1: apples restrictive policies on the appstore. If you want an app that does something that apple doesn't like you won't ever get it on a non-jailbroken idevice. 2: even if there wasn't those restrictive policies the third parties would still have to get arround to porting stuff.
IMO if MS want to make a real success of this they should focus on getting big name windows apps (including thier own stuff like office) to work WELL on a tablet. That way you will have a device that can run any windows app in a pinch and yet also has good tablet functionality.
If you're in the UK, you might also want to look at Maplin Electronics mmm I do sometimes use them when i'm in a pinch. the big problem with maplin is that thier range of components has got worse and worse over the years. I do still sometimes use them in a pinch though.
they were even willing to sell single resistors for 7p each, rather expensive for a resistor, seems they have gone up to 17p. Buy 10 at that price and you are up to nearly what rapid want for 100 equivilent resistors*
because you don't have to buy thousands at a time Thousands is a bit of an exaggeration but yeah resistors do tend to come in fairly high quantities (50-100 is a typical pack size).
A bigger problem is minimum order values and/or small order charges. It's fine when your buying everything for project at once but a real pain when I just want one or two components for a hobby project**.
P.S. I didn't mention maplin or rapid in my previous post because the subject at hand was micrcontrollers and afaict both maplin's and rapid's ranges of microcontrollers seem pretty absymal.
* The resistors maplin sell are actually relatively high spec having a power rating of 0.6W (in the conventional 0.25W body size) and a tolerance of 1%, if all you really needed was a cheap carbon film then maplin's price looks even worse.
** when doing stuff for the uni the uni hs deals that get us out of things like minimum order values and small order charges as well as getting us pretty steep discounts
Except the banks could offer a huge monthly discount to the interest rate on immortality loans, for agreeing to such a huge amount and long payment term. Banks are out there to make money. They aren't going to lend any significant some of money at less than (predicted cost of funds on interbank markets)+(premium to cover predicted risk of default).
Or allow the loan balance to increase over time, instead of decreasing, and roll the unpaid interest each month into a balloon payment every few hundred years. Which would of course be defautled on.
A couple of other recent examples of high quality PC games being given away free (as in beer and in one of the cases below also as in OSS).
warzone2100 was both opensourced and given away (unlike most PC games that get opensourced where the code is given away but the game data remains under a restrictive license) fairly recently though admittedly that is rather an old game (still a good one though)
portal was given away at one point though it was a time limited offer and tied to steam.
Perhaps they came up with a margin of error Generally with this kind of thing you don't estimate a hard edged margin of error but rather you estimate a distribution of probabilities, usually a gausian one.
Combine the probailitity distributions of all your inputs and you get a probability distrbution for your output.
I'd agree about taking the numbers with a pinch of salt though since doing it well is reliant on having good knowlage of the error distributions of all your measurements and calculation methods.
No I don't because I don't think it would help anyone.
As you increase the length of a mortgage the cost asymtopically approaches that of an interest only mortgage. In other words beyond a certain point (exactly what point depends on the interest rates) increasing the length of a mortgage will have negligable impact on the monthly cost of the mortgage.
Who even makes 8-bit microcontrollers? Atmel and microchip mainly at least if by 8-bit you mean 8-bit data (most microcontrollers at that level are harvard architectures with wider program memory than data memory).
I haven't used the atmel stuff (which he is using) myself so I can't comment on how easy it is to get. Most of the microchip stuff i've used is held as stock by both RS and Farnell (the two major prototyping parts vendors here in the UK) and microchip will also sell it you directly (though the shipping and handling charges are annoying). In the US you'd probablly want to look at digikey, mouser or newark as a source.
While this is true, the text back in those days was pretty barebones. I couldn't find a screenshot of what the TV output looks like from this device. Well they say they took the firmware for the microcontroller that does the display from the tellymate project ( http://www.batsocks.co.uk/products/Other/TellyMate.htm ) and that project has screenshots (which look pretty barebones) and figures (38x25 which is fairly similar to teletext's 40x24).
Looks like he is only doing composite video not RF. this will reduce the cost and avoid certain reliability (apparently old fasioned RF modulators aren't really stable enough for modern tuners to lock to) and compatbility issues but will mean that some really old TVs aren't compatible.
The trouble is if your cost system is out of whack then people with limited budgets will make descisions that are good for them under the current system but would be bad descisions if things were more accurately accounted for.
