"But google/kaaza/whatever are all cohesive. You're computer knows the IPs of several other computers connected to you"
Let's take these one at a time.
As with any web site, Google knows the IP addresses of anyone using the site. So they can know who searches for pr0n or mp3's, etc. But they can't know whether you actually went to the site they provided an URL to, and performed a download, as that's between your web browser and the file server.
KaZaA knows the IP addresses of the computers you're connected to. But it's hard to know who's doing what on the entire network, because it's so large and fairly well distributed. Of course, if a musician searches for their copyrighted song, they can see the IP's of everyone who broadcasts it, so running KaZaA isn't too clever. At least the odds of any _one_ search finding you is small.
Each BitTorrent torrent is perfectly centrally coordinated. So if I see an illegal file being served using BitTorrent, and I want to know who's downloading or serving it, I can easily know the IP's of everyone downloading or serving the file.
All I have to do is:
1) Download the torrent file from the web site. Of course, I know the web site's address, DNS records, etc. One down.
2) Run TorrentSpy (or a good BT client) to see the tracker for the file. This I know the IP address of the guy running the tracker that coordinates the downloads. Two down.
3) I run a BT client and open the Torrent file. This initiates a connection to the tracker, which then tells me the IP addresses of all of the other people uploading the file, as well as their download status and all sorts of other interesting things. Three through 2,500 down.:-)
And if I get bored, I can write a program to watch the popular BT web sites for my copyrighted material.
"BitTorrent is still better for piracy because the men in black need to know if your torrent is illicit in order to know if they should log your ip. The are millions of torrents out there. they cant track them all."
It's not hard to watch the popular BT web sites for a list of artist and album names. BTSearch makes it even easier. Putting RSS on Trackers makes it trivially easy. So yes, if people had to watch all those sites, it'd be a lot of work. Computers, however, are pretty good at automating repetitive tasks.
So yes, I suppose that you can't quite assume that "TheBeatlesCompleteCollection.torrent" contains copyrighted material. But you can capture all of the IP's and start a download, then after you verify that the torrent file does contain copyrighted music, issue Cease & Desist letters to everyone.
"And THAT is why this is unsuitable for your average user. They want to download it and expect it to work, regardless of how many other people are sharing it at the time"
For large files being legitimately distributed, this isn't an issue, as the publisher can (and obviously should) leave up a permanent 'seed' server.
"When's the last time you had a large file you were getting from an FTP site just disappear mid-download..."
I'd say that if you're running an FTP site, the BitTorrent equivalent is running a Tracker and a seed. So the file is always at least as available as via FTP, but if the file is popular it gets more availble.
There's little info on the web site, so if anyone knows what this 'protocol' really is, I'd love to know how this proposal is any better than what's already available? For example, all Drupal web sites (www.drupal.org) support a shared login scheme, so if you have a 'drupal distributed authentication' you can log into any drupal site (that choses to support drupal login) with one signon. Even better, once you've created an account on any drupal site, that site can serve as your authentication to any other site, so you can choose who you want to trust. So if I log in as 'joe@remote.delphiforums.com' and my Delphi Forums password, the drupal site will check with Delphi Forums to validate the password, then create the account and log me in. This capability has been in Drupal for _years_.
"torrents should start to be paired with PAR files to create a far more robust method of fetching large files"
This doesn't make any sense. Torrents are completely reliable -- they already have block and file level hashing and automatic re-downloading of blocks in case of transmission errors, etc. The only time you won't get a complete torrent is if there are no complete copies of the file being served. Adding error correcting codes (e.g. PAR files) would make the total file larger, and only recover from incomplete torrents that are _almost_ complete (i.e. would have been complete if the PAR file hadn't made it 15% larger). Just make sure that anything you're downloading has a couple of seeds before starting the download.;-)
I can't say that it's "do to the seeding algorithms" but it's true that there's no encryption or hiding in BitTorrent -- it's pretty fundamental to the protocol's efficiency that everyone downloading a given torrent is given everyone else's IP address so that they can exchange data. This is why BitTorrent is great for moving large _legitimate_ files, and not so clever to use for "piracy". You might as well wear a red shirt on (original) Star Trek.:-)
A) Label pays radio station to play song B) Radio station plays song they're paid to play instead of what the DJ or music director want to play C) Ignorant Masses (ie, You) hear less music, Same ads, and are generally happy.
"Why did atom even come into existance? Was not RSS already established, or is there some kind of deficiancy in RSS that i'm missing here?"
Why did RSS even come into existance? Was not ICE already established? (NB: ICE was in production in 1998, before which it was submitted to the W3C, etc. See the ICE web site for more history.
