That timeline you link to might be good for an overview, but it isn't very good on details. For example, it says that Europeans easily defeated the Aztec and Incas through superior technology, which is rubbish. It wasn't until disease wiped out huge chunks of the Aztec and Inca populations, including most of their armies, that the Europeans were able to win. Before that, Europeans lost almost every hostile encounter with native-americans.
The native-americans, all throughout both North and South America, were much more susceptable to disease than the Europeans, due to two factors.
First, all the native-americans descend from a relatively small group of people that migrated to the Americas around 30k-40k years ago. As a result, there was much less diversity in their immune systems. By one kind of immune system profiling measure commonly used, for instance, Africa shows 35 different immune profiles, and the Americas only had around 17. Furthermore, they were better distributed in Africa. Any two random Africans would only have something like a 2% chance of having the same profile, whereas in the Americas, the chances were something like 30%.
The consequence of this is that when exposed to a new disease, much less of the native-american population was naturally immune, compared to Asian, African, or European populations, and a given infected person was much more likely to run into people that it would spread to.
Second, native-americans did not domesticate as many kinds of animals. They are lactose intolerant for the most part, so didn't domesticate animals to raise for milk, and for meat, they preferred to hunt. (They did in effect raise meat animals, but they did it through large scale manipulation of the environment to increase the wild populations, rather than raising small domesticated populations, so they were never in close extended contact with the animals they used). In Europe, Asia, and Africa, populations picked up a host of diseases from animals long ago, and adapted.
The result was that the native-americans were doomed from the moment any outsiders contacted them. By the time Eurpean settlers started arriving in serious numbers, native populations had already been devestated by disease. In New England, for example, up to 90-95% of the natives had died by the time the Pilgrims arrived. When European explorers first reached the Pacific coast, they found village after village filled with dead people, victims of smallpox. It had spread from the East coast, via trade between native-american groups.
Besides greatly weakening whole groups of native-americans, the wave of disease changed the relationships between different groups of native-americans. For example, in New England, the group that controlled the area the Pilgrims landed had been hit harder by disease than another nearby group, which they had long been enemies of. This is one of the main reasons the first group decided to help the Pilgrims, instead of driving them away. They feared that their enemies could capitalize on the disease and win, and hoped that they could get the Pilgrims as allies.
"Just because the kernel can load your module doesn't mean your modules is GPLed."
Yes, that's exactly what it means
If your argument were correct, then it would be a violation of GPL for any distribution to make these drivers available. Yes, the major distributions do make them available for download, and no one, not even the FSF, says they are violation the GPL.
The whole discussion of micro-kernel vs monolithic kernel is totally pointless. All popular OS kernels are monolithic. We can get back to the debate when we have a working fast microkernel in the market that is actually competitive
What about QNX? Or pretty much every other real-time system?
Or how about pretty much every system where provable security or reliability is needed?
How about a time frame that is put into place? Lets say you can't patent or trademark an idea/logo if it's been used for over a year. That would solve a lot of problems.
Think about the purpose of trademarks, and you'll see that this is not a good proposal.
You don't need to break RSA - just replace the DLL that handles RSA with one that does nothing. Remember the PC is compromised - so the virus/spyware maker can do that and I think they have done it in the past
Uhm...I think you've gotten a bit out of phase here. He was talking about diddling DNS servers or caches, to get it so that uncompromised computers could be infected via fake MS updates. The signature stops this.
Because using an IP address for the program to access causes problems if your server's IP changes. Simple as that.
If your antivirus vendor is on such a shoestring budget that they can't afford a stable IP address, or are so slow at getting updates out that when the IP address does change, they can't get their software updated in time, then you need a better antivirus vendor!
Best would be for the AV software to use several methods to find its update servers. Start with a built-in IP address. If that doesn't work (either it can't reach the server, or it doesn't find valid updates there--and updates should be digitally signed, of course), then it should try a DNS lookup, and see if that works. If that doesn't give it a valid update server, then it should do a DNS lookup itself, instead of relying on the system resolver, by going to the root servers and working its way down.
