True, but Fielding is the wanker who calls out google and then asks to do the same, the privacy commissioner isn't involved in the retention stuff(at least at present).
That's true, but if Rudd needs the greens to pass anything else he wants(and the greens can actually deliver it unlike now), he's too much of an operator to piss them off over something like this.
Personally I'm hoping for a situation where labor can only pass legislation with the help of the greens. That should tone down the crazy of the greens, and tone down the nanny of labour.
Of course it would be even better if we could get that combination plus a liberal party who had some policy other than "oppose everything" so that some debates went right and some debates went left depending on the interests of the country.
His name is Stephen Conroy, and he's not a privacy commissioner, he's the IT minister(for lack of a better phrase). Whether he is individually a loon, or whether he's the front man for Rudd's lunatic policies, or whether the government is just making him froth about lunatic policies to please the few right wing loons in the senate they need to get anything done I really don't know.
That said he's still better than the guy we had a while back who said Australia didn't need faster internet because all it would be used for would be to download porn. Australia is simultaneously the most forward thinking and the most backward place I've ever lived.
Here in Oz we have a choice between the current party who have a particular bent towards nanny stating but otherwise aren't too bad, the Liberal party who are no longer liberal and seem to support the idea of moving back to the 1950's, xenophobia, conservative religious values, privatizing things even the US hasn't privatized, and bending over backwards for big business(and is also a direct continuation of the bugger we voted out last time), and the Greens, who are one of those parties who have a lot of really good ideas, but who are also raving lunatics.
So we have the choice of giving up our freedom, giving up our freedom, or giving up everything else in exchange for our freedom. It's not a whole lot different than the upcoming US election except that our lunatic fringe party is on the left whereas your lunatic fringe party is on the right.
You're right, XP did come out in 2001, which is not quite a decade(though close enough). The bigger problem isn't so much the age of release, but the age of the design.
Whether people believe it or not, Microsoft seems to be starting to get a clue and has come an awful long way from where they were in the late 90's when XP was being designed. A lot of the really stupid things they did which made XP as insecure as it was, aren't part of the design of the Vista/7 line of Operating Systems.
I didn't say it was irrelevant or that Microsoft shouldn't fix it. I said that fixing XP should be a lower priority than fixing an equivalent bug in a more modern version of Windows.
XP is an incredibly old OS, and its design is fundamentally flawed when it comes to security. It has never been and will never be secure. I know Microsoft still sells it(though that'll be done in the next couple months), but that doesn't mean that anyone sane should be buying it. Vista was a bit of a dog, but Windows 7 isn't.
As for a week, a company as large as Microsoft takes a week to do anything whatsoever, and isn't going to commit to any kind of deadline until it's had a chance to look at it. They didn't say they wouldn't fix it, they didn't even say they wouldn't fix it within 60 days, they said they wouldn't commit to 60 days until they'd had a chance to investigate it. No software company(or any other company) with support commitments would do it any differently.
5 days makes him a dickhead, I don't care who he did it to, or what they've done in the past. No one fixes a bug in 5 days, and people do use windows help, some of them even use the remote assist feature.
The help system in Windows XP is for all intents and purposes IE, remember that XP is old enough that it was made during the period where Microsoft were obsessed with making everything part of the core OS.
A bug for an OS which is two versions behind current and almost a decade old, should not be higher priority than fixing current versions of the software. 5 days is also far too short a time for a company the size of Microsoft to even get a team together to look at the problem, let alone come up with an adequate solution, properly test that solution, distribute that solution and get that solution tested and deployed by customers.
This guy was a dickhead and if he'd done it to anyone other than Microsoft he'd have been burned at the stake, ffs 5 days?
I know that from a purely technical point of view O(c) is O(1), but when c is 1000,000 times slower than O(1) then you're making an ass of yourself for anything other than a really gigantic data set.
That's really the big problem with using O() notation in the first place. Because it's entirely concerned with scaling, all constants get filtered out on the presumption that for a sufficiently large data set the constants don't matter. The problem is that when you constants get particularly large, you need a fairly large data set before that's actually true.
