Unfortunately, not becoming law is a new trick that the last government learned can be very effective. See the Internet Watch Foundation for an example. There is no law requiring ISPs to use their blacklists (which, in the past, has included things like random Wikipedia pages), but the major ISPs were told that there would be such a law with some nasty penalties for them if they did not voluntarily start using it. If you complain about it to your MP, you get a form letter back telling you that the IWF is a voluntary body not controlled by the government, so it's not their responsibility.
It's also worth remembering that the margin on luxury products tends to be higher than on essentials. If someone with kids is spending $100 on food and other essentials, and someone else is spending $50 on toys, then the retailer may well be making more profit from the second person.
You don't even have to buy anything. One of my friends asked for my opinion on some shoes she was considering buying and sent me the Amazon link to two options. For the next year, Amazon was convinced that I wanted to buy ladies shoes and dresses.
Not necessarily. Gender identity and sexual preference are not always related. There are some transgender people who are attracted to members of the gender that they identify with, rather than the one that they were born with, which makes life very difficult for them.
So then, you get proper circuit design/sim software, something which Minecraft is not.
I disagree. Building something with a tool that is not designed for it is often a much better learning experience. Building an ALU in a circuit design package is relatively easy - most of the tedious stuff is done for you and there are prepackaged components for a lot of what you want. Building it one switch at a time is a far less productive use of your time, but gives you a much better understanding of how everything really works (although the fact that time and values are discrete in minecraft hides the messy analogue stuff from you).
The goal here is not to build a graphing calculator. That would be pointless - you can buy a much better one very cheaply. The point is to really understand how such a device works so that he could build one if he needed to. That's not to say I wouldn't suggest that his next project be done in Verilog, but there's something to be said from starting with the wrong tool for the job. If nothing else, it teaches you to appreciate the good tools a lot more...
I think you've misunderstood how it works for Big Pharma; curing something does not generate them bucketloads of money
Not immediately, no. The problem with cancer is that it often kills people at a relatively young age. Take the person from just up the thread who had a friend die of leukaemia aged 22. Think how much the industry would have made from him if he'd lived even an averagely healthy life - mostly only needing treatment for minor illnesses - until the age of 70. It will dwarf what they could make from him in a few months of chemotherapy. Even people dying of cancer in their 40s and 50s are a problem for pharma companies: they're dying before they hit the most profitable time of their lives!
Curing one disease doesn't mean that people will stop needing the services of the pharmaceutical industry. Quite the reverse: the longer they live, the more things you can sell them to ensure that they keep on living.
22 is fine as a lifespan for a 'natural' human. You're able to reproduce by about the age of 12, so that gives 10 years to raise your children to the point where they can survive on their own. With a tribe of overlapping ages, they'll then have the support of some other adults for the next few years until they are fully (physically) mature. The short generation, it doesn't take as long for beneficial mutations to spread to the entire tribe, so you have an advantage in evolutionary speed.
For some reason, people often conflate 'natural' with 'good'.
Sell an expensive treatment, patient dies a few months later.
Sell a cure, patient lives for 20-30 more years, suffers from a large number of minor ailments and a few major ones, keeps earning money for 20-30 years and spending some proportion of it on drugs, becomes a long-term revenue stream.
Which one will the evil profit-driven capitalists pick? In fact, there's a third option:
Competitor sells a cure, we don't make any money from selling the treatment.
Minecraft is basically a 3D cellular automata simulator. You set cells to different values ('materials') and then in the next update cycle their properties will change depending on their value and the values of their neighbours. Because cellular automata of this nature are Turing complete it is possible to implement any kind of computer in Minecraft. Because of the visual representation of the blocks, it is possible to create visual representations.
This is no more of an achievement than creating a graphical calculator from LEDs and transistors. Which is to say, it's very impressive.
I suppose that if this treatment is as effective as advertised, we will soon see just how powerful Big Pharma's influence is.
