People who are prepared to pay for dedicated links with service level agreements will get better service than people who buy cheap "broadband" connections.
At least patents only last about 20 years, which while an annoyingly long time that is short enough that the original physical media on which content is sold and the devices to read it will likely last that long.
Copyrights OTOH seem to keep receiving retroactive extensions and afaict the anti-cirumvention stuff has no expiry at all so they are much bigger threats to preservation of our cultural history.
I remember spending ages googling for "cgi proxy" and looking for one that the filter hadn't categorised as "proxy avoidance systems". I wasn't actually interested in porn but many other sites were blocked too on the connections I was using.
The biggest problem I see with this is it's essentiall imposing an extra burden on operators of wifi hotspots. For the big guys who sell wifi hotspot services this is unlikely to be a problem, for the little guys who want to offer free wifi independently it is likely to drive up their costs significantly as rather than just buying an off the shelf access point they have to buy some kind of commercial filtering system as well.
There is also the question of who is responsible when someone does manage to get porn through the clearly flawed filter.
According to the reference manual there is a "boot button" which will force microSD boot and it will also boot from microSD if the eMMC is "empty" (they don't seem to define exactly what is meant by "empty".
The Pi has set a new benchmark for how cheap this sort of device "should" be. Other than the slight processor upgrade and the addition of the emmc nearly every change had cost-cutting as at least part of the reason behind it to bring the device down from "twice the cost of a Pi" to "slightly more expensive than the Pi but in the same ballpark".
I suspect the extra cost of a gigabit phy and magjack over a 10/100 phy and magjack fitted neither the budget nor the goals of the device.
Afaict the only way to buy the thing is direct from hardkernel in korea so that means I have to pay a pretty steep shipping charge AND then pay the carrier a fee for collecting the VAT*. Also iirc the serial console port is somewhat strange** and AIUI while some people are working on framebuffer console support it doesn't work out of the box (and even if it did the thing uses the u-boot based bootloader which you can only talk to over serial) which pretty much makes their serial debug board mandatory if you want to do any tinkering with the OS (rather than just download someone else's media center build and run it). The PSU connector is also as you say relatively unusual and the HDMI connector is a micro type. So overall even if I already have a microsd card and regular perhipherals arround the real cost ends up substantially higher (on the order of double) than the advertised price.
Don't get me wrong, I have a U2 and if you want a reasonablly affordable arm board with lots of CPU power and 2GB of ram and are prepared to put up with using USB for everything the U2 is a good buy. That is not the goal of either the beaglebone or the Pi. The Pi was designed to provide a computer cheap enough that kids could own it themselves. The beaglebone was designed for hardware hackers who want to put an embedded linux system in their project. Their designs reflect those goals.
* I also have to pay the VAT itself of course but I have to pay that regardless. ** 1.8V and on a weird connector rather than a generic pin header.
Despite the fact that two of those modules are listed in the "desktop memory" section they are all listed as being "registered". Afaict at least in the intel world "registered" memory can only be used with "server" platforms.
We don't even have true 64-bit x86-64 processors yet. While programmers are told to* treat pointers as 64-bit in the current implementation (reffered to as a "48-bit implementation" there are only 47 usable bits for user-mode pointers**. That is enough to map 128 terabytes to one process, afaict the most ram you can currently get in a PC architecture machine is 2 terabytes.
If we assume the largest available memory size doubles every 1.5 years and we want to be able to map all the memory to one process then we have 9 years until the current implementation is used up and another 24 years after that before a "full 64-bit" (with one bit used to distinguish between kernel and user mode) implementation is used up.
* Of course just because programmers are told to do something doesn't mean they will http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=642750 ** A 48th bit is used to differentiate kernel and user addresses. The number is then sign-extended to produce a 64-bit number.
Every 64-bit platform i'm aware of still has a 32-bit int. There may be some software that will waste memory on unix like systems when it uses "long" (which is typically 64-bit on 64-bit unix like systems) where a 32-bit value is fine but I doubt that is significant in the grand scheme of things.
