It seems to me that there's an amusing lag in regulatory / legal processes. Seems like if you're an MP or a judge, the dangerous, illegal, subversive stuff if "Stuff that's happened since I was 21". The established Right Way that we all acknowledge as Truth is "How things were before I was 21". In a decade or two we're perhaps going to see MPs who have a more realistic attitude to internet censorship, copyright, etc (modulo lobbying by business, of course!) and the terrifying new technology that's corrupting our children will be augmented reality, neural uplinks, or fully functional Terminators or something - whatever the alarmingly young and unruly kids are currently playing with.
As others have noted, it's apparently not possible to assign 18+ (or equivalent) ratings to games in Australia, so it's not necessarily like the classification people are censoring it - you make a game that "should" have an 18+ rating, they're not able to give it that, they have no choice but to refuse classification. Who is responsible for this rule? It's they who is/are responsible for the video games being banned.
Craptastic as the Intel cards may be, in overall performance terms, I could happily take any of the integrated parts by Intel that has decent Linux support on my next desktop, even if that meant a massive reduction in performance. I have an Xbox 360 for playing games on and I would love for my desktop to Just Work as well as my Eee does with Linux. That said, with ATI cards getting better and better support under Linux it is quite possible that they'll be the best option by the time I upgrade again...
My main like is the fact that there's a common API (with multiple language bindings) for programming core desktop components as well as widgets. Some aspects of their implementation really annoy me. *cough* cashew *cough*;-)
I think scripted widgets are effectively sandboxed, which should be easy because errors can be caught. The random widgets you download from the internet are generally in this category. If their language interpreter crashes you're probably still in trouble but hopefully that's less likely... Native widgets aren't sandboxed which does increase the probability of desktop explosions. I've certainly had a few of those, although part of that may be due to my dodgy system.
Personally I only have about the same number of native widgets running that I used to have as taskbar applets in KDE 3.x, which I'd imagine were also sharing a process with the Kicker at least. So it's not *necessarily* much worse, though that was already a bit more monolithic than a really lightweight window manager would be. However, there's probably now (many) more lines of code to do the same work KDE 3 did, plus the desktop that displays on the root window lives in the same process as the other native plasma code:-/ So that means more bug exposure, probably.
I've known the topic of sandboxing the native widgets come up and the response was "one day", probably in the "someone could do this but the core developers feel they have more important stuff to do" sense.
I found earlier releases of KDE 4 slowed things down particularly, I think it might have improved since then. It's not as snappy as I'd like but it's actually fairly usable now with my KDE 4.3 on Kubuntu Jaunty. I expect it would improve if I installed the proprietary NVidia drivers, or even just the Nouveau drivers.
Interestingly, I find that KDE 4.2 runs quite nicely on my Eee 701 with 2GB RAM as long as I don't run loads of apps - and that's with KWin compositing enabled and a few fairly heavyweight things running. The expose and desktop switcher are quite useful on the tiny screen.
I've seen various claims that KDE4 is much happier if there's compositing available but I'm not sure how accurate they are.
Plasma isn't just that thing for making desktop widgets of dubious usefulness. What KDE has actually done is, in my opinion, a fairly smart design move regardless of whether you like their implementation.
Desktop widgets aren't applications, they are people extending the functionality of their desktop. What the KDE folks saw was that a well-designed API could be used to write the desktop UI itself (task bar, clock, pager, whatever), the things we used to use taskbar applets for (media player control, etc) and the flashy new desktop widgets. Instead of having a basic desktop and plastering a widget API on top, they've gone and unified the whole thing so you can use the same API to write taskbar applets, widgets or write replacement taskbars or... whatever. The various desktop elements are separate building blocks (plasmoids) that can be assembled together. They've also produced loads of bindings for this API to give folks the chance to write stuff in their favourite language.
The plasma netbook interface then takes some of the default building plasmoids, adds some new ones and then glues them together in a different way. So you can get a similar family look and similar functionality (and, fundamentally, the same desktop) but in a way that's optimised for a different form factor of device. I think that's actually pretty neat and somewhat reminiscent of the way you can configure and compile the core Linux kernel down for tiny machines or up to big iron whilst still getting the benefits of a common codebase.
There's a load of other cool stuff including a standard set of "data engines" which separate producing data from displaying it, thus making it easy to glue data sources together in interesting ways. Despite the various feature regressions that rewriting the desktop led to, it's a really neat architecture and should hopefully stand them in good stead for the future.