You're assuming that a cure for cancer would not be in the interests of Big Pharma. I don't think that is necessarily true. Most current cancer treatments are expensive (also often expensive to produce, so not a huge profit margin) and are only sold for shortish treatment periods. Either the patient recovers or dies. A proper cure could easily see people living 20+ years longer. Old people tend to be a better market for pharmaceuticals (of the non-recreational kind, anyway) and so the profit from having people not die from cancer is likely to be greater than the loss from not selling them cancer treatments.
Blood transfusions are pretty routine these days. If you need to be in hospital receiving regular transfusions for a little while until the tumours are gone then that would suck, but it would still be vastly better than any of the alternatives...
you cant just take text-based algorithms and "apply them" to voice and video;) the domains are completely different.
I believe the grandparent was talking about Skype's IM component, not the voice / video conferencing capabilities. I was quite surprised recently to discover that a lot of my non-geeky friends were completely unaware of instant messaging until they encountered it via Skype and Facebook - the IM was the main feature that they said they liked about Facebook, and when I asked what made it better than other IM systems it became clear that they thought that Facebook was the first thing to provide this kind of functionality.
That's where feature and bug bounties come in. A new feature may only be worth $20 to one person, and then it's probably not worth implementing. It may be worth $20 to 100 people though, and then it becomes a lot more interesting. If it's worth $20 to 1,000 people, then that's even better.
Some projects have a mechanism for doing this already. New features are listed along with their costs, and when the total pledge value reaches that amount a developer implements them and gets paid.
You can not, however, send messages to a specific person, nor can you restrict access to anything you post in your journal. I'd say this means that Slashdot is more of a public forum than a social network.
My cheapish all-in-one printer can scan things and produce PDFs, so a PDF might be hand written and then scanned - it doesn't necessarily have to have involved a computer at all. That said, the reason that a lot of places require CVs in Word format is that they are going to run them through context-insensitive keyword matching to pair you with potential positions. These places are generally worth avoiding.
If you really want to mess with them, recall that Word documents support OLE. You can therefore embed a PDF in a Word document. It will then be stored as a PDF bytestream and as an image (in case they don't have a PDF application). It will look correct when they open it in Word, but none of their other tools will be able to extract text from it.
Not really. Skimming the patent, it seems that it is describing a preprocessing stage where you try running all static constructors, record their results, and generate a simplified version. This would be equally applicable to C++ - if all that the constructor does in a global is touch memory then you can just record the memory layout and stick that in the executable. It's quite neat, but I'm not entirely sure it counts as non-obvious, because it's really just an application of constant propagation, which all compilers have done for decades.
640KB is not enough for everyone, but it is enough for some people. It's just about enough to edit long text documents and simple images. 6.4MB is enough for more, it's enough for long rich text documents and occasional more complex images. 64MB is enough for more, it's enough for editing large photos. 640MB is enough for nondestructively editing complex photos and enough for simple video editing. 6.4GB is enough for realtime nondestructive video editing. 64GB is enough for... pretty much anything we use computers for today.
There will always be new applications that require more RAM, but often they are fairly niche products. The same is true of bandwidth. Waiting to load web pages irritates everyone. Waiting to load images irritates most people. Not being able to stream video irritates fewer people. Not being able to stream HD video irritates even fewer. Not being able to interactively stream 3D volume datasets probably only irritates a few dozen...
Almost there, but not quite. iPlayer HD is 3.6Mb/s. That's possible with HSPA, but only if you're relatively close to a fairly uncontended tower. BluRay quality video is around 30-50Mb/s. That isn't possible with HSPA.
Phones don't have that resolution yet, but they're getting there, and I wouldn't be surprised if picoprojectors became a lot more common...