Malware does target antivirus software, so AV software should be written to not trust the system it is on.
Let us say that Joe User gets a piece of Malware, so he decides to visit a security company to find a solution to his problem. However, the malware has modified his hosts file to block security company web pages from being accessed, which is extremely typical. Joe User is not experienced enough to even know there is a hosts file that he could change back
This is why antivirus/antispyware software should check for updates by IP address. If it can't find the update servers, only then should it do a DNS lookup, and then it should do it with its own built-in resolver, that starts at the root servers and works its way down.
And it's not much use for network traffic - a 650MB CD of random data lasts only minutes on a 10Mbit link
In practice, it would last much longer, since most computers aren't saturating a 10 Mbit link. Checking the actual data counts on assorted computers I use, I find about a gigabyte a day is typical. Furthermore, a lot of that is traffic from websites or downloads, which is public information that would not need to be encrypted. If we excluded from the encryption traffic to/from port 80 on the remote side, 650 MB of random data would last quite a while.
This sounds like a horrible way to generate a one time pad to me. A quick Googling indicates that there are about 12000 quasars. You need to pick a quasar obersvable at the same time by both parties, so that will cut the number down to around 6000, best case. Given a message, an attacker who merely knows what day the message was sent on only has to consider 86400 seconds/day * 1000000 possible_start_times/second * 6000 possible_keys/possible_start_time, which is less than 2^49 possible_keys/day.
So, with microsecond synchronization, and the attacker only knowing what day the key is from, this scheme is only worth about 49 bits. Cryptographers consider that to be pathetic.
Fscking rich snob. You know, this git travelled around the world, donates money to fight diseases in 3rd world countries, but seems to have this wild belief that these backwaters are going to have telecommunications to each school and house, let alone broadband
Actually, in earlier stories on Gates' view of the $100 laptop, he is clearly aware that they don't have adequate telecommunications, and said that what they need is not laptops, but cell phones and the associated infrastructure. He said what we should be making and giving them cheaply are basically cell phones that you can hook up to a TV and keyboard and use as a computer.
They are using PestPatrol's database, from way before CA bought PestPatrol. It's woefully inaccurate and out of date. SiteAdvisor is an interesting idea, but worthless in its current form.
I'm pretty disappointed with this move by RIM. I was hoping that by going all the way through the courts we might get some serious patent reform out of the whole mess
This might not have been a good case for prompting patent reform. I haven't read the patents myself, but from the discussion on the TWiT podcast, they were saying that the patents actually looked pretty legitimate, and were only likely to be overturned because of the immense pressure the government was putting on to keep their Blackberries going.
In other words, it's not clear that NTP is that bad guy here, and the RIM is the good guy.
Other than the part where they talk about switching to UPS because FedEx increased their rates?
I'm glad they switched to UPS. I live in an apartment. Here is what I've observed FedEx's delivery algorithm to be.
1. If I'm not home, and the package is small, leave a note on the door that they will try tomorrow. Do not leave it with the apartment manager, like every other delivery company would. Sometimes they explain that the item requires a signature.
2. Ignore any note I leave for them saying explicitly to leave it with the manager.
3. If the item is big or obviously expensive, and especially if it has a big note on it that says "SIGNATURE REQUIRED!", leave it at the door. My 17" PowerBook, for example, was simply left at my door. Once, a long time ago, I apparently checked the box on the tag they left that said it was OK to leave an item without getting a signature, and they remember that whenever they have an expensive item or a "SIGNATURE REQUIRED!" required item, and forget it for cheap items.
So, net result with FedEx: if I order something like a USB cable, it means I'll have to drive to the FedEx facility to get it. If I order something expensive, like a computer, it means I'll have to take a day off of work to receive it, or let it sit on my porch unguarded for hours.
A couple mail-order places have lost a couple thousand dollars worth of orders from me because they use FedEx.
eriously "A pallet is a wooden or plastic platform that can be picked up using a forklift; palletized cargo is cargo placed on a pallet, which is how Newegg's inventory is shipped to them." is considered a story?