People create new protocols that duplicate existing ones all the time. Unfortunately. It often comes down to politics and personalities.
That being said, it's possible to overcome this forking and work together, or to at least find areas of cooperation. ICE2, for example, adopts some aspects of RSS -- you can use ICE2 to perform push, incremental delivery of RSS items, for example, which makes things quite a bit more efficient.
I saw the same thing way back when with VAXen -- I remember when my school bought an expensive CPU upgrade, which was "installed" by having the tech reboot the VAX from a (8") floppy disk that loaded new firmware, making the machine 30% faster!
I ran into the same thing with my TI calculator -- I bought the base model, and took it apart and found out that it was identical to the high end model with all the fancy trig functions, etc., except that there were no plastic buttons on the front panel -- the contacts underneath were complete, etc. So I cut a few holes in the front of the calculator and added my own 'buttons'. Worked great!
"Was wondering what the potential was for using Linux on fairly standard PC hardware to run an Oracle server. Is anyone actually using one in a production set up and if so what number of users/size of database/applications are they using."
I've used Oracle on Linux and BSD in production, and it's worked out well. While Oracle's licenses aren't cheap, it doesn't hurt to save on an NT or UNIX license (both are quite expensive if you actually pay for a real internet server license), so it's a meaningful savings. Oracle's low end licenses are actually pretty cheap (a single CPU license of Oracle database cost less than a single CPU unlimited user NT or Solaris OS license, not to mention MS SQL Server, the last time I looked). Of course, Oracle costs a ton once you start buying multi-CPU, Enterprise licenses, but with some ingenuity you can do a surprising amount with Oracle's cheapest database.
But the real reason to run the database using Linux or BSD is consistency -- the last few years it's become standard to run your front-end servers on a free OS, and it's convenient to run the back-end on the same OS. If your team is more productive using Oracle over a free OS than a proprietary OS, it's hard to argue that you should spend the money.
The other scenario where it makes sense to run Oracle on a free OS is at the very high end, where Oracle has some really cool looking clustering tools. If you run a 50 CPU database, you'd hate to have to buy 50 CPU's of UNIX or NT, at a few $K per CPU. (Keep in mind that the XP license you buy in stores can't legally be used to run an internet server -- it has all sorts of license restrictions).
That being said, I certainly have nothing against running Oracle on Solaris (or Digital UNIX) -- I've done both, and been very happy with the performance and scalability. If my company goes under if the database dies, I don't mind spending money on the server OS.
"Microsoft will begin selling a $50 music player that will 'look and feel as good as the iPod' later this year."
Well, the reason that the iPod looks and feels so good is that the iPod has a fast interface (FireWire or USB2), tons of storage (i.e. a hard drive), and that fantastic scroll wheel.
If MS makes a $50 "iPod", that means that it'll have to cost $20 to manufacturer. For $20, it'll be hard to include a $60 hard drive, much less the controls and display, battery, audio circuitry, etc.
The only options I can see for MS to produce a "$50 iPod" is to either
1) produce a horribly limited device (i.e. minimal display, bad controls, minimal storage), or 2) to tie it to a subscription service that subsidizes the player.
I bet they could sell a $299 iPod for $50 if it only played music tied to their Janus DRM, which required you to pay $10 a month forever. They'll argue that all of the MS licensees that sell WMA (i.e. 20% of the digital download market) will provide ever so many more options than Apple's iTMS (that is 80% of the digital download market), and ignore that anybody using MS' DRM is stuck with a bad user experience, and that there are (to put it politely) many other ways to acquire music.
I don't know, but it doesn't appear to be enough to stop people from trying. I get the impression that once an election is certified, nothing will be done to overturn the result so the penalties are financial, which means that they don't matter much compared to being in office for years.
Absentee ballots are a huge source of election fraud.
The best one I've read about goes like this: friendly young person volunteers at old age home. They offer to bring in absentee ballots for the residents so that they don't need to go to the polls (which can be difficult for some older folks). They help people fill out their ballots. They collect the ballots to drop off. They can fill in any "missing" votes. They can lose ballots for the opposition. One hard working volunteer can reportedly generate hundreds of votes for their preferred candidates.
I wouldn't say that Bev "dismisses open source" -- she says that open source by itself doesn't guarantee that you can trust a voting system, and that you still should have a paper trail so that you can audit to detect problems and correct them. But she supports open source voting systems as _also_ being beneficial.
"What I don't get is why non-socialization or bad socialization is considered a valid argument against homeschooling - isn't that the point in the first place?... Socialization is just a pretty word for indoctrination."
