary vehicle for something like 10 years.
And in that case, it's quite possible the higher maintenance costs could bite you as well, and move that target back even further.
That is true. However, there is another reason...
The electric motors used in hybrids have a maximum speed of about 40MPH. After that, you're driving on all-gasoline, no matter what the driving conditions (eg. highways with stop-lights, congested freeways, etc.).
No, it's GOOD for the public, because the competition puts pressure for lower prices and more features on the manufacturers. If they had no competition, the players would still cost $1,000 each, possibly more.
Everyone wants to make China the big bogey man, but it has no basis in reality.
The only notable things China have done involve simply repurposing existing technology. For SVCD, they used very old CD technology, with MPEG-2 for video and audio. For EVD, they are using (now-old) DVD technology, with their home-grown video codec, which is notably inferior to the 15-year old MPEG-2 standard.
10 years down the line, China might come-up with some highdef format. By then, HD-DVD and/or Blu-ray will be dirt cheap, and heavily entrenched. And China's format will probably be entirely based on one of the two existing formats, anyhow.
Call me when 50GBs of bandwidth, and equivalent ammounts of hard drive space, are cheaper (and faster) than a 12cm piece of plastic.
GM has made many mistakes in the past few years, but it's quite sad that it's Toyota that is taking it's place.
I've seen many GM vehicles, and many Toyotas, and would you like to guess which have the most problems? Toyota seems to be popular because of hype and advertising, rather than cars that are actually (mechanically) good.
The vocal supporters of Toyotas that I've heard from are always comparing their current-generation Toyotas to some GM vehicle from 20 years ago. It's amazing how easily things could have gone the other way, and quite sad to see how incredibly biased the public is in their value judgements.
First: Copyright is not Patents.
Second: In your senario, there's no reason not to license the music you want to sample.
Third: There's a world of difference between creating something and recycling something that already exists.
Fourth: The person in question probably wouldn't be putting their effort into making the song, if they didn't think they could have some control over it... usually money (directly or indirectly).
Cause and effect???
Why are GNOME/KDE the two most popular GUIs? Could it be BECAUSE they copy Windows?
I'd say GNOME (default layout) is much more MacOS-like, than it is Windows-like.
And XFce seems to be up-and-comming. Perhaps we'll see more than just the "fast" or "mini" distros shipping with XFce as the default WM soon.
How are these two even remotely related?
Why can a restaurant sell a burger for $10 when it's raining outside?
A) You could say the same thing about ANY specialized subject. A judge and jury have no idea if cigarettes are addictive or not.
B) Many recent studies have shown that the large majority of doctors and hospitals more or less deserve to be sued for malpractice. They cut corners, hire FAR fewer nurses than required, and other similar cost-saving practices which directly correlate to a higher death-rate.
I think I agree with what you were going for (banning doctors convicted of malpractice) but I can't envision any way this could work in the real world... yet.
It sounds good on paper, but in the real world leads to corruption (no-bid contracts, sub-par equipment delivered, accounting failures, etc), more money and more dead soldiers due to maintaining the old, cheap, low-tech equipment longer than it should have been in-service.
Lots should be cut-out of the federal budgets, but it takes tremendous ammounts of work to figure out exactly what and where, without compromising performance.
Very difficult to write such legislation without huge loopholes companies can squeeze through with bare-minimal effort.
And to enforce it??? You're going to need to set-up another government office, with more paid employees.
Too large and bulky, IMHO.
Slightly-used Psions are still readily available. The included EPOC (now Symbian) Office suite is damn-near as good as full-fledged desktop software, including full formatting, embedding drawings or spreadsheets into documents, etc.
You can print ANYTHING directly from the unit to infrared or serial printers. There is even a 3rd party PDF printer (shareware), as well as a GPLd PDF viewer.
As for durability, the (spring) hinges on the 5s have an unfortunate tendancy to break, but the unit will continue to function, with just a bit more hassle when opening/closing the unit. Never heard of any problems on other units.