The code itself is usually slightly bigger on x86-64 than on x86 which is probablly what the GP was reffering to but in the grand scheme of things code is usually pretty small and the greater efficiencies for position independent code offset this by reducing the chance of multiple copies of the same code being loaded at once due to load time relocations.
The real problem is pointer heavy code. If a program uses data structures that are mostly made up of pointers (or integers that could potentially contain a typecasted pointer and therefore need to be pointer-sized) then those data structures will nearly double in size on x64.
Seriously, anytime Asus is feeling poor, they can release a Crosshair motherboard that takes 64 GB or perhaps 128 GB of RAM.
How much ram we can put in our desktops is not really up to motherboard manufacturers like ASUS, it's up to the CPU and RAM manufacturerers.
Current intel mainstream desktop CPUs support four DIMMs and current high end high end desktop CPUs support eight DIMMS. Afaict the largest DIMM of desktop memory* currently available is 8GB. So the current limit is 32GB for mainstream desktop and 64GB for high end desktop. I belive that the high end desktop stuff theoretically supports 128GB but noone makes the DIMMs needed to do it yet.
Workstation/server platforms can take a lot more than that both through supporting more DIMMs and through supporting types of DIMM that come in higher capacities. I've seen systems that claim support for up to 2TB of ram.
AIUI an x86 process running on an x64 linux kernel gets damn near 4GB of usable virtual address space. I presume the same applies to x32 processes running on that same kernel.
On a 32-bit kernel as you say 3GB is typical. There were "4G/4G" patches at one stage to increase this but afaict they never made it into mainline.
In my experience most hardware that works with other versions of x64 windows works fine on XP x64. The only two exceptions I ran into was the data translation DT9816 (which worked with some APIs but not others, go figure) and the NI mydaq (for which the software refused to install at all). Remember from a driver point of view XP x64 is basically the same as server 2003 x64 so all the core hardware that is used in both clients and servers is well supported.
As for adoption I know of a few dedicated simulation/number crunching boxes at university running it but I don't know anyone else who uses it as the OS on their main office desktop.
The cynic in me wonders if that is related to companies buying new computers with windows 8 and wanting to run windows XP on them. Afaict the main reason to buy windows volume licenses is that they offer greater downgrade rights than OEM copies.
There are two types of applet, "trusted" and "untrusted"
"untrusted" applets show no warning on startup and are run in a sandbox that is supposed to limit their access to your computer and network. Unfortunately that sandbox has proven time and time again to have bugs that provide ways for the code inside to "escape" the sandbox and do what it wants to your computer.
"trusted" applets show that warning on startup. Then if the user clicks yes the applet gets the ability to do whatever the hell it likes.
Nothing about bitcoin would prevent fractional reserve banking, however the nature of bitcoin would make it trivial for a reckless bank to extend their reserve all the way down to 0% and not disclose this until and unless a run on said bank caused a bank failure.
For fractional reserve banking to work the the bank needs to have a means of investing (whether through loans or equity) which has a near certainty of a positive overall return in the currency they are working in. Given bitcoins wild swings in value relative to other currencies and comodities this is likely to be virtually impossible to ensure.
As such i'd expect that even if it was run honestly and transparently any fractional reserve bitcoin bank is very likely to go bankrupt.
I also expect that we will see people running what they claim are fractional reserve bitcoin banks but are really bitcoin ponzi schemes.
If you rely on someone else to store a bitcoin balance for you rather than keeping the bitcoins yourself you are an idiot.
Massive ammounts of contracts are written for pre-determined ammounts of dollars, euros, pounds etc. That makes the value of those currencies somewhat "sticky", yes it does change over time but usually fairly slowly. If it changes too rapidly governments will sometimes step in to halt the shift.(e.g. http://www.guardian.co.uk/business/2011/sep/06/switzerland-pegs-swiss-franc-euro )
Bitcoin has none of that, it just floats to whereever supply and demand take it. Speculators consistute some of that supply and demand but not all of it. Depending on their strategy they may either increase or decrease volatility
Looks like packages.debian.org isn't updating properly. Currently wheezy has a 3.2.41 based kernel. My understanding is they intend to follow the 3.2 kernel series in wheezy.
