No, Khamenei is part of the extreme faction that I mentioned in my post. The current posturing by the USA is driving support to him and others like him.
Finland has Nokia and just across the border there is Ericsson. This means that the local telephone companies get to test a lot of new networking hardware before it goes into mass production, which lowers their costs. The problem in much of the rest of the world is that phone companies sold data plans for early smartphones that had tiny screens and could just about do web browsing if you had a lot of patience and just about managed email. An unlimited plan for one of these phones was well under 1GB - irrespective of how fast the network was, you just couldn't consume much data on them. Then people took the same plans and started using modern smartphones with them. A single YouTube video on a modern phone will use more data than a month with an early Blackberry, for example. The networks were very slow at adapting, and are now trying to readjust their prices to levels that actually make sense.
One of the important prerequisites for a free market is informed customers. I have no problem at all with ISPs providing caps. If everyone saturated their connections 24/7 then they would not be able to provide the service, and the cost of actually providing that amount of bandwidth to everyone would be far greater than most customers are able to afford. The problem is not the capping, it is misleading advertising. If you are going to offer 5GB per month, advertise 5GB per month. If you are going to offer 50GB per month, advertise 50GB per month. If you are going to deploy a transparent proxy that resamples images, specify that. If you are going to block access to certain sites, or only permit HTTP traffic, don't say that you provide Internet access. Tell people exactly what service you will provide and allow potential customers to decide whether they think it is worth what you are charging.
Of course that sort of glosses over the assistance the US provided the UK prior to entering the war as well.
You mean assistance like loaning money, loans that were only paid off a few years ago? Or perhaps assistance like allowing British nuclear physicists to work on the Manhattan Project, and then not allowing them to share any of the results with the UK after the war?
Except it's just a simple translation which you are welcome to verify
The literal translation was to wipe the nation of Israel off the map. You can interpret that in several ways. One is to believe that it means killing everyone there. This is the spin that you are reporting. The other is to realise that Iran has never accepted the legitimacy of Israel as a state and wishes to see the state of Israel eliminated as a legal entity. This is the interpretation promoted by Iran.
It's also important to consider the context of the speech. Iran is a Persian country surrounded by Arabic countries. There is a couple of thousand years of history of conflict between the two groups (and not just distant history - remember Iraq invading Iran?). The only way Iran survives in the region is by reminding everyone else that Israel is a more important enemy than them, and that they're on everyone else's side. Picture the scrawny kid that hangs around with the bullies and shouts insults at other people to show how much they are a part of the group. That's Iran's international political status, but what about internally? They have a complex political structure, where the fundamentalist muslims are not a majority, but are influential. The leadership has to, at least publicly, appear to be acceptable to this group or lose power, in much the same way that the US President has to pander to the Christian right in America.
There are other similarities. Both countries have a moderately sane leadership that needs to pander to extreme fundamentalists to remain in power. The odd thing is that the current US strategy could have been designed to destabilise Iran and push the moderate elements out of power.
It doesn't just need to be a rocket, it needs to be a really big rocket. To deflect something potentially civilisation-destroying, you want to do it as early as possible. Inside lunar orbit is much too late. That means that you need to get a rocket out to it, and then match velocities (just ramming it won't be enough) and deliver the explosive to a sensible location. Ideally, you want to attach to the side and push it, rather than try to blow it up. We're talking something that would make the Apollo rockets look small...
libgnustep-base-dev clearly includes files like NSObject.h.
So? NSObject.h is part of the Foundation framework. It isn't part of the Objective-C runtime. The relevant headers are things like objc.h and runtime.h. The relevant library is libobjc.so.
Unless Debian went out of their way to write their own shims to reproduce the functionality of the GNUstep runtime based on the GNU objc runtime (which uses Object instead of NSObject), it seems far more likely to me that the GNUstep runtime is included in the libgnustep-base package
Uh, what? That doesn't even make sense.
Both runtimes can co-exist just fine, to the best of my knowledge.
Not linked to the same binary. They implement a largely overlapping set of symbols. Linking both causes massive breakage.
But first you should make sure you're correct, because, as I said, it seems extremely implausible.
Look at the source for the GNUstep runtime. Now look in the Debian package. Note that it is not there. Perhaps you should actually check that you know what you are talking about before being patronising.