That's cool, I wasn't aware of that. Seems like it might be useful for scientific code. I know Fortran includes the ability to declare pure functions and (I think) verify their purity, which sounds useful.
Though I imagine the problem for larger computations could be that you're limited to writing functional-style code, at which point you might be better off using a language designed for it? What optimizations does gcc actually do on pure functions, having identified them? I think it would be very interesting to tie this with something like Apple's Grand Central Dispatch (in an automated way, of course) to get computations parallelised magically to whatever degree the hardware can support.
Because there are tons of vested interests. SAP is based in the EU, so there's the possibility they're lobbying the EC on this one. One assumes that Oracle / Sun are lobbying US senators (and politicians in the EU for that matter?). The EU, as the article points out, works under different rules and with a different viewpoint - Oracle and Sun agreed to be bound by local laws when they entered the European markets. The EU probably has a political interest in seeming to stand up to the US, though you'd hope the regulators wouldn't be swayed into unprofessional behaviour by that. The US has an interest in avoiding a precedent where the EU has power over one of their companies. Sun and Oracle are probably trying to dodge awkward questions and hope for the EU to cave. Really, there's no reason to believe 100% that anyone is acting entirely in good faith here, especially given we don't have access to all the information.
We're seeing an interesting consequence of the increasingly interconnected world, though, in that we're reaping business advantages from setting up shop in multiple large markets but in turn companies are then subject to multiple jurisdictions regardless of their country of origin. It seems like the EU and US regulators working together on a decision might be more appropriate, given neither of them has absolute authority to give the go ahead. A co-operative solution to regulation decisions would make a certain amount of sense since it's de facto what we have now. It's surely in nobody's interests for the decision to be left hanging.
Heh, to be honest I'm mostly in the same boat as you - how can Haskell do side-effect stuff whilst still being purely functional. And what on earth is a Monad!?
A programmer friend of mine used to say that using monads to emulate state would be a bit like the following: imagine you want to emulate state but you're programming functionally. You want variables but you can't have them. Instead you have a big tuple of values representing the contents of all the "variables" and you pass this tuple to every function call and get back a new version with the variables updated. You're still programming purely functionally but you've got this emulation of mutable state.
But I mentioned this to another friend who has a PhD in compsci, specialising in compiler design and he looked at me like I was insane - so apparently that's not a good representation of how they behave. Still, it at least gets you a feel for how you could have imperative constructs like variables without breaking the rules of pure functional programming. Maybe that makes the idea of monads seem a little less impossible, even if it doesn't help explain how they work?
Other "functional" languages actually have an imperative subset (e.g. ML has variables, loop constructs, etc). Haskell is distinguished from those partly by being purely functional, so I presume that it really *is* purely functional by some definition, including the monads. But I can't say I understand how.
I had to study the type system of ML as an undergrad and it had this property of looking really opaque to me until I'd stared at it for hours and hours, then suddenly I went "Ooooh, I see!". Monads have a similar feel to me but I've never taken the time to stare them down. That said I imagine that, like the ML type system, they're much easier to use than to fully understand.
Well, pure functional languages are (potentially) good for concurrency in general. Because they have no mutable variables in the usual sense, it doesn't actually matter what order functions are evaluated in (other than the fact that callers cannot continue until their callees return). You can't do this in C or Java because it might be necessary for one function to see a variable modified by another. In a functional language, any dependencies are explicit call-return relationships (well, ish, they typically do have some non-functional constructs otherwise it's hard to do IO!), so in principle it's quite easy to split up a program's work across multiple CPUs (or machines) and not worry about whether they need to talk to each other.
Haskell, along with the ML family of languages, also has an amazing type checker that is waaay more sophisticated than most other languages. I think most people who've played around with these languages do start to feel that often "If it compiles, it's bug-free". Obviously that's not something you can rely on, since the compiler can't know what you meant to do. But it is true that the type system is *way* more useful at detecting bugs at compile-time than for any conventional language I've used.
It's bizarre but there still seems to be this perception that the police are a fine bunch of chaps who will universally do their best to apply the rules sensibly and fairly. There are plenty of police officers who that description applies to, I'm sure - but that's not an excuse for lawmakers and the justice system to assume it holds universally true.