This is completely wrong. Socialization is learning how to interact with other people. Home schooled kids may benefit in many ways (if the parents are good at it), but they really lose out on this front -- in a group school you spend a lot of time with other people, being forced to learn how to get along with them. In home schooling, you spend most of your time at home, with your parents and siblings. Yes, good home schoolers try to spend some time in groups (parks, science museums, etc.) but the occasional group outing is still very limited compared to spending all day with tons of other kids.
"You don't have to be a Christian to see the value in homeschooling."
But if this gets widely adopted, it turns into: 1. Aim cameraphone at barcode. 2. Click 'go there' button 3. Camera takes picture, find code, decodes, and launches web browser at URL.
If this is implemented as an app for the Treo 600, it wins.
The system I've seen didn't print out an error -- it lit a light warning that the ballot was an overvote (meaning you voted for multiple candidates, which invalidates the vote) or an undervote (meaning you voted for no candidates, which is a common mistake, though is sometimes intentional so it's not a rejection, just a warning).
This immediate feedback that allows voters to correct errors is why these systems have a much lower error rate than the systems that collect ballots and batch scan them after the polls close.
"And the other 3? Apple should at least point to the relevant advisory."
Apple did. I'll quote more of the knowledge base article:
"* CoreFoundation: Fixes CAN-2004-0428 to improve the handling of an environment variable. Credit to aaron@vtty.com for reporting this issue.
* Apache 2: Fixes CAN-2003-0020, CAN-2004-0113 and CAN-2004-0174 by updating to Apache 2 to version 2.0.49.
* RAdmin: Fixes CAN-2004-0429 to improve the handling of large requests
* AppleFileServer: Fixes CAN-2004-0430 to improve the handling of long passwords. Credit to Dave G. from @stake for reporting this issue.
* IPSec: Fixes CAN-2004-0155 and CAN-2004-0403 to improve the security of VPN tunnels. IPSec in Mac OS X is not vulnerable to CAN-2004-0392."
Admittedly this is listed in the knowledge base article, not in the consumer description of the patch, but it doesn't seem unreasonable that a sysadmin would read the KB article for the patch before installing it.
"While Apple seems to be patching fairly regularly, the last security update (the group of 4) was a little lacking in that it offered no explanations... As I work in IT, I'm often left installing patches with Apple with no clue what they're doing under the hood"
Apple's description of the patch was rather terse (AppleFileServer: Fixes CAN-2004-0430 to improve the handling of long passwords. Credit to Dave G. from @stake for reporting this issue."), but it provides the reference (CAN-2004-0430) that provides full details. Admittedly, this did require a google search, or reading the usual advisory lists. But it's certainly not hidden from anyone who wants the detail.
"Clear Channel declined to play the Dixie Chicks because of their stupid hate speech."
The Dixie Chicks didn't make any hate speech -- they stated their opposition to Bush's invasion of Iraq. That's political speech, the protection of which is central to our Constitution.
"remaining 92%"
Clear Channel owns over 1,200 radio stations and 37 television stations, with investments in 240 radio stations globally, and Clear Channel Entertainment (aka SFX, one of their more well-known subsidiaries) owns and operates over 200 venues nationwide. They are in 248 of the top 250 radio markets, controlling 60% of all rock programming. They outright own the tours of musicians like Janet Jackson, Aerosmith, Pearl Jam, Madonna and N'Sync. They own the network which airs Rush Limbaugh, Dr. Laura, Casey Kasem, and the Fox Sports Radio Network. With 103,000,000 listeners in the U.S. and 1,000,000,000 globally (1/6 of the world population), this powerful company has grown unchecked, using their monopoly to control the entire music industry.
(from http://www.clearchannelsucks.org/)
And, relevant to this discussion, they've got their own company to sell live recorded CD's of concerts (i.e. they copied the independent startup that implemented it first) and the ClearChannel company has the exclusive deal with all ClearChannel venues. How do you think the other company will do?
It also solves two pragmatic problems: - How do you sell music to people who don't carry RAM cards when going to bars to hear bands play? - How do you deal with all of the different card formats? A Kiosk with slots for CF, SD/MMC, USB, FireWire iPods, new iPods, etc., will be unusable.
"I'm no expert on the matter but if I buy a CD from Russia that's $15 cheaper than in my country, do I have to pay again to the local RIAA just because I got it cheaper?... If I buy legal online music from Russia instead say.. iTunes, does it make the songs illegal?"