Are you saying that warty warthog is not negative?;)
More recently i've noticed a pattern though, LTS releases get very positive and usually stable sounding names (Dapper drake, Hardy heron, Lucid lynx, Precise pangolin). The releases in between tend to get names that imply more instability.
Even if we ignore the argument about whether users should have the freedom to mix free and non-free components in a peice of software. One big problem with the GPL IMO is it has no exceptions for software under other free licenses. So I can have two peices of software both of which are "free" per the FSFs definition yet i'm forbidden from mixing them to create a single larger work.
Hurd didn't make the cut for the wheezy release, apparently the hurd porters plan some sort of unofficial release alongside wheezy but what form this will take (hurd is not and never has been in testing for wheezy) has not been made clear.
AIUI kfreebsd will be included but is still considered to be a "techology preview" rather than a full production release. They seem to be shipping both 8.x and 9.x kernels.
In an ideal world software vendors wouldn't put users in a position of choosing between trusting their internet connection and not getting the software. Certification authorities would make damn sure they were issuing certificates to the right entity. Credit card companies would move away from a system where the dominent way of making an online payment is to give the vendor a code that lets them take unlimited money from your account. Users would directly enter the https url or at least carefully check the https url they had been redirected to after entering the http url. Legitimate vendors wouldn't spread their services across multiple domains so it would be easy to tell the difference between the vendor's legitimate site and a scam domain. Websites that were going to use a third party payment gateway would bring users into a secure area of their own site before giving them the url of the payment gateway.
And of course even if we take "fastest" to mean "highest bandwidth" it's only true if you limit yourself to connections that are sold at consumer/small buiness prices.
Sure, however once the speed of the connection from your house to your ISPs PoP gets over a certain speed it stops being a significant factor in the overall user experiance. Other factors including limited bandwidth at the server, bandwidth limitations in the ISPs (both client side and server side) networks, limited speed of your hard drive, human time to select things in the user interface, time for servers to process requests and so-on become the limiting factors on the overall performance.
Amusingly if you are booking in advance it's occasionally cheaper to travel first class than standard class on trains. This is due to the way each type of advance ticket is sold in limited numbers and (at least for manchester to london) the cheapest first class advance tickets are slightly cheaper than standard class off-peak tickets (I bought one of these once). There are also relatively cheap weekend upgrades available on some routes (I haven't actually ever bought one of these though i've been tempted a couple of times).
And the experiance on virgin trains is lovely, large seats, free wifi, free refreshments, seat groups with proper tables for 1, 2 or 4 people (while standard class only has proper tables for four people and "airline style" seats for two peoples with those flip down things that are too small to put a laptop on). The ride is smooth and the curves are gentle meaing motion sickness is far less of a problem when doing stuff on a train than when doing stuff in a car.
Overall if you either get a good advance deal or are rich enough that the cost is neligable to you it's hard to beat first class train travel as a comfortable way of getting to/from london.
OTOH I don't know why anyone would buy a first class ticket from manchester to leeds because the "first class" on those trains is not deserving of the name
What happens at least with O2 is that the balance itself lasts practically forever* BUT you get bonuses for topping up which are time limited and you can also buy "bolt ons" using your account balance which again are time limited. They also have call charging models where the first few minuites of calls on any given day cost a lot more than later minuites. Similally for data if you don't have a current package data is charged at a high rate up to a daily charging cap (£1 IIRC) and then there are no automatic data charges on that day thereafter**.
So provided you keep mobile data turned off you can keep a mobile active while spending very little (nothing if other people call you from time to time) but if you then take that mobile, enable mobile data and start making calls and browsing the internet with it you can easilly burn up a couple of pounds (or more) of credit in one session.
* IIRC you have to use the account in some way (calls both incoming and outgoing count, I think other stuff does too) every few months to stop the account expiring. ** There is a daily data cap after which they will cut your data off until either the next day starts, you buy a data bolt on or you top up and get data as a reward. I'm not sure if how much data this cap represents is documented anywhere (I wasn't aware of it until I hit it and i've only hit it once).
Slashdot Mirror
I wonder if the military will get better.
People who are prepared to pay for dedicated links with service level agreements will get better service than people who buy cheap "broadband" connections.