I should possibly mention that I am the maintainer of the GNUstep Objective-C runtime, so there's a pretty good chance that I know what I'm talking about...
A grand a year is assuming a class of at least about 250 students and all of them buying the book (instead of borrowing it from the library or sharing). And that's for at least a month's worth of work - at the speed most lecturers write it's 2-3 months. Hardly a good return on investment.
I absolutely agree that one can focus too much on basic arithmetic facts. On the other hand, if you can't do them at all, it's really hard to get anything out of higher math.
I think there are two problems. One is teaching too much arithmetic early, and the other is calling it maths. This means that students think that maths is boring. My mother used to teach primary school children. Her terms and mine didn't quite line up, so I'd often go in at the end of her term and help out, and I kept doing this when I was at university (and had even shorter terms). One term, I taught them some game theory. I will always remember the comment from one of them: 'I thought you were meant to teach us maths. This isn't maths, this is fun.' Game theory is important to understanding a whole raft of things in the real world, but I've never come across a school that teaches it as part of the normal syllabus.
Things like game theory, graph theory, and formal logic are all useful and don't require very much arithmetic. You could easily start teaching them to primary-school children alongside arithmetic, and get them interested in maths early on.
The other answer is that a lot of aircraft simply lose altitude quite quickly when flying upside down. It's pretty common in aerobatics to expect to lose 1000 feet over a short period of inverted flight.
Why would they bother? My first book is used by two university courses that I know of, and the income from that would not even be close to making it worth writing the book. Even if you make everyone in a class of 100 buy the book every year, the author's royalty is going to be pretty small. You'd be lucky to make $1,000/year as a result. Given the time required to write the book, this seems like a pretty bad idea.
Search for Dave Cutler, DEC, Microsoft and Lawsuit. Cutler left DEC and went to work on Windows NT (the WNT acronym being V+1, M+1, S+1 is a coincidence, but an amusing one) and created a system very similar to the one he'd been working on at DEC. DEC sued and eventually they settled. As I recall, one of the terms of the settlement was that Microsoft would support NT on Alpha.
It's more than that. For performance, in Smalltalk-80, each window was allowed to write to a portion of the display buffer directly. The bit bilt operation was invented specifically to get the required performance for Smalltalk-72 and was implemented in microcode for Smalltalk-74. When each window has a rectangle that it is allowed to draw into, it's trivial for the system to do clipping for the drawing - just check that the destination of the bit blit is within that rectangle. Once you have overlapping windows, you have to check that the destination is within that rectangle and isn't within anyone else's rectangle. More importantly, you also have to split up some blits because you can't just copy an entire rectangle, you have to copy 2 subrectangles to get the non-overlapped part of the shape. This imposes a quite serious performance penalty (when you're talking about something like an Alto - less so when you're talking about an 8MHz 68000).
Just use the same textbooks as 30+ years ago. Pre-university mathematics hasn't changed that much.
This is really where the open source model should be shining. If you're buying books to 100,000 students, then really you should be buying the copyright, not paying through the nose for each copy. As an author, I'd happily take a $30K up-front payment to write a textbook and hand over the complete writes to the country's education system. Then can then do a big print run initially, and a smaller run each year to replace ones that wear out. If they need to make corrections, they can print errata pages for the existing copies and just fix them in the new edition so when the old ones wear out they're replaced with ones with the fixes. And, of course, since they own the copyright they can give students PDF versions to keep.
Links from ACs have rel="nofollow" on them, so neither of your posts will do anything to the page rank. This used to be disabled for users with excellent karma, let's see if it still is...
Of course they do. How do you think the JavaScript JIT works? Well, technically it's not assembly, it's machine code, but that's even more architecture-specific.
I wouldn't be surprised if his claim is true. With the G4, Apple put a lot of effort into optimising stuff to use AltiVec because that was the only way of making things faster than their x86 equivalent: AltiVec-optimised code was usually faster than SSE code and was always a lot faster than scalar x86 / x87 code. With the Intel switch, they made a lot less effort to do SSE optimisations. A quad G5 running software optimised for AltiVec is likely to be faster than an i7 using scalar code.
So does VM. No cap at all between 9pm and 10am. And, from what I've heard from people who switched, plus.net has even worse customer support than VM (e.g. two weeks to send an engineer out to fix a problem, when the worst I've seen with VM is 2 days, and that was when calling after 5pm to report it).