By all means, list a few examples... because your claim flies in the face of all logic.
Thanks for the info. I would have looked it up sooner or later...
ESR does make some good points, despite everyone here trying to say he's wrong because of [insert personal pet peave]. He is correct that Linux (and FreeBSD) is far ahead of the competitors on x86-64 hardware. Unfortunately, he doesn't really provide any solutions, just little bits and pieces here and there.
Open Source software really is right there... it can do 99% of everything people need. The problem is: "some assembly required." ALL distros of both Linux and BSD, that I've seen (and I've seen just about all of them), cop-out on actually putting together an operating system, rather than just a random collection of various software. The current problems are numerous, but they could *all* be worked-out by a small team of perhaps 5 people who know what they're doing:
Configuration:
-- First and foremost is GUI configuration. There should be a handful of applications in the menu, which allow you to configure every possible aspect of the system. From your network card profile, aliases, etc., to GRUB/Lilo boot-up options. Many projects/distros have done tiny bits and pieces, but they usually don't work properly, and certainly don't cover all cases. And those that do exist are too often just a glorified text editor, requiring you to already know the option names, and type them into some dialog yourself. EVERYTHING needs to be there. With this small step, configuration could be easier than Windows/Mac.
-- Second is command-line configuration. BSD has it all over Linux on this one. Where on Linux you're editing a dozen files, FreeBSD has this down to effectively 1 (2 other are very rarely, if EVER, needed), and has example files with every possible option listed. The only mistake FreeBSD makes, IMHO, is listing the defaults in a seperate file, rather than including the common ones (commented-out) in the main config file itself. In other words, Linux distros really need to have configuration in a standard location, and everything should be configurable in 2 or 3 files, rather than requiring people to edit several of the their rc.XYZs. Ironically, Slackware was closest to BSD simplicity around v9, but has started moving torwards more and more complexity...
Default Configuration:
-- The default configuration all too often completely sucks. Options that 99.999% of users are going to want, for some reason aren't enabled by default. The distros just pass the buck, and leave it to the users to every time configure things from scratch. For example, take Firefox... Why don't distros set a minimum font size, correct DPI, Download locations, etc.
-- Why must everyone be an expert on modules.conf just to get their hardware working? A nice long default modules.conf packed with EVERYTHING, and sane default options, should be included.
-- Why do included filemanagers so often have no associations? I realize GNOME and KDE's methods are somewhat iffy and not commonly used, but even with the most brain-dead filemanager, you could easily pre-associate file types with all possible applications, so the user need only look at what apps could open the file, and install one of them. You could go slightly further and launch the package manager, opened to the proper section, if none of them are yet installed.
Desktop:
-- Select a toolkit in the beginning, and use it. That means if you select QT, and no QT versions of an app are available, don't install one, nor make an available package. Mixing different toolkits, different desktop environments, different button layouts, etc., just makes for a confusing mess, which can be easily and benefitally avoided. Even if the toolkit you select isn't the one the user may like, the simplicity and other benefits will likely make them happy to accept it anyhow. eg. I may prefer driving on the left side of the road, but I would rather have _everyone_ driving on the right, rather than people chosing for themselves, and therefore having to constantly switch from one to another...
-- The above means you pick your desktop environment, and s
The internet remain free for anyone to use, which is vastly unlike the airwaves...
This "government interference" can be completely circumvented by playing music that is not owned by RIAA/ASCAP. There isn't any similar way to circumvent the restrictions on setting-up your own radio station.
You can avoid those "minimum royalties" if you don't insist on publishing OTHER PEOPLE'S WORKS.
Your willful ignorance of this simple fact is overwhelming.
Yes, I do.
An FCC license is necessary to broadcast over the radio waves. It is not for internet streaming.
Paying fees for copyrighted music is NOT required for internet (radio) streaming, because playing copyrighted material is not necessary.
Fees for the medium != Fees for the content
That's why your comparison is "patently ridiculous".