It's been a while since I did an algorithm class, but my recollection is that O() notation is indeed about the scaling of operations with regards to the input size(n), and generally speaking constants are indeed usually ignored.
The issue however is that all operations are not equal. A moderate speed modern desktop CPU can perform something on the order of 6 and a half billion internal operations in a second, whereas even under the assumptions of the article using a 1 ms access time for hard disk, you can only get 1000 disk operations in a second, a difference of six orders of magnitude.
Knuth's algorithms presume a data access cost of O(1), whereas even on a desktop you're really looking at O(1000000) for some operations and O(1) for others. You still want to minimize operations of course, but if you're accessing disk most of the time, you'd have to have an input size of a million just to break even between the O(n) and O(n^2) algorithms even on a basic desktop.
I'm aware that this is an over simplification since cache and RAM are nowhere near as slow as disk, but the point is that if you're accessing disk a lot then you need a very large n to see a substantial speed increase.
I'd like to live in an advertisement free society too, however I also acknowledge that everyone has to eat, and unless I want to pay cash up front for everything I get, I have to accept a certain amount of advertisement.
That said, there are limits, I don't mind image ads in my web pages, I don't mind text ads, I don't even mind flash ads or basic javascript ads, so long as what they're doing doesn't interfere with my ability to use the page. If I've got to wait 5 seconds every time I open the page while they layer stuff on top of the content I'm trying to read, then I'm going to block it.
The stuff you learn in economics 101 (or for that matter any 101 type course), is never a complete picture of what's going on.
Supply and demand as taught in those kinds of courses is the same as the kinds of physics you get taught where everything is functionally behaving in a vacuum, except for some reason people understand that a ball won't behave exactly the way it does in the formulas when you throw it but they presume that the market will.
First of all, I've never met anyone who just throws out a perfectly good car, most people don't even throw out clunkers, they trade them in, so generally speaking the perfectly good car goes on to become someone else's car so the cost of creation for that car is largely pointless.
Secondly, while going from 25 mpg to 30 mpg isn't huge, it's still a 20% increase in the amount of miles you can drive on a given amount of fuel. Presuming you'd put about 200,000 miles(probably a conservative estimate) on a 30 year old car(and it's going to take a hell of a lot more than "learning how to use a wrench" to get a car to 30 years of age), then you'd be looking at about 1/6th of those miles being essentially free in comparison to the 25 mpg car. That's about 33,333 extra miles you got out of the same amount of fuel, which at 25 mpg would have consumed 1333 gallons of fuel.
That's not exactly small amounts of fuel by any stretch of the imagination, even discounting for loss of fuel efficiency over the years, or the cost of the fuel required to make the parts you'd need to keep a modern car going for that long, or any of the energy saved by recycling the steel or parts in your used car if it isn't drivable.
Now there are certainly arguments as to whether hybrid cars like the Prius are actually a particularly environmental thing, given that making those batteries isn't comparable to building a standard car and they generally get shipped further, and there's certainly a break even point environmentally and economically speaking for disposing of a car entirely, but an increase in mileage is an increase.
Don't get me wrong, if you're looking at replacing a perfectly good car with a new car, knowing how much better that car is is compare to your old car is a very useful thing.
That said, if you need a new car anyway, it really doesn't matter that the difference between a 10 mpg car and a 20 mpg car is higher than the difference between a 50 mpg car and a 33 mpg car. I 50 mpg car is still the best choice. No you probably shouldn't throw out your 6 month old 33 mpg car to get a 50 mpg car, and no you shouldn't say "I can't afford the 50 mpg car so I'm going to stick with the 10 even though I can afford the 20", but while mpg doesn't scale linearly, 50 is still better than 33.
If you were talking about some obscure tax law, or privacy law, or election regulation, you'd probably be right. However I think you underestimate how much people hate cold calling and robodialers in particular.
The news media will be perfectly happy to call the judge on this since they know it'll get ratings.
Corporates can still buy XP till SP1 comes out for Windows 7 and I think that EOL for XP is 2014 or somewhere thereabouts.