At least patents only last about 20 years, which while an annoyingly long time that is short enough that the original physical media on which content is sold and the devices to read it will likely last that long.
Copyrights OTOH seem to keep receiving retroactive extensions and afaict the anti-cirumvention stuff has no expiry at all so they are much bigger threats to preservation of our cultural history.
Such guide will show you how to bypass the filter
I remember spending ages googling for "cgi proxy" and looking for one that the filter hadn't categorised as "proxy avoidance systems". I wasn't actually interested in porn but many other sites were blocked too on the connections I was using.
The biggest problem I see with this is it's essentiall imposing an extra burden on operators of wifi hotspots. For the big guys who sell wifi hotspot services this is unlikely to be a problem, for the little guys who want to offer free wifi independently it is likely to drive up their costs significantly as rather than just buying an off the shelf access point they have to buy some kind of commercial filtering system as well.
There is also the question of who is responsible when someone does manage to get porn through the clearly flawed filter.
According to the reference manual there is a "boot button" which will force microSD boot and it will also boot from microSD if the eMMC is "empty" (they don't seem to define exactly what is meant by "empty".
https://github.com/CircuitCo/BeagleBone-Black/blob/master/BBB_SRM.pdf?raw=true
The Pi has set a new benchmark for how cheap this sort of device "should" be. Other than the slight processor upgrade and the addition of the emmc nearly every change had cost-cutting as at least part of the reason behind it to bring the device down from "twice the cost of a Pi" to "slightly more expensive than the Pi but in the same ballpark".
I suspect the extra cost of a gigabit phy and magjack over a 10/100 phy and magjack fitted neither the budget nor the goals of the device.
Afaict the only way to buy the thing is direct from hardkernel in korea so that means I have to pay a pretty steep shipping charge AND then pay the carrier a fee for collecting the VAT*. Also iirc the serial console port is somewhat strange** and AIUI while some people are working on framebuffer console support it doesn't work out of the box (and even if it did the thing uses the u-boot based bootloader which you can only talk to over serial) which pretty much makes their serial debug board mandatory if you want to do any tinkering with the OS (rather than just download someone else's media center build and run it). The PSU connector is also as you say relatively unusual and the HDMI connector is a micro type. So overall even if I already have a microsd card and regular perhipherals arround the real cost ends up substantially higher (on the order of double) than the advertised price.
Don't get me wrong, I have a U2 and if you want a reasonablly affordable arm board with lots of CPU power and 2GB of ram and are prepared to put up with using USB for everything the U2 is a good buy. That is not the goal of either the beaglebone or the Pi. The Pi was designed to provide a computer cheap enough that kids could own it themselves. The beaglebone was designed for hardware hackers who want to put an embedded linux system in their project. Their designs reflect those goals.
* I also have to pay the VAT itself of course but I have to pay that regardless.
** 1.8V and on a weird connector rather than a generic pin header.
Despite the fact that two of those modules are listed in the "desktop memory" section they are all listed as being "registered". Afaict at least in the intel world "registered" memory can only be used with "server" platforms.
A long time.
We don't even have true 64-bit x86-64 processors yet. While programmers are told to* treat pointers as 64-bit in the current implementation (reffered to as a "48-bit implementation" there are only 47 usable bits for user-mode pointers**. That is enough to map 128 terabytes to one process, afaict the most ram you can currently get in a PC architecture machine is 2 terabytes.
If we assume the largest available memory size doubles every 1.5 years and we want to be able to map all the memory to one process then we have 9 years until the current implementation is used up and another 24 years after that before a "full 64-bit" (with one bit used to distinguish between kernel and user mode) implementation is used up.
* Of course just because programmers are told to do something doesn't mean they will http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=642750
** A 48th bit is used to differentiate kernel and user addresses. The number is then sign-extended to produce a 64-bit number.
Every 64-bit platform i'm aware of still has a 32-bit int. There may be some software that will waste memory on unix like systems when it uses "long" (which is typically 64-bit on 64-bit unix like systems) where a 32-bit value is fine but I doubt that is significant in the grand scheme of things.