At the end of the day, the police are there - in practice - *to catch potential criminals*. Sorting out who is and isn't guilty is not their job, that's the job of the courts (as it should be). So the police don't really have an incentive to be especially fair or reasonable; that's not what we've tasked them with doing. What lawmakers sometimes seem to fail to understand is that if we pressure them to achieve "catch all the terrorists / criminals" then they'll try to do that, even if they "catch" many innocent people too. If we give them new tools to do that then *they will use them*. If the tools we give them are extremely blunt instruments, like the ability to hold innocent people's data on the DNA database, they're going to use them to their fullest extent. If we want them to behave sensibly, the laws need to be more focused and less open to abuse.
It's the same issue with various "anti-terror" laws. Allegedly local councils in the UK have used these to put people under surveillance for reasons unrelated to terrorism (like whether they're using their rubbish bins correctly and whether they live in the locality of a school they have applied to). We gave them overly broad legislation and assumed that they wouldn't use it, even though it helps them to do what they see as their job. None of these organisations can be relied upon to act in the best interests in society because each of them only sees part of the big picture - our politicians are *supposed* to maintain the balance of power with targeted legislation that results in society's best interests being served overall. That goal can't be reached by handing out disproportionate powers indiscriminately.
>>>when I pay taxes I don't consider a state healthcare system to be a theft of my labour.
Even if you never get sick? i.e. You get hit by a car and die, or have a massive heart attack at age 80 and fall over dead. Then you've paid ~$500,000 in Euros towards the hospitals and got (virtually) nothing in return. That's almost as bad as the thousands we Americans were forced to give to AIG - welfare for the corporations.
Well, if I pay into a private health insurance plan and I never get sick, my money gets used to subsidise the treatment of those who did, right? And the other people in the private scheme will only be people who were wealthy enough to afford private insurance, so I'm subsidising relatively wealthy people.
I've still paid money for nothing in return and it's still been used to treat other people. Or do you just mean I should have the right to not pay into *any* health plan and just hope not to get sick?
In that case, I would see your point but I still wouldn't like to live in a country with that system, since I think it would result in a lot of down-and-out people who would be a burden on the economy in other ways (my life is better overall if the reasons for crime, bankruptcies, etc are minimised). It's like paying taxes for the police force - it may never help me directly but if it helps society function well then that benefits everyone. Yes, I actually think it's worth my subsidising other people's healthcare up to a point. Moreover, if the system is run right there are economies of scale to be had so that the cost is not prohibitive.
Personally, if there wasn't state healthcare here I'd be paying into a private insurance plan *anyhow*, since I can't guarantee I'll never be sick or injured. If I pay into a private insurance plan and don't get sick then only richer people benefit from my payments. If I pay into a state healthcare system then all of society benefits and I'll likely get indirect value out of that even if I'm never ill myself.
Consider trying OpenStreetMap for some of your navigation. http://www.openstreetmap.org/ It's a Wikipedia-like effort to create an Open Source-ish map of the world. The project was started in the UK and although coverage is still patchy in some areas it does already have excellent coverage of large parts of London. It includes some data that Google Maps does not have. Due to the Free as in Freedom nature of the project, the website also makes it easy to export PDFs, pngs, etc of areas of the map which you can then store for reference offline (though obviously you can't then search them).
There's a navigation service at http://yournavigation.org/ which has some neat features, like the ability to plot a route for pedestrians or cyclists that will take you down cycleways and footpaths where possible (there was also an elevation profile of your route, so you'll know how hilly it is! That was a GSoC project, I think but it seems not to be available for now).
Disclaimer: I am plugging a project I've contributed to here but it really is pretty good.
PS. There are internet kiosks but you'll need to ask your butler to bring some coal to power them.;-)
I can't speak for other Europeans but: when I pay taxes I don't consider a state healthcare system to be a theft of my labour. Leaving aside moral issues, I think it is maybe preferable to take the up-front financial cost of funding a state healthcare system than to take the social and financial penalties of creating a social and financial underclass who still have "costs" attached for society, even if they're not being paid welfare directly.
Secondly I don't mind because I'm investing money in a system I may need to use - and that system only works because a large enough body of people are paying into it. Insurance doesn't (just) work on the idea that you pay in enough to make your account profitable over time, it attempts to average out the relatively few unprofitable but nonetheless covered individuals across all of the clients with policies. So if you pay into an insurer, it's entirely possible that more sickly people will effectively benefit from your money. It's worth paying in because it gives you the ability to have treatment paid (effectively) out of the other client's fees, should you need to.