It's not a matter of pricing being lower in Russia; the site is pretty obviously illegal. Many of the artists whose work is being sold (e.g. The Beatles, Led Zeppelin) have never agreed to have their work sold digitally, which is why you can't find their music on any legitimate music download service. And of course, if they're selling music for a few cents a track, they're not paying the artists and composers for the recordings.
Even if it were legal in Russia (which it clearly isn't) they clearly wouldn't have the right to sell that music anywhere else, since the companies that have the Russian rights to the music aren't the same as the companies that have the US rights, and those rights are exclusive. So even if the Russian company decided to sell music for almost nothing, they couldn't sell it to US customers, etc. This legal issue is why iTunes only sells to US customers, etc. -- to do things legally, you have to negotiate the rights to sell the music country by country.
"Great! So instead of printing a copy of the classics downloaded from the internet at a few pennies per copy, my child can now use a $1350 laptop"
It seems to me that 'gillbates' is complaining about hypothetical problems, and ignoring how well laptops are actually working out in schools. If I had moderator points (just finished voting a few minutes ago, darn it) I'd mod him down. Instead I'll post.:-)
$1,350 is certainly more than the cost of the textbooks ($350), but not much, and probaly less than the cost of buying 2,000 textbooks that come loaded on the laptops; what's the value in giving the students access to more learning material? Compared to the amount spent per student in most school systems ($thousands) $1,000 is a minor cost if the result is (as it appears to be in the school systems that use latops like this) improved attendence and increased student participation.
Also, in actual practice, kids appear to keep pretty good care of their laptops. On top of this, the machines are purchased with a support contract so the vendor keeps them in working order, provides spares, etc. If kids "hawk their textbooks for drug money" they're responsible for them, so the parents get to pay for it. This appears to be sufficient incentive, since it hasn't been an issue over the last few years that schools systems have been issuing laptops to kids.
"these students won't learn any computer science during this program"
Computers aren't tools for teaching computer science. They're general purpose learning tools -- people don't have to learn CS to research on the internet, to have access to far more textbooks than they could carry, etc. So if they learn english, history, biology, etc., better because they've got a laptop (and this appears to be what happens in schools that are doing this), it looks like a great deal.
"Computers don't teach logic or reason - if they did, a substantial portion of the population would not be making a living teaching inherently stupid machines to perform monotonous tasks."
This is incorrect, except in its simplest form. A computer by itself doesn't teach anything, because it's just a box full of chips sitting on a table. But a computer, loaded with the right software, and used as a tool by a good teacher and eager student, is a wonderful enabler. For example, back when I taught kids programming using Logo they certainly _did_ learn logic and reason, in a way that was simply not possible without a computer.
Of course, computers are also great for teaching writing. It's amazing when kids get access to a computer -- their creativity is unleashed once they don't have to deal with the logistics of writing. For a little kid, it's intimidating knowing that a mistake can cost minutes of painful erasing and rewriting, but using even a simple word processor they can write fearlessly, knowing that they can always erase, rewrite, print out, revise, etc.
So yes, many programmers make a living teaching stupid computers to do interesting things. But many, many more people get to use the resulting software that lets them do creative things like write, edit video, do painting, research, and so on. And that's the value that computers bring.
Well, when there's a will, there's a way. You can use QTConvert to batch convert between any two known QuickTime formats, including from Protected AAC to AAC, Ogg Vorbis, etc. And it comes with scripts that run from within iTunes so that you can select tracks within iTunes and convert them.
"While you're happy to have the choice made for you now, that might not be the case in the future. Could be tricky moving your entire music collection over to another format."
For iTMS, converting to another format is quite easy -- burn to CD (or virtual CD) and RIP into other format.
Yes, converting between lossy formats does cost you some sound quality, but:
- you can't hear it under normal listening conditions, and - there are hacks that let you strip the DRM and leave music in AAC, and
Re:Just found Cringely's article on that, too.
on
Open Voting at OSCON
·
· Score: 1
"There we have it: Yes, we COULD use the Diebold machines (or probably any other brands, since the printer is federally mandated) with the open source software and NO additional hardware (except eventually the cover-with-a-slot), obtaining the printed audit trail we want."
This is a good "stop gap" measure so that counties that already bought DRE's could improve them a bit, though I think that when you get into the details it's not what you would really want to use to vote on.
For example, since the record would be printed after the vote is cast, there's no way to do anything about it if the printed vote isn't correct. They can't remove the vote from the machine, and they can't let the voter cast another vote.
And there's no way to know that the vote recorded on the memory card is the same as the vote printed.
But if a half-step is all that some county can afford, at least they'll be able to have some basic capability to audit.
