I have set up a set of chained playlists that works roughly like this:
5star - rating is 5 stars, last played > 24 hours ago 4star - rating is 4 stars, last played > 4 days ago 3star - rating is 3 stars, last played > 2 weeks ago unrated - rating is 0 stars, last played > 6 weeks ago 2star - rating is 2 stars, last played > 12 weeks ago
weighted - playlist is 5star/4star/3star/2star/unrated
Feeding "weighted" into Party Shuffle doesn't seem to produce duplicates and I control what the ratings mean. One star songs don't ever show up.
And in the X Window System you hold down "compose" and enter "|" and "c". Most other extended characters have similarlyl easy-to-remember bindings. And people say UNIX/Linux isn't user-friendly.
But the question was is the availability of ICC the start of cross-platform development?
And the answer is "no".
The start of cross-platform development was back in the '60s with Fortran, a language that - however horrible in general design and in all its ghastly details - was sufficinetly powerful and well standardised to allow people to write working programs that would run on virtually every platform.
It soon became the norm for vendors to make damn sure that developers could easily distinguish portable code from extensions. The current environment where Microsoft's business model depends on people getting enmeshed in a twisty maze of different extensions is a horrible throwback to, well, no... it was never that screwed up even when it was IBM and the seven dwarfs.
If the message is unsolicited and trying to sell you something, you are entitled to assume it's spam, and if I get a message like that I'll assume it's spam too.
What I'm getting at, though, is that if it's spam, if that message hadn't been sent in the hundreds of thousands... it wouldn't have been sent at all. There's no profit in sending ONE unsolicited advertisement and stopping. What makes it happen is that it's so cheap to send advertisements in bulk email, and that the spammer doesn't have to wait for people to ask for it.
Without bulk and without unsolicited it's a problem. Without the sales... it's still a problem. There's lots of non-commercial spam -- religious spam, political spam, hoaxes and jokes -- and spam would still be a problem (albeit a much much smaller one) without money involved. Without bulk, though, it wouldn't exist.
That's what I mean when I say "one message isn't spam". One message may be damned good evidence of spam, but the bulkness is an essential part of what makes it happen and what makes it a problem.
if everyone in China sends one message to you [...]
If everyone in China is doing it, then someone is coordinating them. Whever is doing it is the person responsible for sending the messages, the individuals who actually sent them are just proxies. So, that person has sent however many billions of messages that is.
Well, there goes your email account
Nah, I already drop SMTP from China at the IP level.
No, I'm not kidding. If you're in a netblock that's been identified as being in China, Korea, Argentina, any dynamic IP space, or several ISPs like telefonica.es... you can't send me mail.
I literally can't afford the bandwidth to read the messages to spam-filter them any more.
The trick is that the software is it's own validation signature. You don't need to prevent the user from running modified software, what you need to do is prevent the user from being able to read the encrypted DRM files with modified software.
You just restated what I said in different words.
It doesn't use any data to prove it hasn't been tampered with.
Sure it does. It uses itself.
The way it works is that the chip uses the hash of the software to generate a crypto key.
How does it get the hash of the software?
By having the OS or the application read or map the image of the application and calculate the hash from that.
If you modify the OS so that when it's perfoming those specific operations it reads a "shadow" copy of the application that hasn't been modified, it will generate the same key that it would if it hasn't been modified. This is a straight analogy to a standard technique in "rootkits", by the way, and you can download examples from any skr1pt k1ddie website.
So the next step is that the OS itself has to be validated the same way, and then any component that's running in privileged mode, so none of them can modify the checksum calculation.
The only alternatives are obfuscation or simply not allowing the user to modify any component of the system that can get between the application, the disk, and the CPU.
The only secret is the key, which is how most security is usually done. ITunes isn't open source, but the DRM is not obfuscated in any way.
If the DRM isn't obfuscated, there's nothing stopping you from reading the key, reading the encrypted file, and writing it out without eny encryption. The only thing keeping you from reading the key is that the software hides it from you... and if the software is open source, it can't hide it.
That's where TPM comes in. TPM establishes an environment that is unmodifiable from the application back to the chip on the motherboard containing the key. And to make that work you need to prevent the user from modifying the OS, the application, or any drivers the application (or the OS on its behalf) might need to use to read data it uses to checksum itself or however else it's supposed to let the chip know it's not been tampered with.
Which is a pretty tough trick, when your software is open-source.
Now, the interesting part comes when you start to move over to a new OS.
That's why you shuld start doing that first.
For your server apps, like Postfix, start setting them up on a server before you quit using Windows as your desktop. Figure out the way the OS works on a system where you don't need much of an environment to get in and use it... not one where you're trying to maintain a whole window system and everything.