No, that is considered a "sentence". People called "writers" put together groups of related sentences to form "paragraphs", and groups of related "paragraphs" are what is considered a story.
Considering that Symantec recommended uninstalling Spybot S&D to "protect" Norton Ghost (this after their antivirus gave Spybot a false positive -- you'd think they'd be a bit more careful), I think it's cool they've been hoisted by their own retard...er petard
Spybot also occasionally flags competing products incorrectly as spyware. The difference is that when Microsoft does it, the vendors of those products can follow a simple procedure: fill out a form on Microsoft's site, and Microsoft will evaluate, and fix the problem. Good luck getting anyone connected to Spybot to talk to you. Spybot rarely corrects their mistakes.
I know of cases where Microsoft has fixed such a mistake in the next update, whereas Spybot has taken over a year.
There is very little legacy.NET code out there, and if you're writing new code, why lock your client into a platform?[...]I don't intend to waste my time learning it, because it's dead-end technology for a dying platform
Well,.NET runs on Windows using Microsoft's.NET implementation, and on Unix and OS X using Mono's implementation. That's an acceptable level of being locked to a platform for most people.
I've got one of these new iMacs, and have been able to check out the speed in person. What people are overlooking is that the speed tests in the reviews are focusing on a few apps that are particularly bad for emulation, such as Photoshop, and even in that case, Photoshop comes out acceptable for quite a lot of tasks. It's not anything a pro would want to use--but pros are not the targets for the iMac. It looks like CPU-bound tasks are roughly about half as fast as they would be on a G5 at about the same clock speed. Most things aren't CPU-bound, and so the hit is smaller.
For things that the intended users of iMacs will use, the performance is fine under emulation. Here's what I've observed, in comparision to my 17" G4 PowerBook, and my 1.8 GHz G5 PowerMac. I've got a Radeon 9800 Pro in the G5, and previously had a GeForce FX 5200 in it.
Word on the iMac feels faster than on the PowerBook, and comparable to the G5. (And Word on the iMac totally kicks the ass of OpenOffice 2 on my Athlon 64 Linux box...).
World of Warcraft on the iMac is faster than on the PowerBook, and faster than on the G5 with the FX 5200, and slower than on the G5 with the Radeon 9800 Pro. It is the video card that is the main factor here, not CPU performance.
As for native apps, such as Safari, Mail, iLife, they are much much much faster than on my PowerMac. X launches in about 1/4 of the time, for example.
Summary: for most non-pro users, the new iMac will be the fastest Mac they've ever seen.
Apple management will fail in its attempts to thwart the hackers
All they have to do is make it so that the time it takes to crack is longer than the time between important updates. They seem to have managed this with iTunes...hymn/jhymn has been broken with iTunes 6 for quite a while now. With iTunes, if you are using jhymn, and an Apple updates breaks it, all you lose is the ability to remove the DRM from newly purchased songs. All your old stuff keeps working. Minor hassle, at most.
With the OS, though, updates will probably make cracked copies stop working. So, if you are running a cracked copy, it will go like this: install update, OS breaks, wait days, weeks, or months for new crack. That won't be acceptable to many people. (Of course, you can avoid installing updates until you know there is a crack for them).
Slashdot Mirror
The native-americans, all throughout both North and South America, were much more susceptable to disease than the Europeans, due to two factors.
First, all the native-americans descend from a relatively small group of people that migrated to the Americas around 30k-40k years ago. As a result, there was much less diversity in their immune systems. By one kind of immune system profiling measure commonly used, for instance, Africa shows 35 different immune profiles, and the Americas only had around 17. Furthermore, they were better distributed in Africa. Any two random Africans would only have something like a 2% chance of having the same profile, whereas in the Americas, the chances were something like 30%.
The consequence of this is that when exposed to a new disease, much less of the native-american population was naturally immune, compared to Asian, African, or European populations, and a given infected person was much more likely to run into people that it would spread to.