I'm currently paying VM £21/month for their cheapest broadband service. The most stringent part of the cap is 1500MB/day between 4pm and 9pm, which I can occasionally go over if I watch an iPlayer HD film in the evening. After 9pm there is no cap. With Demon, I'd have a 50GB/month cap for a slightly higher price (closer to the VM price for 20Mb/s, which comes with a 7000MB/day cap for peak times) once you factor in the cost of BT line rental. 1.5GB/day for 30 days is 45GB, so the total cap for Demon is only very slightly greater than the evening-peak-time cap for VM. If I saturated my line at off-peak times with VM, and used up to the cap at on-peak times, then I would be using 67GB per day. In other words, VM lets me download more per day than Demon does per month (I've never actually done this, but I have downloaded over 20GB in a single day when backing up a remote server to my home machine).
Why does it take you 2.8 GB of RAM to do something that used to be done in 32 MB with room to spare?
In 1994 / 1995, the web was just about HTML 2.0, but mainly HTML 1.0 plus browser extensions. I've written code that will happily handle pretty much any web page from that era (except frames), and it was under a thousand lines of code. Web pages were designed to load in 10 seconds on a 14.4kbps modem, so were typically very small and contained very few images, no scripts, and certainly no video.
Now, HTML documents are large. This Slashdot page is 1MB of text plus markup alone. On your 32MB machine, it would be using a lot of the memory just to contain the text. If you tried to build a DOM tree then you'd likely be using another 3-4MB. And that's before you even start rendering images. Go to a photo sharing site and you'll see images that are 40+MB when uncompressed - your 32MB machine will start thrashing just showing the image...
If you're unhappy with Virgin Media, who have you found in the UK that doesn't have equally (or more!) aggressive caps? Some ADSL providers have bigger caps for the same speed, but they charge twice as much.
I also wonder how many of these downloaders are using the software for illegal sharing. It sounds like the sort of system that would be great for sharing files in a small company (easier to configure than a file server and VPN) or sharing photos with friends. Hopefully the FreedomBox will ship something similar...
Slashdot Mirror
No, Khamenei is part of the extreme faction that I mentioned in my post. The current posturing by the USA is driving support to him and others like him.
Finland has Nokia and just across the border there is Ericsson. This means that the local telephone companies get to test a lot of new networking hardware before it goes into mass production, which lowers their costs. The problem in much of the rest of the world is that phone companies sold data plans for early smartphones that had tiny screens and could just about do web browsing if you had a lot of patience and just about managed email. An unlimited plan for one of these phones was well under 1GB - irrespective of how fast the network was, you just couldn't consume much data on them. Then people took the same plans and started using modern smartphones with them. A single YouTube video on a modern phone will use more data than a month with an early Blackberry, for example. The networks were very slow at adapting, and are now trying to readjust their prices to levels that actually make sense.
One of the important prerequisites for a free market is informed customers. I have no problem at all with ISPs providing caps. If everyone saturated their connections 24/7 then they would not be able to provide the service, and the cost of actually providing that amount of bandwidth to everyone would be far greater than most customers are able to afford. The problem is not the capping, it is misleading advertising. If you are going to offer 5GB per month, advertise 5GB per month. If you are going to offer 50GB per month, advertise 50GB per month. If you are going to deploy a transparent proxy that resamples images, specify that. If you are going to block access to certain sites, or only permit HTTP traffic, don't say that you provide Internet access. Tell people exactly what service you will provide and allow potential customers to decide whether they think it is worth what you are charging.
Of course that sort of glosses over the assistance the US provided the UK prior to entering the war as well.
You mean assistance like loaning money, loans that were only paid off a few years ago? Or perhaps assistance like allowing British nuclear physicists to work on the Manhattan Project, and then not allowing them to share any of the results with the UK after the war?
Except it's just a simple translation which you are welcome to verify
The literal translation was to wipe the nation of Israel off the map. You can interpret that in several ways. One is to believe that it means killing everyone there. This is the spin that you are reporting. The other is to realise that Iran has never accepted the legitimacy of Israel as a state and wishes to see the state of Israel eliminated as a legal entity. This is the interpretation promoted by Iran.