The code itself is usually slightly bigger on x86-64 than on x86 which is probablly what the GP was reffering to but in the grand scheme of things code is usually pretty small and the greater efficiencies for position independent code offset this by reducing the chance of multiple copies of the same code being loaded at once due to load time relocations.
The real problem is pointer heavy code. If a program uses data structures that are mostly made up of pointers (or integers that could potentially contain a typecasted pointer and therefore need to be pointer-sized) then those data structures will nearly double in size on x64.
Seriously, anytime Asus is feeling poor, they can release a Crosshair motherboard that takes 64 GB or perhaps 128 GB of RAM.
How much ram we can put in our desktops is not really up to motherboard manufacturers like ASUS, it's up to the CPU and RAM manufacturerers.
Current intel mainstream desktop CPUs support four DIMMs and current high end high end desktop CPUs support eight DIMMS. Afaict the largest DIMM of desktop memory* currently available is 8GB. So the current limit is 32GB for mainstream desktop and 64GB for high end desktop. I belive that the high end desktop stuff theoretically supports 128GB but noone makes the DIMMs needed to do it yet.
Workstation/server platforms can take a lot more than that both through supporting more DIMMs and through supporting types of DIMM that come in higher capacities. I've seen systems that claim support for up to 2TB of ram.
* DDR3, unregistered non-ecc.
3GB typically.
AIUI an x86 process running on an x64 linux kernel gets damn near 4GB of usable virtual address space. I presume the same applies to x32 processes running on that same kernel.
On a 32-bit kernel as you say 3GB is typical. There were "4G/4G" patches at one stage to increase this but afaict they never made it into mainline.
In my experience most hardware that works with other versions of x64 windows works fine on XP x64. The only two exceptions I ran into was the data translation DT9816 (which worked with some APIs but not others, go figure) and the NI mydaq (for which the software refused to install at all). Remember from a driver point of view XP x64 is basically the same as server 2003 x64 so all the core hardware that is used in both clients and servers is well supported.
As for adoption I know of a few dedicated simulation/number crunching boxes at university running it but I don't know anyone else who uses it as the OS on their main office desktop.
Volume licensing is way, way up.
The cynic in me wonders if that is related to companies buying new computers with windows 8 and wanting to run windows XP on them. Afaict the main reason to buy windows volume licenses is that they offer greater downgrade rights than OEM copies.
There are two types of applet, "trusted" and "untrusted"
"untrusted" applets show no warning on startup and are run in a sandbox that is supposed to limit their access to your computer and network. Unfortunately that sandbox has proven time and time again to have bugs that provide ways for the code inside to "escape" the sandbox and do what it wants to your computer.
"trusted" applets show that warning on startup. Then if the user clicks yes the applet gets the ability to do whatever the hell it likes.
Nothing about bitcoin would prevent fractional reserve banking, however the nature of bitcoin would make it trivial for a reckless bank to extend their reserve all the way down to 0% and not disclose this until and unless a run on said bank caused a bank failure.
For fractional reserve banking to work the the bank needs to have a means of investing (whether through loans or equity) which has a near certainty of a positive overall return in the currency they are working in. Given bitcoins wild swings in value relative to other currencies and comodities this is likely to be virtually impossible to ensure.
As such i'd expect that even if it was run honestly and transparently any fractional reserve bitcoin bank is very likely to go bankrupt.
I also expect that we will see people running what they claim are fractional reserve bitcoin banks but are really bitcoin ponzi schemes.
If you rely on someone else to store a bitcoin balance for you rather than keeping the bitcoins yourself you are an idiot.
Massive ammounts of contracts are written for pre-determined ammounts of dollars, euros, pounds etc. That makes the value of those currencies somewhat "sticky", yes it does change over time but usually fairly slowly. If it changes too rapidly governments will sometimes step in to halt the shift.(e.g. http://www.guardian.co.uk/business/2011/sep/06/switzerland-pegs-swiss-franc-euro )
Bitcoin has none of that, it just floats to whereever supply and demand take it. Speculators consistute some of that supply and demand but not all of it. Depending on their strategy they may either increase or decrease volatility
Looks like packages.debian.org isn't updating properly. Currently wheezy has a 3.2.41 based kernel. My understanding is they intend to follow the 3.2 kernel series in wheezy.
and their choice of words is not negative.