Second, native-americans did not domesticate as many kinds of animals. They are lactose intolerant for the most part, so didn't domesticate animals to raise for milk, and for meat, they preferred to hunt. (They did in effect raise meat animals, but they did it through large scale manipulation of the environment to increase the wild populations, rather than raising small domesticated populations, so they were never in close extended contact with the animals they used). In Europe, Asia, and Africa, populations picked up a host of diseases from animals long ago, and adapted.
The result was that the native-americans were doomed from the moment any outsiders contacted them. By the time Eurpean settlers started arriving in serious numbers, native populations had already been devestated by disease. In New England, for example, up to 90-95% of the natives had died by the time the Pilgrims arrived. When European explorers first reached the Pacific coast, they found village after village filled with dead people, victims of smallpox. It had spread from the East coast, via trade between native-american groups.
Besides greatly weakening whole groups of native-americans, the wave of disease changed the relationships between different groups of native-americans. For example, in New England, the group that controlled the area the Pilgrims landed had been hit harder by disease than another nearby group, which they had long been enemies of. This is one of the main reasons the first group decided to help the Pilgrims, instead of driving them away. They feared that their enemies could capitalize on the disease and win, and hoped that they could get the Pilgrims as allies.
I believe Archos invented the HDD-based MP3 player.
Yes, that's exactly what it means
If your argument were correct, then it would be a violation of GPL for any distribution to make these drivers available. Yes, the major distributions do make them available for download, and no one, not even the FSF, says they are violation the GPL.
What about QNX? Or pretty much every other real-time system?
Or how about pretty much every system where provable security or reliability is needed?
Think about the purpose of trademarks, and you'll see that this is not a good proposal.
Uhm...I think you've gotten a bit out of phase here. He was talking about diddling DNS servers or caches, to get it so that uncompromised computers could be infected via fake MS updates. The signature stops this.
If your antivirus vendor is on such a shoestring budget that they can't afford a stable IP address, or are so slow at getting updates out that when the IP address does change, they can't get their software updated in time, then you need a better antivirus vendor!
Best would be for the AV software to use several methods to find its update servers. Start with a built-in IP address. If that doesn't work (either it can't reach the server, or it doesn't find valid updates there--and updates should be digitally signed, of course), then it should try a DNS lookup, and see if that works. If that doesn't give it a valid update server, then it should do a DNS lookup itself, instead of relying on the system resolver, by going to the root servers and working its way down.
Malware does target antivirus software, so AV software should be written to not trust the system it is on.
This is why antivirus/antispyware software should check for updates by IP address. If it can't find the update servers, only then should it do a DNS lookup, and then it should do it with its own built-in resolver, that starts at the root servers and works its way down.
Which still leaves it one of the best shows on television.
In practice, it would last much longer, since most computers aren't saturating a 10 Mbit link. Checking the actual data counts on assorted computers I use, I find about a gigabyte a day is typical. Furthermore, a lot of that is traffic from websites or downloads, which is public information that would not need to be encrypted. If we excluded from the encryption traffic to/from port 80 on the remote side, 650 MB of random data would last quite a while.
This sounds like a horrible way to generate a one time pad to me. A quick Googling indicates that there are about 12000 quasars. You need to pick a quasar obersvable at the same time by both parties, so that will cut the number down to around 6000, best case. Given a message, an attacker who merely knows what day the message was sent on only has to consider 86400 seconds/day * 1000000 possible_start_times/second * 6000 possible_keys/possible_start_time, which is less than 2^49 possible_keys/day. So, with microsecond synchronization, and the attacker only knowing what day the key is from, this scheme is only worth about 49 bits. Cryptographers consider that to be pathetic.
This is the computer equivalent of a cargo cult.
Actually, in earlier stories on Gates' view of the $100 laptop, he is clearly aware that they don't have adequate telecommunications, and said that what they need is not laptops, but cell phones and the associated infrastructure. He said what we should be making and giving them cheaply are basically cell phones that you can hook up to a TV and keyboard and use as a computer.
They are using PestPatrol's database, from way before CA bought PestPatrol. It's woefully inaccurate and out of date. SiteAdvisor is an interesting idea, but worthless in its current form.