Slashdot Mirror
"But google/kaaza/whatever are all cohesive. You're computer knows the IPs of several other computers connected to you"
:-)
Let's take these one at a time.
As with any web site, Google knows the IP addresses of anyone using the site. So they can know who searches for pr0n or mp3's, etc. But they can't know whether you actually went to the site they provided an URL to, and performed a download, as that's between your web browser and the file server.
KaZaA knows the IP addresses of the computers you're connected to. But it's hard to know who's doing what on the entire network, because it's so large and fairly well distributed. Of course, if a musician searches for their copyrighted song, they can see the IP's of everyone who broadcasts it, so running KaZaA isn't too clever. At least the odds of any _one_ search finding you is small.
Each BitTorrent torrent is perfectly centrally coordinated. So if I see an illegal file being served using BitTorrent, and I want to know who's downloading or serving it, I can easily know the IP's of everyone downloading or serving the file.
All I have to do is:
1) Download the torrent file from the web site. Of course, I know the web site's address, DNS records, etc. One down.
2) Run TorrentSpy (or a good BT client) to see the tracker for the file. This I know the IP address of the guy running the tracker that coordinates the downloads. Two down.
3) I run a BT client and open the Torrent file. This initiates a connection to the tracker, which then tells me the IP addresses of all of the other people uploading the file, as well as their download status and all sorts of other interesting things. Three through 2,500 down.
And if I get bored, I can write a program to watch the popular BT web sites for my copyrighted material.
"BitTorrent is still better for piracy because the men in black need to know if your torrent is illicit in order to know if they should log your ip. The are millions of torrents out there. they cant track them all."
It's not hard to watch the popular BT web sites for a list of artist and album names. BTSearch makes it even easier. Putting RSS on Trackers makes it trivially easy. So yes, if people had to watch all those sites, it'd be a lot of work. Computers, however, are pretty good at automating repetitive tasks.
So yes, I suppose that you can't quite assume that "TheBeatlesCompleteCollection.torrent" contains copyrighted material. But you can capture all of the IP's and start a download, then after you verify that the torrent file does contain copyrighted music, issue Cease & Desist letters to everyone.
Isn't networking fun?
"And THAT is why this is unsuitable for your average user. They want to download it and expect it to work, regardless of how many other people are sharing it at the time"
For large files being legitimately distributed, this isn't an issue, as the publisher can (and obviously should) leave up a permanent 'seed' server.
"When's the last time you had a large file you were getting from an FTP site just disappear mid-download..."
I'd say that if you're running an FTP site, the BitTorrent equivalent is running a Tracker and a seed. So the file is always at least as available as via FTP, but if the file is popular it gets more availble.
There's little info on the web site, so if anyone knows what this 'protocol' really is, I'd love to know how this proposal is any better than what's already available? For example, all Drupal web sites (www.drupal.org) support a shared login scheme, so if you have a 'drupal distributed authentication' you can log into any drupal site (that choses to support drupal login) with one signon. Even better, once you've created an account on any drupal site, that site can serve as your authentication to any other site, so you can choose who you want to trust. So if I log in as 'joe@remote.delphiforums.com' and my Delphi Forums password, the drupal site will check with Delphi Forums to validate the password, then create the account and log me in. This capability has been in Drupal for _years_.
http://drupal.org/node/view/312#6790
There's also SharedID.
"torrents should start to be paired with PAR files to create a far more robust method of fetching large files"
;-)
This doesn't make any sense. Torrents are completely reliable -- they already have block and file level hashing and automatic re-downloading of blocks in case of transmission errors, etc. The only time you won't get a complete torrent is if there are no complete copies of the file being served. Adding error correcting codes (e.g. PAR files) would make the total file larger, and only recover from incomplete torrents that are _almost_ complete (i.e. would have been complete if the PAR file hadn't made it 15% larger). Just make sure that anything you're downloading has a couple of seeds before starting the download.
I can't say that it's "do to the seeding algorithms" but it's true that there's no encryption or hiding in BitTorrent -- it's pretty fundamental to the protocol's efficiency that everyone downloading a given torrent is given everyone else's IP address so that they can exchange data. This is why BitTorrent is great for moving large _legitimate_ files, and not so clever to use for "piracy". You might as well wear a red shirt on (original) Star Trek. :-)
You got the steps wrong:
A) Label pays radio station to play song
B) Radio station plays song they're paid to play instead of what the DJ or music director want to play
C) Ignorant Masses (ie, You) hear less music, Same ads, and are generally happy.
Why did RSS even come into existance? Was not ICE already established? (NB: ICE was in production in 1998, before which it was submitted to the W3C, etc. See the ICE web site for more history.