Open source can't support strong DRM, because strong DRM depends on you being unable to modify the OS.
You could build a closed source derivitive of BSD and include DRM components in it, but strong DRM in a GPL kernel violates clause 6 of the GPL.
It may turn out that this is legally unenforcable, but in practical terms an OS that doesn't let you modify the kernel or install unsigned drivers is not open in any real sense. Whether it's a close-source BSD derivitive of an end-run around the GPL the reult is the same.
Konfabulator widgets seem to be WAY less CPU-intensive than Dashboard.
I've noticed that other programs that have gone to webkit from doing straight Aqua/Quartz layout have become slower and more CPU-hungry. It makes sense, Webkit is doing a lot more work, and it's got to re-render a lot more material when a script change changes the content.
Why can't Apple write a driver to format the ipod for HFS+ under Windows?
1. That would require people to install a driver and reboot after installing iTunes, making Apple look bad.
2. It would mean a LOT more chances for people using iPods on slightly shonky Windows machines to have their iPod not work, making Apple look bad.
3. It would mean you coudln't take your iPod and plug it in to someone's computer and use it as a disk drive, if that's what floats your boat, again making Apple look bad.
4. It would cost time and money to implement.
5. The lost sales from 1-3 plus #4 would more than overwhelm any savings on not licensing the patent.
6. As other people noted, they have cross-licensing agreements with Microsoft anyway.
as far as I can tell, they are shipped sans filesystem, and formatted when you install the thing.
They still have to have software in the iPod that knows how to read that format, or the iPod couldn't read its own music. That's what would hypothetically require a license.
As someone else pointed out, Apple has cross-licensing agreements with Microsoft, but other manufacturers don't.
Paying for FAT licenses? BS. They probably exchanged them through cross licensing.
Ah, good point, I should have remembered that detail. Apologies.
Still, Microsoft is soaking people for FAT32, a file system that is such a wretched hive of scum and villainy it makes Mos Eisley look like Blandsville Iowa by comparison, which people only use because it's what they have to do if they're going to support Windows.
Oh, yes, I realise that Microsoft has not been playing a "defensive patent" game. The FAT32 patent as well as their FUD about Linux patents are proof of that. I was mostly noting that this isn't yet "more of the same" so if you haven't burned your XP CDs yet... put down the match.:)
I'm not sure how they got the FAT32 patent, I thought that once you've shipped something you can't turn around and patent it (the idea being that there's no point in encouraging the publication of something that's already been published).
In case you missed it, you don't need to turn in your Windows licenses yet. This is all speculation, and it's not even Microsoft basher speculation for the most part, it just seems to be journos trying to get a scoop by making stuff up.
Though Microsoft has recently created 11th hour license fees on the FAT file system, and I'm sure Apple's paying those on every iPod sold.
neither one offers as much funcitonality as Outlook
Such as being the #1 software for virus and worm distribution in the world?
I banned Outlook and IE (the #2) at work around 1997, based purely on my reaction to the whole design of Active X, Active Desktop, and the rest of the Active culture.
The biggest difference between our email environment and the rest of the company's email environment was that the only way we knew when another email worm was galloping about on the Internet was that we got a bunch of infected junk mail from people at other sites.
We had zero virus infestations. We occasionally had a single infected box, and most of those were people who were running Outlook despite our ban, but I never had to clean more than one box at a time.
I did once have to argue with a contractor over the policy while I was sitting there digging through his files to clean a particularly nasty virus out of his box.
I don't care what the capability of the rest of Outlook is: until they rip the HTML control out of it or implement a version of the HTML control that has no mechanism to run "Active Content", I wouldn't touch it with a 10 foot asbestos pole and I would strongly recommend using it unless you do something horribly desparate like MAC-lock every port and put everyone on a separate VLAN with their own external hardware firewall that QOS-throttles SMTP and scans for surges.
That's what things like Secure Audio Path is all about. Microsoft and Intel and the hardware vendors are working hard to keep our computers from BEING "general purpose computers".
Slashdot Mirror
I put unrated between 2 and 3 stars.
Your wish is my command.
I have set up a set of chained playlists that works roughly like this:
5star - rating is 5 stars, last played > 24 hours ago
4star - rating is 4 stars, last played > 4 days ago
3star - rating is 3 stars, last played > 2 weeks ago
unrated - rating is 0 stars, last played > 6 weeks ago
2star - rating is 2 stars, last played > 12 weeks ago
weighted - playlist is 5star/4star/3star/2star/unrated
Feeding "weighted" into Party Shuffle doesn't seem to produce duplicates and I control what the ratings mean. One star songs don't ever show up.