Also remember that this is likely to be 30Gb/s of total throughput between the cell and all clients (or maybe all in one direction, depending on the antenna design), it's not 30Gb/s to every person (well, it might be, I didn't RTFA). 30Gb/s is way more than one person can use at the moment - during my PhD a few years ago I had a 1Gb/s connection to the Internet and had to be careful if I clicked on large download links because my hard disk couldn't keep up with the network speed and the machine would fill up RAM trying to cache everything and then become unresponsive until the download finished. 100Mb/s, however, is quite easy to saturate without placing any strain on the local machine and 10Mb/s is very easy. Ignoring other overheads, this means that one cell can give 300 people a 100Mb/s connection or 3,000 people a 10Mb/s connection. That means that 3,000 people in the same cell can be streaming HD video without any problems.
You seem to be missing the reason why most software is written in the first place. It's not done because someone wants to make money, it's done because someone wants to use the resulting code. This accounts for 100% of the open source code I've worked on. Some projects I've started because no one else provided the tools I wanted. Some projects other people started and I've contributed to because it's easier than rewriting everything.
Clang is a good example: it was started because Apple wanted a modular [Objective-]C[++] front end that could be used in a compiler, in refactoring tools, for syntax highlighting, and so on. My first contribution was to support the GNU Objective-C ABI - something of no use to Apple, but a great deal of use to me. I didn't get paid for this directly, but having a decent compiler for Objective-C stuff on Linux/*BSD has helped me get paid for other stuff. I've since done paid work on clang for other companies that needed other features implemented, or other systems supported.
Clang wasn't started in the hope that its authors would be able to get people to pay, it was started because its authors' employer needed it. It was open sourced because that helped reduce the cost of development for everyone involved. Apple probably could have kept it proprietary and developed in house, but then companies like Google and ARM, and individuals like myself, wouldn't have contributed anything, so their costs would have been higher.
If you're using ZFS, then the best solution is to use RAID-Z for online storage and then have two external disks which you use zfs send / zfs receive to update. This means that catastrophic failure (e.g. a power supply problem blowing all of the drives in the machine) will still leave you able to recover. Ideally, you should store one drive at home and one elsewhere, so that if someone steals your computer then they don't get the data.
Slashdot Mirror
Unfortunately, not becoming law is a new trick that the last government learned can be very effective. See the Internet Watch Foundation for an example. There is no law requiring ISPs to use their blacklists (which, in the past, has included things like random Wikipedia pages), but the major ISPs were told that there would be such a law with some nasty penalties for them if they did not voluntarily start using it. If you complain about it to your MP, you get a form letter back telling you that the IWF is a voluntary body not controlled by the government, so it's not their responsibility.
It's also worth remembering that the margin on luxury products tends to be higher than on essentials. If someone with kids is spending $100 on food and other essentials, and someone else is spending $50 on toys, then the retailer may well be making more profit from the second person.
That's true, but I don't think anyone has shown a strong correlation correlation between sociopathic tendencies and sexual orientation...
You don't even have to buy anything. One of my friends asked for my opinion on some shoes she was considering buying and sent me the Amazon link to two options. For the next year, Amazon was convinced that I wanted to buy ladies shoes and dresses.
Not necessarily. Gender identity and sexual preference are not always related. There are some transgender people who are attracted to members of the gender that they identify with, rather than the one that they were born with, which makes life very difficult for them.
Everybody looks down on someone. I try to be egalitarian and look down on all of you equally.
So then, you get proper circuit design/sim software, something which Minecraft is not.
I disagree. Building something with a tool that is not designed for it is often a much better learning experience. Building an ALU in a circuit design package is relatively easy - most of the tedious stuff is done for you and there are prepackaged components for a lot of what you want. Building it one switch at a time is a far less productive use of your time, but gives you a much better understanding of how everything really works (although the fact that time and values are discrete in minecraft hides the messy analogue stuff from you).
The goal here is not to build a graphing calculator. That would be pointless - you can buy a much better one very cheaply. The point is to really understand how such a device works so that he could build one if he needed to. That's not to say I wouldn't suggest that his next project be done in Verilog, but there's something to be said from starting with the wrong tool for the job. If nothing else, it teaches you to appreciate the good tools a lot more...