People create new protocols that duplicate existing ones all the time. Unfortunately. It often comes down to politics and personalities.
That being said, it's possible to overcome this forking and work together, or to at least find areas of cooperation. ICE2, for example, adopts some aspects of RSS -- you can use ICE2 to perform push, incremental delivery of RSS items, for example, which makes things quite a bit more efficient.
I saw the same thing way back when with VAXen -- I remember when my school bought an expensive CPU upgrade, which was "installed" by having the tech reboot the VAX from a (8") floppy disk that loaded new firmware, making the machine 30% faster!
I ran into the same thing with my TI calculator -- I bought the base model, and took it apart and found out that it was identical to the high end model with all the fancy trig functions, etc., except that there were no plastic buttons on the front panel -- the contacts underneath were complete, etc. So I cut a few holes in the front of the calculator and added my own 'buttons'. Worked great!
"Was wondering what the potential was for using Linux on fairly standard PC hardware to run an Oracle server. Is anyone actually using one in a
production set up and if so what number of users/size of database/applications are they using."
I've used Oracle on Linux and BSD in production, and it's worked out well. While Oracle's licenses aren't cheap, it doesn't hurt to save on an NT or UNIX license (both are quite expensive if you actually pay for a real internet server license), so it's a meaningful savings. Oracle's low end licenses are actually pretty cheap (a single CPU license of Oracle database cost less than a single CPU unlimited user NT or Solaris OS license, not to mention MS SQL Server, the last time I looked). Of course, Oracle costs a ton once you start buying multi-CPU, Enterprise licenses, but with some ingenuity you can do a surprising amount with Oracle's cheapest database.
But the real reason to run the database using Linux or BSD is consistency -- the last few years it's become standard to run your front-end servers on a free OS, and it's convenient to run the back-end on the same OS. If your team is more productive using Oracle over a free OS than a proprietary OS, it's hard to argue that you should spend the money.
The other scenario where it makes sense to run Oracle on a free OS is at the very high end, where Oracle has some really cool looking clustering tools. If you run a 50 CPU database, you'd hate to have to buy 50 CPU's of UNIX or NT, at a few $K per CPU. (Keep in mind that the XP license you buy in stores can't legally be used to run an internet server -- it has all sorts of license restrictions).
That being said, I certainly have nothing against running Oracle on Solaris (or Digital UNIX) -- I've done both, and been very happy with the performance and scalability. If my company goes under if the database dies, I don't mind spending money on the server OS.
"Microsoft will begin selling a $50 music player that will 'look and feel as good as the iPod' later this year."
Well, the reason that the iPod looks and feels so good is that the iPod has a fast interface (FireWire or USB2), tons of storage (i.e. a hard drive), and that fantastic scroll wheel.
If MS makes a $50 "iPod", that means that it'll have to cost $20 to manufacturer. For $20, it'll be hard to include a $60 hard drive, much less the controls and display, battery, audio circuitry, etc.
The only options I can see for MS to produce a "$50 iPod" is to either
1) produce a horribly limited device (i.e. minimal display, bad controls, minimal storage), or
2) to tie it to a subscription service that subsidizes the player.
I bet they could sell a $299 iPod for $50 if it only played music tied to their Janus DRM, which required you to pay $10 a month forever. They'll argue that all of the MS licensees that sell WMA (i.e. 20% of the digital download market) will provide ever so many more options than Apple's iTMS (that is 80% of the digital download market), and ignore that anybody using MS' DRM is stuck with a bad user experience, and that there are (to put it politely) many other ways to acquire music.
I don't know, but it doesn't appear to be enough to stop people from trying. I get the impression that once an election is certified, nothing will be done to overturn the result so the penalties are financial, which means that they don't matter much compared to being in office for years.
But perhaps someone knows for sure?
Absentee ballots are a huge source of election fraud.
The best one I've read about goes like this: friendly young person volunteers at old age home. They offer to bring in absentee ballots for the residents so that they don't need to go to the polls (which can be difficult for some older folks). They help people fill out their ballots. They collect the ballots to drop off. They can fill in any "missing" votes. They can lose ballots for the opposition. One hard working volunteer can reportedly generate hundreds of votes for their preferred candidates.
I wouldn't say that Bev "dismisses open source" -- she says that open source by itself doesn't guarantee that you can trust a voting system, and that you still should have a paper trail so that you can audit to detect problems and correct them. But she supports open source voting systems as _also_ being beneficial.