In particular i'd think overcharging for storage would cause a lot of local badly backed up storage to pop up (at least that is what seems to happen at the uni I go to).
That is true, however a worldwide power grid would be incrediblly expensive to install. Joining america to eurasia would require either long undersea runs or long runs through inhospitable places like sibera.
It sounds to me like they fell back to not setting them and that caused the default memory limit to be used which is too low.
One of the annoyances with the sun JVM is that it has to have a memory limit set (I belive because of the way the implemented GC) and the default is rather low.
A good example is text areas, all browsers I know of support them but not all browsers will correctly handle unicode text in them. More specifically older browsers (most notably IE for the mac and netscape 4.x) convert the text into a platform specific legacy encoding (and in the process destroy anything not present in that encoding), let the user edit it and then convert it back to unicode. This presents a big problem for unicode based wikis (a handful of users having broken browsers can lead to a low of broken pages).
So if you see a browser that doesn't handle it properly you either want to refuse to let them edit or (as mediawiki does) apply an armouring/dearmouring process to the text to allow editing the ascii parts of pages (and if you are detemined the rest too) without breaking the non-ascii parts.
The trouble is due to the way suns GC is designed it essentially needs to work with a limited size pool of memory. Make the pool too big and you end up with a machine that swaps instead of garbage collecting when memory gets tight under high load and/or a java app unnessacerally hogging memory other apps need. Make it too small and you get out of memory errors.
Bloatware like eclipse needs the pool set large than the default, so there is code that sets that if it sees a VM that it thinks needs it setting. The trouble is the change to the vendor string meant that the VM was no longer being identified as one that needed it.
The thing with java is that it installs support for applets, JWS etc. Therefore it is putting itself in a position of high exposure to security threats and hence needs to be kept updated.
Yes large businesses should be regression testing stuff on their standard configurations but that isn't really feasible for those of us running individual desktops.
Should they? Sometimes different implementations of a "standard" behave differently, sometimes that is because of a bug, sometimes it's because of an ambiguous specification in the standard. Sometimes it's because of something beyond the scope of the standard (e.g. suns VM needs to be told how much memory it's allowed to use upfront or memory hungry applications will fail).
When this happens you have to identify what you are running on so you can tailor your apps behaviour to that of the implementation it is running on.
Would another field have been better? possibly but when the documenation doesn't provide any info on what fields should and shouldn't be used for such purposed
1: they are EXPENSIVE, figure you will be paying many times the cost per gigabyte of ordinary drives. Particually if you buy the SAN vendors drives so you get support. This tends to cancel out more efficiant use of space. 2: Even a 1U server has space for a few drives inside, so if you use a SAN with 1U servers it will probablly take up more space than just putting the drives in the servers. Blades would reduce this issue but come with issues of thier own (e.g. vendor lockin) 3: if something does go wrong with a SAN it means everything has problems at once. This can leave all sorts of IT services down for days as IT scramble to first fix the SAN and then fix everything that depends on the SAN (seen this happen at the uni I go to).
I have my doubts on the power consumption front too. Afaict drives are a negligable part of a modern computers power consumption anyway.
And it may well make economic sense too at least if you are talking about a low end server with a pair of SATA drives (though it depends how much your server vendor rips you off on hard drives).
Upgrading drives later has a lot of costs on top of the raw cost of getting the extra drives.
How much does it cost to get hold of those extra drives? (at uni recently someone told me that the total cost of processing a purchase order worked out to about £40 now admittedly some of that is fixed costs but still it makes you think about how you order stuff) How much does it cost for the server monkey's time to add extra drives? How much does it cost for the sysadmin time to reconfigure the box to use those new drives?
He is almost certainly refering to wholesale electricity, not retail.
One of the big problems right now is that all but the largest consumers are decoupled from the price fluctuations of electricty through the day, therefore price fluctuations don't affect the demand side of the supply/demand balance. However bringing time of day pricing to regular consumers opens up a massive can of worms of it's own.
You will note that Apple is not forcing you to get the 12-core version: 8- and 6- core versions are available at higher speeds. Since intel doesn't offer any 3 core CPUs a 6-core machine would have to be a 1x6 configuration so I wouldn't expect it to be any higher clocked than a 12-core (2x6) version.