It's also important to consider the context of the speech. Iran is a Persian country surrounded by Arabic countries. There is a couple of thousand years of history of conflict between the two groups (and not just distant history - remember Iraq invading Iran?). The only way Iran survives in the region is by reminding everyone else that Israel is a more important enemy than them, and that they're on everyone else's side. Picture the scrawny kid that hangs around with the bullies and shouts insults at other people to show how much they are a part of the group. That's Iran's international political status, but what about internally? They have a complex political structure, where the fundamentalist muslims are not a majority, but are influential. The leadership has to, at least publicly, appear to be acceptable to this group or lose power, in much the same way that the US President has to pander to the Christian right in America.
There are other similarities. Both countries have a moderately sane leadership that needs to pander to extreme fundamentalists to remain in power. The odd thing is that the current US strategy could have been designed to destabilise Iran and push the moderate elements out of power.
It doesn't just need to be a rocket, it needs to be a really big rocket. To deflect something potentially civilisation-destroying, you want to do it as early as possible. Inside lunar orbit is much too late. That means that you need to get a rocket out to it, and then match velocities (just ramming it won't be enough) and deliver the explosive to a sensible location. Ideally, you want to attach to the side and push it, rather than try to blow it up. We're talking something that would make the Apollo rockets look small...
We do have these laws, but I have yet to see them enforced against a US-based company. Even one with a significant UK presence, such as Google.
Did you look at the list of files?
Yes...
libgnustep-base-dev clearly includes files like NSObject.h.
So? NSObject.h is part of the Foundation framework. It isn't part of the Objective-C runtime. The relevant headers are things like objc.h and runtime.h. The relevant library is libobjc.so.
Unless Debian went out of their way to write their own shims to reproduce the functionality of the GNUstep runtime based on the GNU objc runtime (which uses Object instead of NSObject), it seems far more likely to me that the GNUstep runtime is included in the libgnustep-base package
Uh, what? That doesn't even make sense.
Both runtimes can co-exist just fine, to the best of my knowledge.
Not linked to the same binary. They implement a largely overlapping set of symbols. Linking both causes massive breakage.
But first you should make sure you're correct, because, as I said, it seems extremely implausible.
Look at the source for the GNUstep runtime. Now look in the Debian package. Note that it is not there. Perhaps you should actually check that you know what you are talking about before being patronising.
I should possibly mention that I am the maintainer of the GNUstep Objective-C runtime, so there's a pretty good chance that I know what I'm talking about...
A grand a year is assuming a class of at least about 250 students and all of them buying the book (instead of borrowing it from the library or sharing). And that's for at least a month's worth of work - at the speed most lecturers write it's 2-3 months. Hardly a good return on investment.
I absolutely agree that one can focus too much on basic arithmetic facts. On the other hand, if you can't do them at all, it's really hard to get anything out of higher math.
I think there are two problems. One is teaching too much arithmetic early, and the other is calling it maths. This means that students think that maths is boring. My mother used to teach primary school children. Her terms and mine didn't quite line up, so I'd often go in at the end of her term and help out, and I kept doing this when I was at university (and had even shorter terms). One term, I taught them some game theory. I will always remember the comment from one of them: 'I thought you were meant to teach us maths. This isn't maths, this is fun.' Game theory is important to understanding a whole raft of things in the real world, but I've never come across a school that teaches it as part of the normal syllabus.
Things like game theory, graph theory, and formal logic are all useful and don't require very much arithmetic. You could easily start teaching them to primary-school children alongside arithmetic, and get them interested in maths early on.
The other answer is that a lot of aircraft simply lose altitude quite quickly when flying upside down. It's pretty common in aerobatics to expect to lose 1000 feet over a short period of inverted flight.
Why would they bother? My first book is used by two university courses that I know of, and the income from that would not even be close to making it worth writing the book. Even if you make everyone in a class of 100 buy the book every year, the author's royalty is going to be pretty small. You'd be lucky to make $1,000/year as a result. Given the time required to write the book, this seems like a pretty bad idea.
Search for Dave Cutler, DEC, Microsoft and Lawsuit. Cutler left DEC and went to work on Windows NT (the WNT acronym being V+1, M+1, S+1 is a coincidence, but an amusing one) and created a system very similar to the one he'd been working on at DEC. DEC sued and eventually they settled. As I recall, one of the terms of the settlement was that Microsoft would support NT on Alpha.