It's not his fault. He works for Verizon...
And he's mistaken, anyhow.
His link to HP.com is for full COLOR print-outs, which is, at best, a tiny minority of the pages in almost all books.
Around 1 cent/page (black) isn't uncommon for professional units.
That's nothing...
Every night 850KHz becomes an extremely strong signal around here... Absolutely no fading to speak of.
850 is KOA in Denver, Colorado, and I'm outside of Los Angeles, CA. That's around 1000 miles, and it's still a stronger signal than any of the stations from Los Angeles, around 50 miles away.
Broadcasting from the top of a mountain automatically gives you the tallest antenna, anywhere.
Horrible analogies.
In fact, YOU are the one (essentially) saying that radar and TV were invented by Tesla...
Tesla did many things with radio waves, but he never transmitted audio, never set-up a public-broadcast radio station (which is the anniversy in question) etc.
Those who discovered electricity aren't responsible for every subsuquent device which happened to require/use electricity to operate. Ditto for radio waves.
I can understand frustration with the fees for internet streaming, but equating copyright fees to FCC licenses is patently ridiculous.
It doesn't matter what the media... YOU CAN'T JUST FREELY COPY OTHER PEOPLE'S WORKS.
No matter what the fees have been established at, the internet pipes remain free and clear, completely contrary to your comparison.
You can have an internet radio station playing only your own music. Or you can privately negotiate lower fees with copyright holders (eg. non-RIAA artists). You don't pay a large fee to "broadcast" on the internet... Anybody can do it. (And Net non-Neutrality can't stop it either.) Just look at the overwhelming number of podcasts available.
That weird guy down the hall, who always has to pick the urinal right next to me...
Wow. I didn't realize I was talking to an insane person. My mistake.
"Reasonable" is entirely in the eye of the beholder.
The fact of the matter remains, just because most systems can't play back high resolution video on a single core, does not mean those writing it are going to make it threaded.
WHAT THE HELL ARE YOU TALKING ABOUT?
He lists three programs which play videos. He does not mention specific codecs.
I have no idea where you get "known to work fine on a reasonably modern single-core system" from. How did you possibly determine that? How did you reach that conclusion?
It's a bit like saying "bicycles are know to be able to exceed the speed limit"...
No, it was a case of someone taking an action movie, and try to pretent it was serious. The end result was to watch everyone walk around for 2 hours, utterly depressed about everyone and everything, for no particular reason. As well as everyone making blindly stupid, and obviously horrible decisions, with no consistency, just to move the plot along.
It was esentially a really crappy Batman movie, trying to also be something akin to (a really crappy) American Beauty, and failing miserably on all counts.
That is just utterly wrong. No matter what the framerate, with an interlaced display you have artifacts like spacial and temporal aliasing, twitter, etc.
With 24fps material on a 60fps interlaced display, you have to put up with things like judder as well, which can look really terrible in panning scenes.
The vast majority of HDTVs are interlaced. Plasmas, LCD, etc., are in the minority.
Direct-view and projection CRTs are still the standard, because of the higher resolution, response-time, contrast, and price.
This shows a complete and utter lack of understand of any relevant video concepts.
Film is converted to interlaced with 3:2 pulldown (aka telecine), and you need significant processing power to reverse it (IVTC) to return it to the original progressive image. Even with unlimited processing power, there has never been an IVTC process divised which does a perfect job. You will always get some artifacts. Plus, I'm willing to bet that the under-powered processors in most HDTVs aren't good enough to even keep those to a minimum.
.
For some reason though, ignorant idiots like yourself (Toshiba shills, perhaps?) repeatedly spout-off on stuff like this whenever HDTV comes up on slashdot, and manage to trick some unsuspecting mods into giving them points for their misinformation.
Perhaps now you can jump to the next one, and say that people can't possibly see any improvement of HDTV over standard definition, or that the DRM on HD-DVD/Blu-ray is so much worse than CSS on DVDs...