Assuming apple offers intel's best processors the fastest 8 core (2x4) configuration would have a faster clock speed than the fastest 12 core (2x6) configuration but only marginally. It wouldn't at all surprise me though if apple were to refuse to sell intels fastest quad-core xeon to make the 6-core option uniqevocically the best one.
There are 1tb SSD drives (OCZ makes one, I believe), but they are in excess of somewhere around $4k Newegg say just over $3K
And it comes in desktop form factor (3.5 inch) whereas you can get a 1TB HDD in large laptop form factor (2.5 inch 13mm high) and a 750GB in regular laptop form factor (2.5 inch 9.5mm high). The largest SSD i've seen in any laptop form factor is 512GB.
more expensive than most people are willing to spend on the computer as a whole. Indeed though desktop users can use a more moderate SSD for the OS etc and a HDD for bulk data.
The big issue is for laptop users who unless it's a monster of a laptop have to choose between a SSD or a HDD.
It's a difficult sort of study to do properly for a few reasons so unless there is strong evidence otherwise (e.g. funding from big media) I'd expect this was simply a case of incompetance.
Reasons why it's a difficult sort of study
1: If you actually download the files to investigate them then you are getting into legally dodgy ground. If you want to download at more than a trickle you will have to upload too which puts you in an even worse position legally. 2: Afaict most legal torrents use their own trackers rather than public ones, so they won't show up in a search that focuses on the big puplic trackers. 3: Afaict big media are trying to deliberately fill P2P systems including bittorrent with fake crap as part of their war on the pirates.
Because as well as capacity there is performance and reliability to consider.
High spin speeds are important to hard drive performance because a large part of the time to perform a small read is the time taken to wait for the HDD to spin to the right position after seeking to the track. Unfortunately high speeds and large platters are not a good combination mechanically.
For SSDs it could be usefull at some point but afaict the main limitation on SSDs right now is that most people aren't prepared to pay what even 1TB of flash costs (cost per gigabyte of SSDs seems relatively constant over a wide range of capacities).
Slashdot Mirror
It's limited by a couple of things.
1: apples restrictive policies on the appstore. If you want an app that does something that apple doesn't like you won't ever get it on a non-jailbroken idevice.
2: even if there wasn't those restrictive policies the third parties would still have to get arround to porting stuff.
IMO if MS want to make a real success of this they should focus on getting big name windows apps (including thier own stuff like office) to work WELL on a tablet. That way you will have a device that can run any windows app in a pinch and yet also has good tablet functionality.
If you're in the UK, you might also want to look at Maplin Electronics
mmm I do sometimes use them when i'm in a pinch. the big problem with maplin is that thier range of components has got worse and worse over the years. I do still sometimes use them in a pinch though.
they were even willing to sell single resistors for 7p each, rather expensive for a resistor,
seems they have gone up to 17p. Buy 10 at that price and you are up to nearly what rapid want for 100 equivilent resistors*
because you don't have to buy thousands at a time
Thousands is a bit of an exaggeration but yeah resistors do tend to come in fairly high quantities (50-100 is a typical pack size).
A bigger problem is minimum order values and/or small order charges. It's fine when your buying everything for project at once but a real pain when I just want one or two components for a hobby project**.
When it comes to prototyping resistors i've found the best thing to do is to get a resistor kit, for prototyping resistors e.g. http://www.rapidonline.com/Electronic-Components/Resistors-Potentiometer/Metal-Film-Resistors/MR25-Metal-film-resistor-kit/65199 .
P.S. I didn't mention maplin or rapid in my previous post because the subject at hand was micrcontrollers and afaict both maplin's and rapid's ranges of microcontrollers seem pretty absymal.
* The resistors maplin sell are actually relatively high spec having a power rating of 0.6W (in the conventional 0.25W body size) and a tolerance of 1%, if all you really needed was a cheap carbon film then maplin's price looks even worse.
** when doing stuff for the uni the uni hs deals that get us out of things like minimum order values and small order charges as well as getting us pretty steep discounts
Except the banks could offer a huge monthly discount to the interest rate on immortality loans, for agreeing to such a huge amount and long payment term.
Banks are out there to make money. They aren't going to lend any significant some of money at less than (predicted cost of funds on interbank markets)+(premium to cover predicted risk of default).
Or allow the loan balance to increase over time, instead of decreasing, and roll the unpaid interest each month into a balloon payment every few hundred years.
Which would of course be defautled on.