It's more than that. For performance, in Smalltalk-80, each window was allowed to write to a portion of the display buffer directly. The bit bilt operation was invented specifically to get the required performance for Smalltalk-72 and was implemented in microcode for Smalltalk-74. When each window has a rectangle that it is allowed to draw into, it's trivial for the system to do clipping for the drawing - just check that the destination of the bit blit is within that rectangle. Once you have overlapping windows, you have to check that the destination is within that rectangle and isn't within anyone else's rectangle. More importantly, you also have to split up some blits because you can't just copy an entire rectangle, you have to copy 2 subrectangles to get the non-overlapped part of the shape. This imposes a quite serious performance penalty (when you're talking about something like an Alto - less so when you're talking about an 8MHz 68000).
Just use the same textbooks as 30+ years ago. Pre-university mathematics hasn't changed that much.
This is really where the open source model should be shining. If you're buying books to 100,000 students, then really you should be buying the copyright, not paying through the nose for each copy. As an author, I'd happily take a $30K up-front payment to write a textbook and hand over the complete writes to the country's education system. Then can then do a big print run initially, and a smaller run each year to replace ones that wear out. If they need to make corrections, they can print errata pages for the existing copies and just fix them in the new edition so when the old ones wear out they're replaced with ones with the fixes. And, of course, since they own the copyright they can give students PDF versions to keep.
crap that doesn't work.
The best way to read Doctorow is to treat his books as essays describing possible futures, not as stories.
Don't tell me they put ASM in a web browser.
Of course they do. How do you think the JavaScript JIT works? Well, technically it's not assembly, it's machine code, but that's even more architecture-specific.
I wouldn't be surprised if his claim is true. With the G4, Apple put a lot of effort into optimising stuff to use AltiVec because that was the only way of making things faster than their x86 equivalent: AltiVec-optimised code was usually faster than SSE code and was always a lot faster than scalar x86 / x87 code. With the Intel switch, they made a lot less effort to do SSE optimisations. A quad G5 running software optimised for AltiVec is likely to be faster than an i7 using scalar code.
So does VM. No cap at all between 9pm and 10am. And, from what I've heard from people who switched, plus.net has even worse customer support than VM (e.g. two weeks to send an engineer out to fix a problem, when the worst I've seen with VM is 2 days, and that was when calling after 5pm to report it).
I'm currently paying VM £21/month for their cheapest broadband service. The most stringent part of the cap is 1500MB/day between 4pm and 9pm, which I can occasionally go over if I watch an iPlayer HD film in the evening. After 9pm there is no cap. With Demon, I'd have a 50GB/month cap for a slightly higher price (closer to the VM price for 20Mb/s, which comes with a 7000MB/day cap for peak times) once you factor in the cost of BT line rental. 1.5GB/day for 30 days is 45GB, so the total cap for Demon is only very slightly greater than the evening-peak-time cap for VM. If I saturated my line at off-peak times with VM, and used up to the cap at on-peak times, then I would be using 67GB per day. In other words, VM lets me download more per day than Demon does per month (I've never actually done this, but I have downloaded over 20GB in a single day when backing up a remote server to my home machine).
Why does it take you 2.8 GB of RAM to do something that used to be done in 32 MB with room to spare?
In 1994 / 1995, the web was just about HTML 2.0, but mainly HTML 1.0 plus browser extensions. I've written code that will happily handle pretty much any web page from that era (except frames), and it was under a thousand lines of code. Web pages were designed to load in 10 seconds on a 14.4kbps modem, so were typically very small and contained very few images, no scripts, and certainly no video.
Now, HTML documents are large. This Slashdot page is 1MB of text plus markup alone. On your 32MB machine, it would be using a lot of the memory just to contain the text. If you tried to build a DOM tree then you'd likely be using another 3-4MB. And that's before you even start rendering images. Go to a photo sharing site and you'll see images that are 40+MB when uncompressed - your 32MB machine will start thrashing just showing the image...
If you're unhappy with Virgin Media, who have you found in the UK that doesn't have equally (or more!) aggressive caps? Some ADSL providers have bigger caps for the same speed, but they charge twice as much.
I also wonder how many of these downloaders are using the software for illegal sharing. It sounds like the sort of system that would be great for sharing files in a small company (easier to configure than a file server and VPN) or sharing photos with friends. Hopefully the FreedomBox will ship something similar...