I think you've misunderstood how it works for Big Pharma; curing something does not generate them bucketloads of money
Not immediately, no. The problem with cancer is that it often kills people at a relatively young age. Take the person from just up the thread who had a friend die of leukaemia aged 22. Think how much the industry would have made from him if he'd lived even an averagely healthy life - mostly only needing treatment for minor illnesses - until the age of 70. It will dwarf what they could make from him in a few months of chemotherapy. Even people dying of cancer in their 40s and 50s are a problem for pharma companies: they're dying before they hit the most profitable time of their lives!
Curing one disease doesn't mean that people will stop needing the services of the pharmaceutical industry. Quite the reverse: the longer they live, the more things you can sell them to ensure that they keep on living.
22 is fine as a lifespan for a 'natural' human. You're able to reproduce by about the age of 12, so that gives 10 years to raise your children to the point where they can survive on their own. With a tribe of overlapping ages, they'll then have the support of some other adults for the next few years until they are fully (physically) mature. The short generation, it doesn't take as long for beneficial mutations to spread to the entire tribe, so you have an advantage in evolutionary speed.
For some reason, people often conflate 'natural' with 'good'.
Let's see, we have two choices:
Which one will the evil profit-driven capitalists pick? In fact, there's a third option:
Minecraft is basically a 3D cellular automata simulator. You set cells to different values ('materials') and then in the next update cycle their properties will change depending on their value and the values of their neighbours. Because cellular automata of this nature are Turing complete it is possible to implement any kind of computer in Minecraft. Because of the visual representation of the blocks, it is possible to create visual representations.
This is no more of an achievement than creating a graphical calculator from LEDs and transistors. Which is to say, it's very impressive.
I suppose that if this treatment is as effective as advertised, we will soon see just how powerful Big Pharma's influence is.
You're assuming that a cure for cancer would not be in the interests of Big Pharma. I don't think that is necessarily true. Most current cancer treatments are expensive (also often expensive to produce, so not a huge profit margin) and are only sold for shortish treatment periods. Either the patient recovers or dies. A proper cure could easily see people living 20+ years longer. Old people tend to be a better market for pharmaceuticals (of the non-recreational kind, anyway) and so the profit from having people not die from cancer is likely to be greater than the loss from not selling them cancer treatments.
Blood transfusions are pretty routine these days. If you need to be in hospital receiving regular transfusions for a little while until the tumours are gone then that would suck, but it would still be vastly better than any of the alternatives...
you cant just take text-based algorithms and "apply them" to voice and video ;) the domains are completely different.
I believe the grandparent was talking about Skype's IM component, not the voice / video conferencing capabilities. I was quite surprised recently to discover that a lot of my non-geeky friends were completely unaware of instant messaging until they encountered it via Skype and Facebook - the IM was the main feature that they said they liked about Facebook, and when I asked what made it better than other IM systems it became clear that they thought that Facebook was the first thing to provide this kind of functionality.
That's where feature and bug bounties come in. A new feature may only be worth $20 to one person, and then it's probably not worth implementing. It may be worth $20 to 100 people though, and then it becomes a lot more interesting. If it's worth $20 to 1,000 people, then that's even better.
Some projects have a mechanism for doing this already. New features are listed along with their costs, and when the total pledge value reaches that amount a developer implements them and gets paid.
Have you ever tried flying first class? You get a large seat, lots of space, a glass of champagne brought to you before takeoff, much better food.
Of course, you're paying over $100/hour for the privilege...
You can not, however, send messages to a specific person, nor can you restrict access to anything you post in your journal. I'd say this means that Slashdot is more of a public forum than a social network.
My cheapish all-in-one printer can scan things and produce PDFs, so a PDF might be hand written and then scanned - it doesn't necessarily have to have involved a computer at all. That said, the reason that a lot of places require CVs in Word format is that they are going to run them through context-insensitive keyword matching to pair you with potential positions. These places are generally worth avoiding.