I was more speculating on a possible convergence - Android going the Chrome way, you might say. Although it'd be a bit surprising if they dropped compatibility completely, I guess. Good to see that the native support for Android has improved, although if they're allowing native code I'm even more perplexed as to why they went the Java route in the first place.
True. The short answer is that I do personally care, so it does affect my purchase decisions. The longer, wider ranging answer is that I don't think making a platform more limited for programmers is necessarily going to result in a better range of apps for the user, so it's not a strategy I want to see spread.
I wonder if this means Android will converge towards a more standard Linux, or if Chrome will converge to become less standard. Or if they'll keep the unique aspects of each and just try to unify stuff like browser code. I don't really fancy a phone that can only run web apps, or a "PC" that can only run Java apps compiled to a weird byte code! I don't really like the way Android has reinvented all of userspace, whereas at least Chrome builds on existing code a bit more. But they are solving different problems, which perhaps explains *some* of the differences...
A man recovering from spinal injuries goes to his hundredth physio session in just a few months. With the physio's help and all his strength he hauls himself upright and takes a step - his wife takes a photograph - before falling painfully to the ground. The insurer looks at the photograph "Congratulations, you're cured because you're standing and you couldn't do that before.".
The medical profession seems to generally agree that depression is a real condition, there's research to demonstrate that there's a real underlying chemical difference in the brains of sufferers. Are you saying that they're all wrong? If you are asserting that, it would be most helpful for you to tell us where you received your medical degree and provide a link to your list of peer-reviewed research papers in the area. If you don't have those then your claim is similar to the people who don't know anything of physics but are utterly convinced that the LHC will destroy the world and that they know better than the scientists who built it. If you do have those then you are, presumably, revolutionizing the state of medical science with your argument?
I'm not really clear whether you're saying depression doesn't exist or just that most people who claim benefits due to depression are lying. I think I've addressed the former. But if you meant the latter - is it really fair behaviour to accuse every single depression claimant of being a fraud because you heard that some people have been frauds? Do you apply the same logic (many X are Y, therefore I assume that all X are Y) in other areas of life?
You make it sound like the insurance company is forced to do this due to business realities, which I don't think is really fair. There are a lot of circumstances in which a business might like to get out of fulfilling the obligation to do what you paid them for - that's a reason for them to weasel out of stuff but it's not an excuse. If the woman has got a diagnosis saying she has depression and the insurance company sees evidence that she's improved then they could ask her to come back for a medical assessment.
To continue using your analogy of someone who supposedly couldn't move but is photographed living it up, unless they put some more thorough effort (than just a few Facebook photos) into examining her case it could equally be like saying "Oh, we saw them standing up even though they were supposed to be crippled" having just found a picture of a physio session where 30 seconds after the photo was taken the person fell down because they weren't strong enough to stand.
It's right (and understandable) for insurers to be careful in what they pay out for and to verify that people's needs are genuine. I just think that due diligence in this case involves a bit more than looking at a couple of photos without context. The cover is a service that you *pay* for after all, so if they're going to bail out on giving you that service they'd better have a solid explanation.
I own an air guitar and it's actually pretty sweet - I can make like I'm rocking out wherever I am. Whereas an air car... I don't see the market. You're at a party and the music's pumping and you just decide to "air drive" to the shops? Not cool.
Some comedian - I think it was Ed Byrne - on the UK satirical news quiz "Have I Got News For You" made a good point about nuclear submarines. Probably misquoting slightly it was:
"We need to spend £20bn on a giant metal sausage under the sea in case one day we decide to destroy the world"
I think that put the whole scheme into perspective a bit.
When Palm tried to use Apple's software in a way Apple didn't like (syncing the Pre with iTunes), it got broken pretty fast - more than once. It's a different situation - Pre is a competitor to the iPhone and Apple aren't obliged to support a competitor. But Apple was going out and changing iTunes to make it not sync with the Pre, it's not like changes Apple made in normal development just *happened* to break the Pre. Whereas nobody would have batted an eyelid if Apple had accidentally broken the Pre, a lot of people felt that deliberately breaking it was not something they were keen on.
Seeing how aggressively they've pursued that case, you can see how they might get accused of doing a similar thing for the iPhone. I can just about imagine that Apple would have the arrogance to take a hard line on iPhone jailbreaking but I'd be mightily surprised if they had the legal and PR foolhardiness to actually deliberately brick devices.
