You need to use common sense. When a company is doing a product that might annoy a bigger company, and they pick a stupid name that cries out "SUE ME!", you should expect that at some point they are going to be turning all kinds of records over to that bigger company as part of the inevitable lawsuit.
You should not give such companies any information that you want them to keep secret.
Uhm...ever looked at a CRT with a magnifier? They've got hard-coded pixels, too.
It's very instructive to watch with a magnifier while playing with the width and height controls, to see how they map the logical pixels to the physical pixels. Do that, and you'll see the real reason that CRT's look good at different resolutions.
Size: approximately the same footprint as a person while in use. Not true for a folded scooter.
Ease of use: obviously, you've never used a bike in a city. Bikes are incredibly awkward as soon as you go indoors with them. Try riding a bike in an elevator, or down a hallway.
The intent with these large jukeboxes is that you put all your music on them and keep it there. Hence, you only transfer a given song once. The initial loading might take a few hours. After that, you can load new songs faster than you can rip them or download them from the net, so USB is fine for that.
Video is a lot bigger than audio. The majority of consumer broadband access is via cable, which usually has a 128 Kb/sec or 256 Kb/sec upload cap per user, and also has an upload cap per cable node. TCP gets very unhappy if the upstream gets saturated...downloads come to a crawl when ACKs can't get through. If a lot of people try to upload things, even with individual caps of 128 Kb/sec, it doesn't take many to hit the node's limit, and bring downloads to a stop.
This means that in addition to the content providers wanting to stop a video napster, the cable internet companies will also want very much to squash it. The impression I got with audio naptster was that the bandwidth usage was low enough to not hurt cable performance, and so the cable companies didn't care that much, and so didn't want to get involved.
Video peer-to-peer will have a much harder time catching on, I think...it won't take many cases of ISPs canceling accounts of people who do it to scare people away.
As others have pointed out, they are not selling the data. They are selling the formating of the data.
However, if you get one of the integrated Tive/DirecTV receivers, and are willing to look around on the net to find some hacks, you can hack it to get the program data from the DirecTV guide. You then don't need to hook it up to a phone, either, which is a plus for some.
I would guess that you could adopt that hack to work with the non-DirecTV units, if you give them some kind of network access. Grab the guide data from yahoo or gist or someplace like that, and write a perl script to convert it to Tivo's format.
From one of the things on their site: Although currently demonstrating its technology on very small bit strings, ZeoSync expects to overcome the existing temporal restraints of its technology and optimize its algorithms to lead to significant changes in how data is stored and transmitted (emphasis added).
Using time travel, high compression of arbitrary data is trivial. Simply record the location (in both space and time) of the computer with the data, and the name of the file, and then replace the file with a note saying when and where it existed. To decompress, you just pop back in time and space to before the time of the deletion and copy the file.
I realize that what I'm about to propose does not work. The challenge is to figure out why
Here's a proposal for a compression scheme that has the following properties:
1. It works on all bit strings of more than one bit.
2. It is lossless and reversible.
3. It never makes the string larger. There are some strings that don't get smaller, but see item #4.
4. You can iterate it, to reduce any string down to 1 bit! You can use this to deal with pesky strings that don't get smaller. After enough iterations, they will be compressed.
OK, here's my algorithm:
Input: a string of N bits, numbered 0 to N-1.
If all N bits are 0, the output is a string of N-1 1's. Otherwise, find the lowest numbered 1 bit. Let its position be i. The output string consists of N bits, as follows:
Bits 0, 1,... i-1 are 1's. Bit i is 0. Bits i+1,..., N-1 are the same as the corresponding input bits.
Again, let me emphasize that this is not a usable compression method!. The fun is finding the flaw.
A lot of people seem to have the idea that something that would be wrong to do in a leaflet or newspaper or on a street corner with a megaphone is OK if you do it on the internet.
It's rather frightening to realize that there are people who only obey the law or social norms because they are too lazy to do otherwise until internet makes it easy.
I don't have enough moderator points to mod down everyone who has said something stupid so far, so I'll post instead.
1. To all the people who are complaining that WMA is a proprietary format: So is DVD.
2. To the people who say no consumer would want this: go to a store and look at how many portable devices there are now that play MP3 and/or WMA files. As a consumer, I'd be quite pleased if the CDs I burn for my RioVolt would play in my home CD or DVD player. WMA is a better format than MP3, in both quality and space taken. Of course consumers will want it.
3. To the people pointing out that WMA supports rights management: duh!
Of course, if you ask me, the compiler should do reordering, but what should I know[...]
The CPU is working with the instructions that are actually executed. The compiler is working with instructions that might be executed. This should let the CPU do some optimizations that the compiler cannot. On the other hand, the compiler has more time to work on the problem than the CPU, so there should be some optimizations that the compiler can do better.
Seems to me that this means that both the compiler and the CPU should be doing this.
OK, this is interesting. I've been trying to figure out how copy protection could violate the law. Things like "fair use" make it so that certain copying is not a copyright violation, but also don't prevent the copyright holder from taking non-copyright measures to stop that copying. The DMCA makes it pretty clear that the idea of copy protection is perfectly legal.