If you really want to mess with them, recall that Word documents support OLE. You can therefore embed a PDF in a Word document. It will then be stored as a PDF bytestream and as an image (in case they don't have a PDF application). It will look correct when they open it in Word, but none of their other tools will be able to extract text from it.
What happened?
Flights that normal people could afford.
Not really. Skimming the patent, it seems that it is describing a preprocessing stage where you try running all static constructors, record their results, and generate a simplified version. This would be equally applicable to C++ - if all that the constructor does in a global is touch memory then you can just record the memory layout and stick that in the executable. It's quite neat, but I'm not entirely sure it counts as non-obvious, because it's really just an application of constant propagation, which all compilers have done for decades.
640KB is not enough for everyone, but it is enough for some people. It's just about enough to edit long text documents and simple images. 6.4MB is enough for more, it's enough for long rich text documents and occasional more complex images. 64MB is enough for more, it's enough for editing large photos. 640MB is enough for nondestructively editing complex photos and enough for simple video editing. 6.4GB is enough for realtime nondestructive video editing. 64GB is enough for... pretty much anything we use computers for today.
There will always be new applications that require more RAM, but often they are fairly niche products. The same is true of bandwidth. Waiting to load web pages irritates everyone. Waiting to load images irritates most people. Not being able to stream video irritates fewer people. Not being able to stream HD video irritates even fewer. Not being able to interactively stream 3D volume datasets probably only irritates a few dozen...
Almost there, but not quite. iPlayer HD is 3.6Mb/s. That's possible with HSPA, but only if you're relatively close to a fairly uncontended tower. BluRay quality video is around 30-50Mb/s. That isn't possible with HSPA.
Phones don't have that resolution yet, but they're getting there, and I wouldn't be surprised if picoprojectors became a lot more common...
Also remember that this is likely to be 30Gb/s of total throughput between the cell and all clients (or maybe all in one direction, depending on the antenna design), it's not 30Gb/s to every person (well, it might be, I didn't RTFA). 30Gb/s is way more than one person can use at the moment - during my PhD a few years ago I had a 1Gb/s connection to the Internet and had to be careful if I clicked on large download links because my hard disk couldn't keep up with the network speed and the machine would fill up RAM trying to cache everything and then become unresponsive until the download finished. 100Mb/s, however, is quite easy to saturate without placing any strain on the local machine and 10Mb/s is very easy. Ignoring other overheads, this means that one cell can give 300 people a 100Mb/s connection or 3,000 people a 10Mb/s connection. That means that 3,000 people in the same cell can be streaming HD video without any problems.
You seem to be missing the reason why most software is written in the first place. It's not done because someone wants to make money, it's done because someone wants to use the resulting code. This accounts for 100% of the open source code I've worked on. Some projects I've started because no one else provided the tools I wanted. Some projects other people started and I've contributed to because it's easier than rewriting everything.
Clang is a good example: it was started because Apple wanted a modular [Objective-]C[++] front end that could be used in a compiler, in refactoring tools, for syntax highlighting, and so on. My first contribution was to support the GNU Objective-C ABI - something of no use to Apple, but a great deal of use to me. I didn't get paid for this directly, but having a decent compiler for Objective-C stuff on Linux/*BSD has helped me get paid for other stuff. I've since done paid work on clang for other companies that needed other features implemented, or other systems supported.
Clang wasn't started in the hope that its authors would be able to get people to pay, it was started because its authors' employer needed it. It was open sourced because that helped reduce the cost of development for everyone involved. Apple probably could have kept it proprietary and developed in house, but then companies like Google and ARM, and individuals like myself, wouldn't have contributed anything, so their costs would have been higher.
If you're using ZFS, then the best solution is to use RAID-Z for online storage and then have two external disks which you use zfs send / zfs receive to update. This means that catastrophic failure (e.g. a power supply problem blowing all of the drives in the machine) will still leave you able to recover. Ideally, you should store one drive at home and one elsewhere, so that if someone steals your computer then they don't get the data.