That said, no one should be using XP. Move to Linux, move to Mac, or upgrade. XP is a steaming pile after all these years and the idiots with the rose coloured glasses obsessing about how great it is and how it works on the computer they bought in the 1990's can go take a running leap.
The rules have to change, that's fine, but they shouldn't change arbitrarily. The success of the iPhone lies in large part with the applications available for it, and those applications are there in large part because of the environment Apple offers to developers.
This means that the agreement which Apple has with iPhone developers is important to their business model. Apple needs to have a written application policy, changes to that policy should be in writing and developers should be notified in advance of any changes.
My point was more that it's not a mineral resource.
If it was a mineral resource we could just recycle what we already have. There might only be a certain amount of iron in the world, but it doesn't go away when we make something out of it, it's still there, and we can reuse it if we need to, there's certainly a maximum amount of it, but we'll never "run out".
Oil on the other hand is an organic compound, with all the baggage which comes along with that. There is no maximum amount, but you can't reuse it either, once you've burnt it it's not oil anymore and you have to wait for more to get produced. Yes because the renewal rate of oil is really slow we can't control its rate of renewal very well, but treating it like it's a mineral resource like iron or gold isn't correct and will lead to faulting thinking.
Oil is not a mineral resource, it is organic not mineral, there is not a finite supply, and it is renewable. A sustainable oil industry is theoretically possible, though of course largely impractical.
Theoretically new oil is being created all the time and will continue to be created for the rest of eternity. The rub of course is that we've used up the majority of the oil created in the last billion years or so in the last century, so our rate of use is quite a bit faster than the rate of resupply.
I think the issue is that Apple keeps changing their minds. It's one thing to build some software send it to Apple and have them reject it. That's part of life, but building a product, selling it to customers, investing in that product and then having Apple change their minds on a previously granted approval is a bit of a hard pill to swallow, and it's not just been this guy who have had that happen.
Slashdot Mirror
True, but Fielding is the wanker who calls out google and then asks to do the same, the privacy commissioner isn't involved in the retention stuff(at least at present).
That's true, but if Rudd needs the greens to pass anything else he wants(and the greens can actually deliver it unlike now), he's too much of an operator to piss them off over something like this.
Personally I'm hoping for a situation where labor can only pass legislation with the help of the greens. That should tone down the crazy of the greens, and tone down the nanny of labour.
Of course it would be even better if we could get that combination plus a liberal party who had some policy other than "oppose everything" so that some debates went right and some debates went left depending on the interests of the country.
His name is Stephen Conroy, and he's not a privacy commissioner, he's the IT minister(for lack of a better phrase). Whether he is individually a loon, or whether he's the front man for Rudd's lunatic policies, or whether the government is just making him froth about lunatic policies to please the few right wing loons in the senate they need to get anything done I really don't know.
That said he's still better than the guy we had a while back who said Australia didn't need faster internet because all it would be used for would be to download porn. Australia is simultaneously the most forward thinking and the most backward place I've ever lived.
Well, unfortunately it's not that simple.
Here in Oz we have a choice between the current party who have a particular bent towards nanny stating but otherwise aren't too bad, the Liberal party who are no longer liberal and seem to support the idea of moving back to the 1950's, xenophobia, conservative religious values, privatizing things even the US hasn't privatized, and bending over backwards for big business(and is also a direct continuation of the bugger we voted out last time), and the Greens, who are one of those parties who have a lot of really good ideas, but who are also raving lunatics.
So we have the choice of giving up our freedom, giving up our freedom, or giving up everything else in exchange for our freedom. It's not a whole lot different than the upcoming US election except that our lunatic fringe party is on the left whereas your lunatic fringe party is on the right.
You're right, XP did come out in 2001, which is not quite a decade(though close enough). The bigger problem isn't so much the age of release, but the age of the design.
Whether people believe it or not, Microsoft seems to be starting to get a clue and has come an awful long way from where they were in the late 90's when XP was being designed. A lot of the really stupid things they did which made XP as insecure as it was, aren't part of the design of the Vista/7 line of Operating Systems.