However, he mentions the AHRA. The interesting part about the AHRA is that it places a tax on certain blank media, and mandated certain copy protection schemes in digital recording hardware. The record companies get the money from the tax. In exchange for this, consumers got some pretty broad music copying rights.
I think the theory he is thinking about is that consumers have bought copying rights via that tax, and so that the record companies can't take steps to stop that copying, since they have accepted the money from that tax.
There's also the tools. Infoworld (I think), in a comparison of CORBA and COM (a couple years ago...things may be different now) concluded that CORBA was a little bit better technically, and COM was a lot easier to use, because Microsoft provided better tools.
The thing about Microsoft is that they often take two or three iterations to get it right, but when they get there, it's right for the programmer. Too many things in the Unix world are right in a theoretical sense, but a complete pain in the ass when you fire up your editor, start the caffeine IV, and try to crank out code.
Microsoft will stick with the "Standard" just like they did with Java, only this time there won't be anyone pig-headed like Sun rejecting their improvements out of spite.
What idiot moderated as "flamebait" the person who said Netscape's problems stemmed from Netscape being a bad browser?
Netscape was a lousy browser. My god, when Microsoft started giving away IE, Netscape would reload the page from the server if you resized the window. Years later, I'm still astounded at how braindead that was.
Before IE was bundled with the OS, you could either get it for free by download, or go to the store and buy it as part of an add-on package, which cost about the same as Netscape. Go back and check the sales figures, and you'll find that this non-free version outsold Netscape, which demolishes the claim that it was IE's freeness that killed Netscape.
What mistakes in TPM? Notice that the best SW movies were SW and TESB. SW was made when Lucas still a man studios could say "no" to and make do things their way. TESB probably was made under fewer constraints, but he still would have had to answer to them.
By ROTJ, Lucas almost certainly had much more control over the movie...and it was the worst of the three.
With TPM, I can't imagine anyone saying no to Lucas for anything. TPM is thus likely to be the closest we've yet seen to Lucas' vision of what Star Wars is.
Yeah, you can get bandwidth cheaper than a T1, by going to much bigger pipes. How big a pipe do you need to get the price down to $0.25 per gig? If the ISP has to aggregate traffic from many areas to get enough for that huge pipe, you have to consider the cost of getting all that traffic to where they have the big pipe. E.g., if they use a T3 to get the traffic from some small town to some regional network center where they have the big pipe, then the cost of bandwidth on a T3 places a lower limit on what it costs the ISP to provide bandwidth to those users.
Adjust prices in the following for whatever things cost in your part of the world. Let's compute a ballpark figure for the cost of bandwidth from the cable company to the internet.
180 Kbyte/second data transfer rate per T1 x 3600 seconds/hour x 24 hours/day x 30.5 days/month = 452 gigabytes/month. So, they need a T1 for every 452 gigabytes of data their customers try to download each month.
A T1 is about $700/month.
So, ballpark figure, it costs them $1.55 for every gigabyte of data their customers download.
At $40/month, this means that even if they had zero costs other than paying for bandwidth to the internet, they would lose money on any customer who downloads more than 25 gigs a month.
Why are you comparing to brand name PCs? In the context of Mac vs. Linux, we have to presume that the user is fairly sophisticed technically, and so has no problem buying a case, motherboard, CPU, etc., and spending a half hour assembling it themselves.
Given that assumption, at any given level of performance on Unix-like things, you can beat any given Mac at a much lower cost.
MacOS X looks nice...but the hardware to run it costs about 3 times as much as comparable hardware to run Linux, and that hardware for Linux can also run Windows.
Slashdot Mirror
You should not give such companies any information that you want them to keep secret.
It's very instructive to watch with a magnifier while playing with the width and height controls, to see how they map the logical pixels to the physical pixels. Do that, and you'll see the real reason that CRT's look good at different resolutions.
Ease of use: obviously, you've never used a bike in a city. Bikes are incredibly awkward as soon as you go indoors with them. Try riding a bike in an elevator, or down a hallway.
How the hell does something whose math is off by over an order of magnitude get moderated up as insightful?
The intent with these large jukeboxes is that you put all your music on them and keep it there. Hence, you only transfer a given song once. The initial loading might take a few hours. After that, you can load new songs faster than you can rip them or download them from the net, so USB is fine for that.
This means that in addition to the content providers wanting to stop a video napster, the cable internet companies will also want very much to squash it. The impression I got with audio naptster was that the bandwidth usage was low enough to not hurt cable performance, and so the cable companies didn't care that much, and so didn't want to get involved.
Video peer-to-peer will have a much harder time catching on, I think...it won't take many cases of ISPs canceling accounts of people who do it to scare people away.
MFC does indeed have some odd characteristics, but most of them are there for good reason, and that book explains them.
However, if you get one of the integrated Tive/DirecTV receivers, and are willing to look around on the net to find some hacks, you can hack it to get the program data from the DirecTV guide. You then don't need to hook it up to a phone, either, which is a plus for some.