And in the X Window System you hold down "compose" and enter "|" and "c". Most other extended characters have similarlyl easy-to-remember bindings. And people say UNIX/Linux isn't user-friendly.
But the question was is the availability of ICC the start of cross-platform development?
And the answer is "no".
The start of cross-platform development was back in the '60s with Fortran, a language that - however horrible in general design and in all its ghastly details - was sufficinetly powerful and well standardised to allow people to write working programs that would run on virtually every platform.
It soon became the norm for vendors to make damn sure that developers could easily distinguish portable code from extensions. The current environment where Microsoft's business model depends on people getting enmeshed in a twisty maze of different extensions is a horrible throwback to, well, no... it was never that screwed up even when it was IBM and the seven dwarfs.
Did you try to set your contrast down or switch to green on black? If so, have you any thoughts on why it didn't help?
By the time Apple quits shipping PPC, the XBox360 and PS3 are likely going to be out. How big a deal do people anticipate it to be to run YDL on them?
If the message is unsolicited and trying to sell you something, you are entitled to assume it's spam, and if I get a message like that I'll assume it's spam too.
What I'm getting at, though, is that if it's spam, if that message hadn't been sent in the hundreds of thousands... it wouldn't have been sent at all. There's no profit in sending ONE unsolicited advertisement and stopping. What makes it happen is that it's so cheap to send advertisements in bulk email, and that the spammer doesn't have to wait for people to ask for it.
Without bulk and without unsolicited it's a problem. Without the sales... it's still a problem. There's lots of non-commercial spam -- religious spam, political spam, hoaxes and jokes -- and spam would still be a problem (albeit a much much smaller one) without money involved. Without bulk, though, it wouldn't exist.
That's what I mean when I say "one message isn't spam". One message may be damned good evidence of spam, but the bulkness is an essential part of what makes it happen and what makes it a problem.
if everyone in China sends one message to you [...]
If everyone in China is doing it, then someone is coordinating them. Whever is doing it is the person responsible for sending the messages, the individuals who actually sent them are just proxies. So, that person has sent however many billions of messages that is.
Well, there goes your email account
Nah, I already drop SMTP from China at the IP level.
No, I'm not kidding. If you're in a netblock that's been identified as being in China, Korea, Argentina, any dynamic IP space, or several ISPs like telefonica.es... you can't send me mail.
I literally can't afford the bandwidth to read the messages to spam-filter them any more.
The trick is that the software is it's own validation signature. You don't need to prevent the user from running modified software, what you need to do is prevent the user from being able to read the encrypted DRM files with modified software.
You just restated what I said in different words.
It doesn't use any data to prove it hasn't been tampered with.
Sure it does. It uses itself.
The way it works is that the chip uses the hash of the software to generate a crypto key.
How does it get the hash of the software?
By having the OS or the application read or map the image of the application and calculate the hash from that.
If you modify the OS so that when it's perfoming those specific operations it reads a "shadow" copy of the application that hasn't been modified, it will generate the same key that it would if it hasn't been modified. This is a straight analogy to a standard technique in "rootkits", by the way, and you can download examples from any skr1pt k1ddie website.
So the next step is that the OS itself has to be validated the same way, and then any component that's running in privileged mode, so none of them can modify the checksum calculation.
The only alternatives are obfuscation or simply not allowing the user to modify any component of the system that can get between the application, the disk, and the CPU.
One message isn't spam. Microsoft is welcome to send one message to me. At least if that's all they do... send one message. To me.
If they send one message to 100,000 people, that's not one message any more. That's 100,000 messages.
If 100,000 people send one message back to Microsoft saying "take me off your list" that's still not one message, that's 100,000 messages.
No, one message isn't spam. But I don't think that they really mean "one message". Do you?
The only secret is the key, which is how most security is usually done. ITunes isn't open source, but the DRM is not obfuscated in any way.
If the DRM isn't obfuscated, there's nothing stopping you from reading the key, reading the encrypted file, and writing it out without eny encryption. The only thing keeping you from reading the key is that the software hides it from you... and if the software is open source, it can't hide it.
That's where TPM comes in. TPM establishes an environment that is unmodifiable from the application back to the chip on the motherboard containing the key. And to make that work you need to prevent the user from modifying the OS, the application, or any drivers the application (or the OS on its behalf) might need to use to read data it uses to checksum itself or however else it's supposed to let the chip know it's not been tampered with.
Which is a pretty tough trick, when your software is open-source.
Now, the interesting part comes when you start to move over to a new OS.
That's why you shuld start doing that first.
For your server apps, like Postfix, start setting them up on a server before you quit using Windows as your desktop. Figure out the way the OS works on a system where you don't need much of an environment to get in and use it... not one where you're trying to maintain a whole window system and everything.