I didn't say it was irrelevant or that Microsoft shouldn't fix it. I said that fixing XP should be a lower priority than fixing an equivalent bug in a more modern version of Windows.
XP is an incredibly old OS, and its design is fundamentally flawed when it comes to security. It has never been and will never be secure. I know Microsoft still sells it(though that'll be done in the next couple months), but that doesn't mean that anyone sane should be buying it. Vista was a bit of a dog, but Windows 7 isn't.
As for a week, a company as large as Microsoft takes a week to do anything whatsoever, and isn't going to commit to any kind of deadline until it's had a chance to look at it. They didn't say they wouldn't fix it, they didn't even say they wouldn't fix it within 60 days, they said they wouldn't commit to 60 days until they'd had a chance to investigate it. No software company(or any other company) with support commitments would do it any differently.
5 days makes him a dickhead, I don't care who he did it to, or what they've done in the past. No one fixes a bug in 5 days, and people do use windows help, some of them even use the remote assist feature.
The help system in Windows XP is for all intents and purposes IE, remember that XP is old enough that it was made during the period where Microsoft were obsessed with making everything part of the core OS.
Because some people(admitedly not very many) actually use HCP.
A bug for an OS which is two versions behind current and almost a decade old, should not be higher priority than fixing current versions of the software. 5 days is also far too short a time for a company the size of Microsoft to even get a team together to look at the problem, let alone come up with an adequate solution, properly test that solution, distribute that solution and get that solution tested and deployed by customers.
This guy was a dickhead and if he'd done it to anyone other than Microsoft he'd have been burned at the stake, ffs 5 days?
I know that from a purely technical point of view O(c) is O(1), but when c is 1000,000 times slower than O(1) then you're making an ass of yourself for anything other than a really gigantic data set.
That's really the big problem with using O() notation in the first place. Because it's entirely concerned with scaling, all constants get filtered out on the presumption that for a sufficiently large data set the constants don't matter. The problem is that when you constants get particularly large, you need a fairly large data set before that's actually true.
It's a bit more complicated than that.
It's been a while since I did an algorithm class, but my recollection is that O() notation is indeed about the scaling of operations with regards to the input size(n), and generally speaking constants are indeed usually ignored.
The issue however is that all operations are not equal. A moderate speed modern desktop CPU can perform something on the order of 6 and a half billion internal operations in a second, whereas even under the assumptions of the article using a 1 ms access time for hard disk, you can only get 1000 disk operations in a second, a difference of six orders of magnitude.
Knuth's algorithms presume a data access cost of O(1), whereas even on a desktop you're really looking at O(1000000) for some operations and O(1) for others. You still want to minimize operations of course, but if you're accessing disk most of the time, you'd have to have an input size of a million just to break even between the O(n) and O(n^2) algorithms even on a basic desktop.
I'm aware that this is an over simplification since cache and RAM are nowhere near as slow as disk, but the point is that if you're accessing disk a lot then you need a very large n to see a substantial speed increase.
I'd like to live in an advertisement free society too, however I also acknowledge that everyone has to eat, and unless I want to pay cash up front for everything I get, I have to accept a certain amount of advertisement.
I don't use adblock for this very reason.
That said, there are limits, I don't mind image ads in my web pages, I don't mind text ads, I don't even mind flash ads or basic javascript ads, so long as what they're doing doesn't interfere with my ability to use the page. If I've got to wait 5 seconds every time I open the page while they layer stuff on top of the content I'm trying to read, then I'm going to block it.
True enough.
I just get frustrated by the number of people who believe they understand everything because they understand supply and demand.
The stuff you learn in economics 101 (or for that matter any 101 type course), is never a complete picture of what's going on.
Supply and demand as taught in those kinds of courses is the same as the kinds of physics you get taught where everything is functionally behaving in a vacuum, except for some reason people understand that a ball won't behave exactly the way it does in the formulas when you throw it but they presume that the market will.
BP actually have a really shitty environmental record.
What they have is a really good marketing department.
That's not entirely true.
First of all, I've never met anyone who just throws out a perfectly good car, most people don't even throw out clunkers, they trade them in, so generally speaking the perfectly good car goes on to become someone else's car so the cost of creation for that car is largely pointless.