A couple of other recent examples of high quality PC games being given away free (as in beer and in one of the cases below also as in OSS).
warzone2100 was both opensourced and given away (unlike most PC games that get opensourced where the code is given away but the game data remains under a restrictive license) fairly recently though admittedly that is rather an old game (still a good one though)
portal was given away at one point though it was a time limited offer and tied to steam.
Perhaps they came up with a margin of error
Generally with this kind of thing you don't estimate a hard edged margin of error but rather you estimate a distribution of probabilities, usually a gausian one.
Combine the probailitity distributions of all your inputs and you get a probability distrbution for your output.
I'd agree about taking the numbers with a pinch of salt though since doing it well is reliant on having good knowlage of the error distributions of all your measurements and calculation methods.
No I don't because I don't think it would help anyone.
As you increase the length of a mortgage the cost asymtopically approaches that of an interest only mortgage. In other words beyond a certain point (exactly what point depends on the interest rates) increasing the length of a mortgage will have negligable impact on the monthly cost of the mortgage.
Who even makes 8-bit microcontrollers?
Atmel and microchip mainly at least if by 8-bit you mean 8-bit data (most microcontrollers at that level are harvard architectures with wider program memory than data memory).
I haven't used the atmel stuff (which he is using) myself so I can't comment on how easy it is to get. Most of the microchip stuff i've used is held as stock by both RS and Farnell (the two major prototyping parts vendors here in the UK) and microchip will also sell it you directly (though the shipping and handling charges are annoying). In the US you'd probablly want to look at digikey, mouser or newark as a source.
While this is true, the text back in those days was pretty barebones. I couldn't find a screenshot of what the TV output looks like from this device.
Well they say they took the firmware for the microcontroller that does the display from the tellymate project ( http://www.batsocks.co.uk/products/Other/TellyMate.htm ) and that project has screenshots (which look pretty barebones) and figures (38x25 which is fairly similar to teletext's 40x24).
Looks like he is only doing composite video not RF. this will reduce the cost and avoid certain reliability (apparently old fasioned RF modulators aren't really stable enough for modern tuners to lock to) and compatbility issues but will mean that some really old TVs aren't compatible.
The trouble is if your cost system is out of whack then people with limited budgets will make descisions that are good for them under the current system but would be bad descisions if things were more accurately accounted for.
In particular i'd think overcharging for storage would cause a lot of local badly backed up storage to pop up (at least that is what seems to happen at the uni I go to).
That is true, however a worldwide power grid would be incrediblly expensive to install. Joining america to eurasia would require either long undersea runs or long runs through inhospitable places like sibera.
It sounds to me like they fell back to not setting them and that caused the default memory limit to be used which is too low.
One of the annoyances with the sun JVM is that it has to have a memory limit set (I belive because of the way the implemented GC) and the default is rather low.
A good example is text areas, all browsers I know of support them but not all browsers will correctly handle unicode text in them. More specifically older browsers (most notably IE for the mac and netscape 4.x) convert the text into a platform specific legacy encoding (and in the process destroy anything not present in that encoding), let the user edit it and then convert it back to unicode. This presents a big problem for unicode based wikis (a handful of users having broken browsers can lead to a low of broken pages).
So if you see a browser that doesn't handle it properly you either want to refuse to let them edit or (as mediawiki does) apply an armouring/dearmouring process to the text to allow editing the ascii parts of pages (and if you are detemined the rest too) without breaking the non-ascii parts.
Not exactly
The trouble is due to the way suns GC is designed it essentially needs to work with a limited size pool of memory. Make the pool too big and you end up with a machine that swaps instead of garbage collecting when memory gets tight under high load and/or a java app unnessacerally hogging memory other apps need. Make it too small and you get out of memory errors.
Bloatware like eclipse needs the pool set large than the default, so there is code that sets that if it sees a VM that it thinks needs it setting. The trouble is the change to the vendor string meant that the VM was no longer being identified as one that needed it.
That's my understanding of the issue anyway.
The thing with java is that it installs support for applets, JWS etc. Therefore it is putting itself in a position of high exposure to security threats and hence needs to be kept updated.
Yes large businesses should be regression testing stuff on their standard configurations but that isn't really feasible for those of us running individual desktops.
Should they?
Sometimes different implementations of a "standard" behave differently, sometimes that is because of a bug, sometimes it's because of an ambiguous specification in the standard. Sometimes it's because of something beyond the scope of the standard (e.g. suns VM needs to be told how much memory it's allowed to use upfront or memory hungry applications will fail).