Open source can't support strong DRM, because strong DRM depends on you being unable to modify the OS.
You could build a closed source derivitive of BSD and include DRM components in it, but strong DRM in a GPL kernel violates clause 6 of the GPL.
It may turn out that this is legally unenforcable, but in practical terms an OS that doesn't let you modify the kernel or install unsigned drivers is not open in any real sense. Whether it's a close-source BSD derivitive of an end-run around the GPL the reult is the same.
Konfabulator widgets seem to be WAY less CPU-intensive than Dashboard.
I've noticed that other programs that have gone to webkit from doing straight Aqua/Quartz layout have become slower and more CPU-hungry. It makes sense, Webkit is doing a lot more work, and it's got to re-render a lot more material when a script change changes the content.
Why can't Apple write a driver to format the ipod for HFS+ under Windows?
1. That would require people to install a driver and reboot after installing iTunes, making Apple look bad.
2. It would mean a LOT more chances for people using iPods on slightly shonky Windows machines to have their iPod not work, making Apple look bad.
3. It would mean you coudln't take your iPod and plug it in to someone's computer and use it as a disk drive, if that's what floats your boat, again making Apple look bad.
4. It would cost time and money to implement.
5. The lost sales from 1-3 plus #4 would more than overwhelm any savings on not licensing the patent.
6. As other people noted, they have cross-licensing agreements with Microsoft anyway.
It should not be hard to develop another FS.
If you use your iPod on a Mac, you can format it in HFS+. So, Apple doesn't need to develop a new FS, they already have one. No problem, right?
The problem is that you can't access it from Windows if you do that.
Whoops.
as far as I can tell, they are shipped sans filesystem, and formatted when you install the thing.
They still have to have software in the iPod that knows how to read that format, or the iPod couldn't read its own music. That's what would hypothetically require a license.
As someone else pointed out, Apple has cross-licensing agreements with Microsoft, but other manufacturers don't.
Paying for FAT licenses? BS. They probably exchanged them through cross licensing.
Ah, good point, I should have remembered that detail. Apologies.
Still, Microsoft is soaking people for FAT32, a file system that is such a wretched hive of scum and villainy it makes Mos Eisley look like Blandsville Iowa by comparison, which people only use because it's what they have to do if they're going to support Windows.
Unless you're trying to imply that human beings are incapable of acting like the most intelligent creature on the planet as some say we are.
You have it the wrong way around.
Human beings are capable of acting like the most intelligent creature on the planet.
The intelligent thing to do was to stay the hell away from this mess. So who does that leave?
Oh, yes, I realise that Microsoft has not been playing a "defensive patent" game. The FAT32 patent as well as their FUD about Linux patents are proof of that. I was mostly noting that this isn't yet "more of the same" so if you haven't burned your XP CDs yet... put down the match. :)
I'm not sure how they got the FAT32 patent, I thought that once you've shipped something you can't turn around and patent it (the idea being that there's no point in encouraging the publication of something that's already been published).
In case you missed it, you don't need to turn in your Windows licenses yet. This is all speculation, and it's not even Microsoft basher speculation for the most part, it just seems to be journos trying to get a scoop by making stuff up.
Though Microsoft has recently created 11th hour license fees on the FAT file system, and I'm sure Apple's paying those on every iPod sold.
Anyway, this looks like a better story.
neither one offers as much funcitonality as Outlook
Such as being the #1 software for virus and worm distribution in the world?
I banned Outlook and IE (the #2) at work around 1997, based purely on my reaction to the whole design of Active X, Active Desktop, and the rest of the Active culture.
The biggest difference between our email environment and the rest of the company's email environment was that the only way we knew when another email worm was galloping about on the Internet was that we got a bunch of infected junk mail from people at other sites.
We had zero virus infestations. We occasionally had a single infected box, and most of those were people who were running Outlook despite our ban, but I never had to clean more than one box at a time.
I did once have to argue with a contractor over the policy while I was sitting there digging through his files to clean a particularly nasty virus out of his box.
I don't care what the capability of the rest of Outlook is: until they rip the HTML control out of it or implement a version of the HTML control that has no mechanism to run "Active Content", I wouldn't touch it with a 10 foot asbestos pole and I would strongly recommend using it unless you do something horribly desparate like MAC-lock every port and put everyone on a separate VLAN with their own external hardware firewall that QOS-throttles SMTP and scans for surges.
Wow. I can predict the past!
That's what things like Secure Audio Path is all about. Microsoft and Intel and the hardware vendors are working hard to keep our computers from BEING "general purpose computers".