"What I don't get is why non-socialization or bad socialization is considered a valid argument against homeschooling - isn't that the point in the first place? ... Socialization is just a pretty word for indoctrination."
This is completely wrong. Socialization is learning how to interact with other people. Home schooled kids may benefit in many ways (if the parents are good at it), but they really lose out on this front -- in a group school you spend a lot of time with other people, being forced to learn how to get along with them. In home schooling, you spend most of your time at home, with your parents and siblings. Yes, good home schoolers try to spend some time in groups (parks, science museums, etc.) but the occasional group outing is still very limited compared to spending all day with tons of other kids.
"You don't have to be a Christian to see the value in homeschooling."
True, you can be a Christian _or_ a Libertarian!
But if this gets widely adopted, it turns into:
1. Aim cameraphone at barcode.
2. Click 'go there' button
3. Camera takes picture, find code, decodes, and launches web browser at URL.
If this is implemented as an app for the Treo 600, it wins.
The system I've seen didn't print out an error -- it lit a light warning that the ballot was an overvote (meaning you voted for multiple candidates, which invalidates the vote) or an undervote (meaning you voted for no candidates, which is a common mistake, though is sometimes intentional so it's not a rejection, just a warning).
This immediate feedback that allows voters to correct errors is why these systems have a much lower error rate than the systems that collect ballots and batch scan them after the polls close.
"And the other 3? Apple should at least point to the relevant advisory."
Apple did. I'll quote more of the knowledge base article:
"* CoreFoundation: Fixes CAN-2004-0428 to improve the handling of an environment variable. Credit to aaron@vtty.com for reporting this issue.
* Apache 2: Fixes CAN-2003-0020, CAN-2004-0113 and CAN-2004-0174 by updating to Apache 2 to version 2.0.49.
* RAdmin: Fixes CAN-2004-0429 to improve the handling of large requests
* AppleFileServer: Fixes CAN-2004-0430 to improve the handling of long passwords. Credit to Dave G. from @stake for reporting this issue.
* IPSec: Fixes CAN-2004-0155 and CAN-2004-0403 to improve the security of VPN tunnels. IPSec in Mac OS X is not vulnerable to CAN-2004-0392."
Admittedly this is listed in the knowledge base article, not in the consumer description of the patch, but it doesn't seem unreasonable that a sysadmin would read the KB article for the patch before installing it.
"While Apple seems to be patching fairly regularly, the last security update (the group of 4) was a little lacking in that it offered no explanations ... As I work in IT, I'm often left installing patches with Apple with no clue what they're doing under the hood"
Apple's description of the patch was rather terse (AppleFileServer: Fixes CAN-2004-0430 to improve the handling of long passwords. Credit to Dave G. from @stake for reporting this issue."), but it provides the reference (CAN-2004-0430) that provides full details. Admittedly, this did require a google search, or reading the usual advisory lists. But it's certainly not hidden from anyone who wants the detail.
"Clear Channel declined to play the Dixie Chicks because of their stupid hate speech."
The Dixie Chicks didn't make any hate speech -- they stated their opposition to Bush's invasion of Iraq. That's political speech, the protection of which is central to our Constitution.
"remaining 92%"
Clear Channel owns over 1,200 radio stations and 37 television stations, with investments in 240 radio stations globally, and Clear Channel Entertainment (aka SFX, one of their more well-known subsidiaries) owns and operates over 200 venues nationwide. They are in 248 of the top 250 radio markets, controlling 60% of all rock programming. They outright own the tours of musicians like Janet Jackson, Aerosmith, Pearl Jam, Madonna and N'Sync. They own the network which airs Rush Limbaugh, Dr. Laura, Casey Kasem, and the Fox Sports Radio Network. With 103,000,000 listeners in the U.S. and 1,000,000,000 globally (1/6 of the world population), this powerful company has grown unchecked, using their monopoly to control the entire music industry.
(from http://www.clearchannelsucks.org/)
And, relevant to this discussion, they've got their own company to sell live recorded CD's of concerts (i.e. they copied the independent startup that implemented it first) and the ClearChannel company has the exclusive deal with all ClearChannel venues. How do you think the other company will do?
It also solves two pragmatic problems:
- How do you sell music to people who don't carry RAM cards when going to bars to hear bands play?
- How do you deal with all of the different card formats? A Kiosk with slots for CF, SD/MMC, USB, FireWire iPods, new iPods, etc., will be unusable.
"I'm no expert on the matter but if I buy a CD from Russia that's $15 cheaper than in my country, do I have to pay again to the local RIAA just because I got it cheaper? ... If I buy legal online music from Russia instead say.. iTunes, does it make the songs illegal?"