Secondly, while going from 25 mpg to 30 mpg isn't huge, it's still a 20% increase in the amount of miles you can drive on a given amount of fuel. Presuming you'd put about 200,000 miles(probably a conservative estimate) on a 30 year old car(and it's going to take a hell of a lot more than "learning how to use a wrench" to get a car to 30 years of age), then you'd be looking at about 1/6th of those miles being essentially free in comparison to the 25 mpg car. That's about 33,333 extra miles you got out of the same amount of fuel, which at 25 mpg would have consumed 1333 gallons of fuel.
That's not exactly small amounts of fuel by any stretch of the imagination, even discounting for loss of fuel efficiency over the years, or the cost of the fuel required to make the parts you'd need to keep a modern car going for that long, or any of the energy saved by recycling the steel or parts in your used car if it isn't drivable.
Now there are certainly arguments as to whether hybrid cars like the Prius are actually a particularly environmental thing, given that making those batteries isn't comparable to building a standard car and they generally get shipped further, and there's certainly a break even point environmentally and economically speaking for disposing of a car entirely, but an increase in mileage is an increase.
Don't get me wrong, if you're looking at replacing a perfectly good car with a new car, knowing how much better that car is is compare to your old car is a very useful thing.
That said, if you need a new car anyway, it really doesn't matter that the difference between a 10 mpg car and a 20 mpg car is higher than the difference between a 50 mpg car and a 33 mpg car. I 50 mpg car is still the best choice. No you probably shouldn't throw out your 6 month old 33 mpg car to get a 50 mpg car, and no you shouldn't say "I can't afford the 50 mpg car so I'm going to stick with the 10 even though I can afford the 20", but while mpg doesn't scale linearly, 50 is still better than 33.
If you were talking about some obscure tax law, or privacy law, or election regulation, you'd probably be right. However I think you underestimate how much people hate cold calling and robodialers in particular.
The news media will be perfectly happy to call the judge on this since they know it'll get ratings.
Corporates can still buy XP till SP1 comes out for Windows 7 and I think that EOL for XP is 2014 or somewhere thereabouts.
That said, no one should be using XP. Move to Linux, move to Mac, or upgrade. XP is a steaming pile after all these years and the idiots with the rose coloured glasses obsessing about how great it is and how it works on the computer they bought in the 1990's can go take a running leap.
The rules have to change, that's fine, but they shouldn't change arbitrarily. The success of the iPhone lies in large part with the applications available for it, and those applications are there in large part because of the environment Apple offers to developers.
This means that the agreement which Apple has with iPhone developers is important to their business model. Apple needs to have a written application policy, changes to that policy should be in writing and developers should be notified in advance of any changes.
My point was more that it's not a mineral resource.
If it was a mineral resource we could just recycle what we already have. There might only be a certain amount of iron in the world, but it doesn't go away when we make something out of it, it's still there, and we can reuse it if we need to, there's certainly a maximum amount of it, but we'll never "run out".
Oil on the other hand is an organic compound, with all the baggage which comes along with that. There is no maximum amount, but you can't reuse it either, once you've burnt it it's not oil anymore and you have to wait for more to get produced. Yes because the renewal rate of oil is really slow we can't control its rate of renewal very well, but treating it like it's a mineral resource like iron or gold isn't correct and will lead to faulting thinking.
Oil is not a mineral resource, it is organic not mineral, there is not a finite supply, and it is renewable. A sustainable oil industry is theoretically possible, though of course largely impractical.
Theoretically new oil is being created all the time and will continue to be created for the rest of eternity. The rub of course is that we've used up the majority of the oil created in the last billion years or so in the last century, so our rate of use is quite a bit faster than the rate of resupply.
Sort of.
I think the issue is that Apple keeps changing their minds. It's one thing to build some software send it to Apple and have them reject it. That's part of life, but building a product, selling it to customers, investing in that product and then having Apple change their minds on a previously granted approval is a bit of a hard pill to swallow, and it's not just been this guy who have had that happen.