This might not have been a good case for prompting patent reform. I haven't read the patents myself, but from the discussion on the TWiT podcast, they were saying that the patents actually looked pretty legitimate, and were only likely to be overturned because of the immense pressure the government was putting on to keep their Blackberries going.
In other words, it's not clear that NTP is that bad guy here, and the RIM is the good guy.
I'm glad they switched to UPS. I live in an apartment. Here is what I've observed FedEx's delivery algorithm to be.
1. If I'm not home, and the package is small, leave a note on the door that they will try tomorrow. Do not leave it with the apartment manager, like every other delivery company would. Sometimes they explain that the item requires a signature.
2. Ignore any note I leave for them saying explicitly to leave it with the manager.
3. If the item is big or obviously expensive, and especially if it has a big note on it that says "SIGNATURE REQUIRED!", leave it at the door. My 17" PowerBook, for example, was simply left at my door. Once, a long time ago, I apparently checked the box on the tag they left that said it was OK to leave an item without getting a signature, and they remember that whenever they have an expensive item or a "SIGNATURE REQUIRED!" required item, and forget it for cheap items.
So, net result with FedEx: if I order something like a USB cable, it means I'll have to drive to the FedEx facility to get it. If I order something expensive, like a computer, it means I'll have to take a day off of work to receive it, or let it sit on my porch unguarded for hours.
A couple mail-order places have lost a couple thousand dollars worth of orders from me because they use FedEx.
No, that is considered a "sentence". People called "writers" put together groups of related sentences to form "paragraphs", and groups of related "paragraphs" are what is considered a story.
Spybot also occasionally flags competing products incorrectly as spyware. The difference is that when Microsoft does it, the vendors of those products can follow a simple procedure: fill out a form on Microsoft's site, and Microsoft will evaluate, and fix the problem. Good luck getting anyone connected to Spybot to talk to you. Spybot rarely corrects their mistakes.
I know of cases where Microsoft has fixed such a mistake in the next update, whereas Spybot has taken over a year.
Going beyond search engines, there are a couple funny things going on with tinyurl. Try tinyurl.com/dick. That's pretty funny.
Then try tinyurl.com/cunt. That one is just mean.
Well, .NET runs on Windows using Microsoft's .NET implementation, and on Unix and OS X using Mono's implementation. That's an acceptable level of being locked to a platform for most people.
Yes, it is. Plurality doesn't mean what you think it means, apparently.
Even more surprising, 100% of Elves in World of Warcraft are played by Humans!
And after that, will it be getting someone lost in a TomTom commercial?
For things that the intended users of iMacs will use, the performance is fine under emulation. Here's what I've observed, in comparision to my 17" G4 PowerBook, and my 1.8 GHz G5 PowerMac. I've got a Radeon 9800 Pro in the G5, and previously had a GeForce FX 5200 in it.
Word on the iMac feels faster than on the PowerBook, and comparable to the G5. (And Word on the iMac totally kicks the ass of OpenOffice 2 on my Athlon 64 Linux box...).
World of Warcraft on the iMac is faster than on the PowerBook, and faster than on the G5 with the FX 5200, and slower than on the G5 with the Radeon 9800 Pro. It is the video card that is the main factor here, not CPU performance.
As for native apps, such as Safari, Mail, iLife, they are much much much faster than on my PowerMac. X launches in about 1/4 of the time, for example.
Summary: for most non-pro users, the new iMac will be the fastest Mac they've ever seen.
All they have to do is make it so that the time it takes to crack is longer than the time between important updates. They seem to have managed this with iTunes...hymn/jhymn has been broken with iTunes 6 for quite a while now. With iTunes, if you are using jhymn, and an Apple updates breaks it, all you lose is the ability to remove the DRM from newly purchased songs. All your old stuff keeps working. Minor hassle, at most.
With the OS, though, updates will probably make cracked copies stop working. So, if you are running a cracked copy, it will go like this: install update, OS breaks, wait days, weeks, or months for new crack. That won't be acceptable to many people. (Of course, you can avoid installing updates until you know there is a crack for them).