I would guess that you could adopt that hack to work with the non-DirecTV units, if you give them some kind of network access. Grab the guide data from yahoo or gist or someplace like that, and write a perl script to convert it to Tivo's format.
I wonder if it is strong enough to take a 19" monitor sitting on top? Under the monitor would be a perfect place for it on my desktop.
Using time travel, high compression of arbitrary data is trivial. Simply record the location (in both space and time) of the computer with the data, and the name of the file, and then replace the file with a note saying when and where it existed. To decompress, you just pop back in time and space to before the time of the deletion and copy the file.
Here's a proposal for a compression scheme that has the following properties:
1. It works on all bit strings of more than one bit.
2. It is lossless and reversible.
3. It never makes the string larger. There are some strings that don't get smaller, but see item #4.
4. You can iterate it, to reduce any string down to 1 bit! You can use this to deal with pesky strings that don't get smaller. After enough iterations, they will be compressed.
OK, here's my algorithm:
Input: a string of N bits, numbered 0 to N-1.
If all N bits are 0, the output is a string of N-1 1's. Otherwise, find the lowest numbered 1 bit. Let its position be i. The output string consists of N bits, as follows:
Bits 0, 1, ... i-1 are 1's. Bit i is 0. Bits i+1, ..., N-1 are the same as the corresponding input bits.
Again, let me emphasize that this is not a usable compression method!. The fun is finding the flaw.
It's rather frightening to realize that there are people who only obey the law or social norms because they are too lazy to do otherwise until internet makes it easy.
1. To all the people who are complaining that WMA is a proprietary format: So is DVD.
2. To the people who say no consumer would want this: go to a store and look at how many portable devices there are now that play MP3 and/or WMA files. As a consumer, I'd be quite pleased if the CDs I burn for my RioVolt would play in my home CD or DVD player. WMA is a better format than MP3, in both quality and space taken. Of course consumers will want it.
3. To the people pointing out that WMA supports rights management: duh!
The CPU is working with the instructions that are actually executed. The compiler is working with instructions that might be executed. This should let the CPU do some optimizations that the compiler cannot. On the other hand, the compiler has more time to work on the problem than the CPU, so there should be some optimizations that the compiler can do better.
Seems to me that this means that both the compiler and the CPU should be doing this.
However, he mentions the AHRA. The interesting part about the AHRA is that it places a tax on certain blank media, and mandated certain copy protection schemes in digital recording hardware. The record companies get the money from the tax. In exchange for this, consumers got some pretty broad music copying rights.
I think the theory he is thinking about is that consumers have bought copying rights via that tax, and so that the record companies can't take steps to stop that copying, since they have accepted the money from that tax.
The thing about Microsoft is that they often take two or three iterations to get it right, but when they get there, it's right for the programmer. Too many things in the Unix world are right in a theoretical sense, but a complete pain in the ass when you fire up your editor, start the caffeine IV, and try to crank out code.
Microsoft will stick with the "Standard" just like they did with Java, only this time there won't be anyone pig-headed like Sun rejecting their improvements out of spite.
Netscape was a lousy browser. My god, when Microsoft started giving away IE, Netscape would reload the page from the server if you resized the window. Years later, I'm still astounded at how braindead that was.
Before IE was bundled with the OS, you could either get it for free by download, or go to the store and buy it as part of an add-on package, which cost about the same as Netscape. Go back and check the sales figures, and you'll find that this non-free version outsold Netscape, which demolishes the claim that it was IE's freeness that killed Netscape.
No, they are assuming that those people who have jobs can afford $25 for a portable CD player to use at work.
By ROTJ, Lucas almost certainly had much more control over the movie...and it was the worst of the three.
With TPM, I can't imagine anyone saying no to Lucas for anything. TPM is thus likely to be the closest we've yet seen to Lucas' vision of what Star Wars is.
I fear what the next film could be.
So?
Yeah, you can get bandwidth cheaper than a T1, by going to much bigger pipes. How big a pipe do you need to get the price down to $0.25 per gig? If the ISP has to aggregate traffic from many areas to get enough for that huge pipe, you have to consider the cost of getting all that traffic to where they have the big pipe. E.g., if they use a T3 to get the traffic from some small town to some regional network center where they have the big pipe, then the cost of bandwidth on a T3 places a lower limit on what it costs the ISP to provide bandwidth to those users.
180 Kbyte/second data transfer rate per T1 x 3600 seconds/hour x 24 hours/day x 30.5 days/month = 452 gigabytes/month. So, they need a T1 for every 452 gigabytes of data their customers try to download each month.
A T1 is about $700/month.
So, ballpark figure, it costs them $1.55 for every gigabyte of data their customers download.
At $40/month, this means that even if they had zero costs other than paying for bandwidth to the internet, they would lose money on any customer who downloads more than 25 gigs a month.
TANSTAAFL applies to internet bandwidth.
Given that assumption, at any given level of performance on Unix-like things, you can beat any given Mac at a much lower cost.
MacOS X looks nice...but the hardware to run it costs about 3 times as much as comparable hardware to run Linux, and that hardware for Linux can also run Windows.