When this happens you have to identify what you are running on so you can tailor your apps behaviour to that of the implementation it is running on.
Would another field have been better? possibly but when the documenation doesn't provide any info on what fields should and shouldn't be used for such purposed
However SANs have issues of thier own
1: they are EXPENSIVE, figure you will be paying many times the cost per gigabyte of ordinary drives. Particually if you buy the SAN vendors drives so you get support. This tends to cancel out more efficiant use of space.
2: Even a 1U server has space for a few drives inside, so if you use a SAN with 1U servers it will probablly take up more space than just putting the drives in the servers. Blades would reduce this issue but come with issues of thier own (e.g. vendor lockin)
3: if something does go wrong with a SAN it means everything has problems at once. This can leave all sorts of IT services down for days as IT scramble to first fix the SAN and then fix everything that depends on the SAN (seen this happen at the uni I go to).
I have my doubts on the power consumption front too. Afaict drives are a negligable part of a modern computers power consumption anyway.
And it may well make economic sense too at least if you are talking about a low end server with a pair of SATA drives (though it depends how much your server vendor rips you off on hard drives).
Upgrading drives later has a lot of costs on top of the raw cost of getting the extra drives.
How much does it cost to get hold of those extra drives? (at uni recently someone told me that the total cost of processing a purchase order worked out to about £40 now admittedly some of that is fixed costs but still it makes you think about how you order stuff)
How much does it cost for the server monkey's time to add extra drives?
How much does it cost for the sysadmin time to reconfigure the box to use those new drives?
He is almost certainly refering to wholesale electricity, not retail.
One of the big problems right now is that all but the largest consumers are decoupled from the price fluctuations of electricty through the day, therefore price fluctuations don't affect the demand side of the supply/demand balance. However bringing time of day pricing to regular consumers opens up a massive can of worms of it's own.
In my experiance optical mice work just fine on clothing and beds.
You will note that Apple is not forcing you to get the 12-core version: 8- and 6- core versions are available at higher speeds.
Since intel doesn't offer any 3 core CPUs a 6-core machine would have to be a 1x6 configuration so I wouldn't expect it to be any higher clocked than a 12-core (2x6) version.
Assuming apple offers intel's best processors the fastest 8 core (2x4) configuration would have a faster clock speed than the fastest 12 core (2x6) configuration but only marginally. It wouldn't at all surprise me though if apple were to refuse to sell intels fastest quad-core xeon to make the 6-core option uniqevocically the best one.
neither do decent optical mice in my experiance.
There are 1tb SSD drives (OCZ makes one, I believe), but they are in excess of somewhere around $4k
Newegg say just over $3K
And it comes in desktop form factor (3.5 inch) whereas you can get a 1TB HDD in large laptop form factor (2.5 inch 13mm high) and a 750GB in regular laptop form factor (2.5 inch 9.5mm high). The largest SSD i've seen in any laptop form factor is 512GB.
more expensive than most people are willing to spend on the computer as a whole.
Indeed though desktop users can use a more moderate SSD for the OS etc and a HDD for bulk data.
The big issue is for laptop users who unless it's a monster of a laptop have to choose between a SSD or a HDD.
It's a difficult sort of study to do properly for a few reasons so unless there is strong evidence otherwise (e.g. funding from big media) I'd expect this was simply a case of incompetance.
Reasons why it's a difficult sort of study
1: If you actually download the files to investigate them then you are getting into legally dodgy ground. If you want to download at more than a trickle you will have to upload too which puts you in an even worse position legally.
2: Afaict most legal torrents use their own trackers rather than public ones, so they won't show up in a search that focuses on the big puplic trackers.
3: Afaict big media are trying to deliberately fill P2P systems including bittorrent with fake crap as part of their war on the pirates.
Because as well as capacity there is performance and reliability to consider.
High spin speeds are important to hard drive performance because a large part of the time to perform a small read is the time taken to wait for the HDD to spin to the right position after seeking to the track. Unfortunately high speeds and large platters are not a good combination mechanically.
For SSDs it could be usefull at some point but afaict the main limitation on SSDs right now is that most people aren't prepared to pay what even 1TB of flash costs (cost per gigabyte of SSDs seems relatively constant over a wide range of capacities).