It's not a matter of pricing being lower in Russia; the site is pretty obviously illegal. Many of the artists whose work is being sold (e.g. The Beatles, Led Zeppelin) have never agreed to have their work sold digitally, which is why you can't find their music on any legitimate music download service. And of course, if they're selling music for a few cents a track, they're not paying the artists and composers for the recordings.
Even if it were legal in Russia (which it clearly isn't) they clearly wouldn't have the right to sell that music anywhere else, since the companies that have the Russian rights to the music aren't the same as the companies that have the US rights, and those rights are exclusive. So even if the Russian company decided to sell music for almost nothing, they couldn't sell it to US customers, etc. This legal issue is why iTunes only sells to US customers, etc. -- to do things legally, you have to negotiate the rights to sell the music country by country.
"Great! So instead of printing a copy of the classics downloaded from the internet at a few pennies per copy, my child can now use a $1350 laptop"
:-)
It seems to me that 'gillbates' is complaining about hypothetical problems, and ignoring how well laptops are actually working out in schools. If I had moderator points (just finished voting a few minutes ago, darn it) I'd mod him down. Instead I'll post.
$1,350 is certainly more than the cost of the textbooks ($350), but not much, and probaly less than the cost of buying 2,000 textbooks that come loaded on the laptops; what's the value in giving the students access to more learning material? Compared to the amount spent per student in most school systems ($thousands) $1,000 is a minor cost if the result is (as it appears to be in the school systems that use latops like this) improved attendence and increased student participation.
Also, in actual practice, kids appear to keep pretty good care of their laptops. On top of this, the machines are purchased with a support contract so the vendor keeps them in working order, provides spares, etc. If kids "hawk their textbooks for drug money" they're responsible for them, so the parents get to pay for it. This appears to be sufficient incentive, since it hasn't been an issue over the last few years that schools systems have been issuing laptops to kids.
"these students won't learn any computer science during this program"
Computers aren't tools for teaching computer science. They're general purpose learning tools -- people don't have to learn CS to research on the internet, to have access to far more textbooks than they could carry, etc. So if they learn english, history, biology, etc., better because they've got a laptop (and this appears to be what happens in schools that are doing this), it looks like a great deal.
"Computers don't teach logic or reason - if they did, a substantial portion of the population would not be making a living teaching inherently stupid machines to perform monotonous tasks."
This is incorrect, except in its simplest form. A computer by itself doesn't teach anything, because it's just a box full of chips sitting on a table. But a computer, loaded with the right software, and used as a tool by a good teacher and eager student, is a wonderful enabler. For example, back when I taught kids programming using Logo they certainly _did_ learn logic and reason, in a way that was simply not possible without a computer.
Of course, computers are also great for teaching writing. It's amazing when kids get access to a computer -- their creativity is unleashed once they don't have to deal with the logistics of writing. For a little kid, it's intimidating knowing that a mistake can cost minutes of painful erasing and rewriting, but using even a simple word processor they can write fearlessly, knowing that they can always erase, rewrite, print out, revise, etc.
So yes, many programmers make a living teaching stupid computers to do interesting things. But many, many more people get to use the resulting software that lets them do creative things like write, edit video, do painting, research, and so on. And that's the value that computers bring.
Well, when there's a will, there's a way. You can use QTConvert to batch convert between any two known QuickTime formats, including from Protected AAC to AAC, Ogg Vorbis, etc. And it comes with scripts that run from within iTunes so that you can select tracks within iTunes and convert them.
"While you're happy to have the choice made for you now, that might not be the case in the future. Could be tricky moving your entire music collection over to another format."
For iTMS, converting to another format is quite easy -- burn to CD (or virtual CD) and RIP into other format.
Yes, converting between lossy formats does cost you some sound quality, but:
- you can't hear it under normal listening conditions, and
- there are hacks that let you strip the DRM and leave music in AAC, and
"There we have it: Yes, we COULD use the Diebold machines (or probably any other brands, since the printer is federally mandated) with the open source software and NO additional hardware (except eventually the cover-with-a-slot), obtaining the printed audit trail we want."
This is a good "stop gap" measure so that counties that already bought DRE's could improve them a bit, though I think that when you get into the details it's not what you would really want to use to vote on.
For example, since the record would be printed after the vote is cast, there's no way to do anything about it if the printed vote isn't correct. They can't remove the vote from the machine, and they can't let the voter cast another vote.
And there's no way to know that the vote recorded on the memory card is the same as the vote printed.
But if a half-step is all that some county can afford, at least they'll be able to have some basic capability to audit.