Are you saying that warty warthog is not negative? ;)
More recently i've noticed a pattern though, LTS releases get very positive and usually stable sounding names (Dapper drake, Hardy heron, Lucid lynx, Precise pangolin). The releases in between tend to get names that imply more instability.
Even if we ignore the argument about whether users should have the freedom to mix free and non-free components in a peice of software. One big problem with the GPL IMO is it has no exceptions for software under other free licenses. So I can have two peices of software both of which are "free" per the FSFs definition yet i'm forbidden from mixing them to create a single larger work.
Hurd didn't make the cut for the wheezy release, apparently the hurd porters plan some sort of unofficial release alongside wheezy but what form this will take (hurd is not and never has been in testing for wheezy) has not been made clear.
AIUI kfreebsd will be included but is still considered to be a "techology preview" rather than a full production release. They seem to be shipping both 8.x and 9.x kernels.
In an ideal world software vendors wouldn't put users in a position of choosing between trusting their internet connection and not getting the software. Certification authorities would make damn sure they were issuing certificates to the right entity. Credit card companies would move away from a system where the dominent way of making an online payment is to give the vendor a code that lets them take unlimited money from your account. Users would directly enter the https url or at least carefully check the https url they had been redirected to after entering the http url. Legitimate vendors wouldn't spread their services across multiple domains so it would be easy to tell the difference between the vendor's legitimate site and a scam domain. Websites that were going to use a third party payment gateway would bring users into a secure area of their own site before giving them the url of the payment gateway.
Sadly we don't live in that world
And of course even if we take "fastest" to mean "highest bandwidth" it's only true if you limit yourself to connections that are sold at consumer/small buiness prices.
Sure, however once the speed of the connection from your house to your ISPs PoP gets over a certain speed it stops being a significant factor in the overall user experiance. Other factors including limited bandwidth at the server, bandwidth limitations in the ISPs (both client side and server side) networks, limited speed of your hard drive, human time to select things in the user interface, time for servers to process requests and so-on become the limiting factors on the overall performance.
Amusingly if you are booking in advance it's occasionally cheaper to travel first class than standard class on trains. This is due to the way each type of advance ticket is sold in limited numbers and (at least for manchester to london) the cheapest first class advance tickets are slightly cheaper than standard class off-peak tickets (I bought one of these once). There are also relatively cheap weekend upgrades available on some routes (I haven't actually ever bought one of these though i've been tempted a couple of times).
And the experiance on virgin trains is lovely, large seats, free wifi, free refreshments, seat groups with proper tables for 1, 2 or 4 people (while standard class only has proper tables for four people and "airline style" seats for two peoples with those flip down things that are too small to put a laptop on). The ride is smooth and the curves are gentle meaing motion sickness is far less of a problem when doing stuff on a train than when doing stuff in a car.
Overall if you either get a good advance deal or are rich enough that the cost is neligable to you it's hard to beat first class train travel as a comfortable way of getting to/from london.
OTOH I don't know why anyone would buy a first class ticket from manchester to leeds because the "first class" on those trains is not deserving of the name
What happens at least with O2 is that the balance itself lasts practically forever* BUT you get bonuses for topping up which are time limited and you can also buy "bolt ons" using your account balance which again are time limited. They also have call charging models where the first few minuites of calls on any given day cost a lot more than later minuites. Similally for data if you don't have a current package data is charged at a high rate up to a daily charging cap (£1 IIRC) and then there are no automatic data charges on that day thereafter**.
So provided you keep mobile data turned off you can keep a mobile active while spending very little (nothing if other people call you from time to time) but if you then take that mobile, enable mobile data and start making calls and browsing the internet with it you can easilly burn up a couple of pounds (or more) of credit in one session.
* IIRC you have to use the account in some way (calls both incoming and outgoing count, I think other stuff does too) every few months to stop the account expiring.
** There is a daily data cap after which they will cut your data off until either the next day starts, you buy a data bolt on or you top up and get data as a reward. I'm not sure if how much data this cap represents is documented anywhere (I wasn't aware of it until I hit it and i've only hit it once).