Slashdot Mirror
It seems to me that there's an amusing lag in regulatory / legal processes. Seems like if you're an MP or a judge, the dangerous, illegal, subversive stuff if "Stuff that's happened since I was 21". The established Right Way that we all acknowledge as Truth is "How things were before I was 21". In a decade or two we're perhaps going to see MPs who have a more realistic attitude to internet censorship, copyright, etc (modulo lobbying by business, of course!) and the terrifying new technology that's corrupting our children will be augmented reality, neural uplinks, or fully functional Terminators or something - whatever the alarmingly young and unruly kids are currently playing with.
As others have noted, it's apparently not possible to assign 18+ (or equivalent) ratings to games in Australia, so it's not necessarily like the classification people are censoring it - you make a game that "should" have an 18+ rating, they're not able to give it that, they have no choice but to refuse classification. Who is responsible for this rule? It's they who is/are responsible for the video games being banned.
Craptastic as the Intel cards may be, in overall performance terms, I could happily take any of the integrated parts by Intel that has decent Linux support on my next desktop, even if that meant a massive reduction in performance. I have an Xbox 360 for playing games on and I would love for my desktop to Just Work as well as my Eee does with Linux. That said, with ATI cards getting better and better support under Linux it is quite possible that they'll be the best option by the time I upgrade again...
Nurse, get another ream of US Letter in Tray 2, stat!
My main like is the fact that there's a common API (with multiple language bindings) for programming core desktop components as well as widgets. Some aspects of their implementation really annoy me. *cough* cashew *cough* ;-)
I think scripted widgets are effectively sandboxed, which should be easy because errors can be caught. The random widgets you download from the internet are generally in this category. If their language interpreter crashes you're probably still in trouble but hopefully that's less likely... Native widgets aren't sandboxed which does increase the probability of desktop explosions. I've certainly had a few of those, although part of that may be due to my dodgy system.
Personally I only have about the same number of native widgets running that I used to have as taskbar applets in KDE 3.x, which I'd imagine were also sharing a process with the Kicker at least. So it's not *necessarily* much worse, though that was already a bit more monolithic than a really lightweight window manager would be. However, there's probably now (many) more lines of code to do the same work KDE 3 did, plus the desktop that displays on the root window lives in the same process as the other native plasma code :-/ So that means more bug exposure, probably.
I've known the topic of sandboxing the native widgets come up and the response was "one day", probably in the "someone could do this but the core developers feel they have more important stuff to do" sense.
I found earlier releases of KDE 4 slowed things down particularly, I think it might have improved since then. It's not as snappy as I'd like but it's actually fairly usable now with my KDE 4.3 on Kubuntu Jaunty. I expect it would improve if I installed the proprietary NVidia drivers, or even just the Nouveau drivers.
Interestingly, I find that KDE 4.2 runs quite nicely on my Eee 701 with 2GB RAM as long as I don't run loads of apps - and that's with KWin compositing enabled and a few fairly heavyweight things running. The expose and desktop switcher are quite useful on the tiny screen.
I've seen various claims that KDE4 is much happier if there's compositing available but I'm not sure how accurate they are.
Plasma isn't just that thing for making desktop widgets of dubious usefulness. What KDE has actually done is, in my opinion, a fairly smart design move regardless of whether you like their implementation.
Desktop widgets aren't applications, they are people extending the functionality of their desktop. What the KDE folks saw was that a well-designed API could be used to write the desktop UI itself (task bar, clock, pager, whatever), the things we used to use taskbar applets for (media player control, etc) and the flashy new desktop widgets. Instead of having a basic desktop and plastering a widget API on top, they've gone and unified the whole thing so you can use the same API to write taskbar applets, widgets or write replacement taskbars or ... whatever. The various desktop elements are separate building blocks (plasmoids) that can be assembled together. They've also produced loads of bindings for this API to give folks the chance to write stuff in their favourite language.
The plasma netbook interface then takes some of the default building plasmoids, adds some new ones and then glues them together in a different way. So you can get a similar family look and similar functionality (and, fundamentally, the same desktop) but in a way that's optimised for a different form factor of device. I think that's actually pretty neat and somewhat reminiscent of the way you can configure and compile the core Linux kernel down for tiny machines or up to big iron whilst still getting the benefits of a common codebase.
There's a load of other cool stuff including a standard set of "data engines" which separate producing data from displaying it, thus making it easy to glue data sources together in interesting ways. Despite the various feature regressions that rewriting the desktop led to, it's a really neat architecture and should hopefully stand them in good stead for the future.
That's cool, I wasn't aware of that. Seems like it might be useful for scientific code. I know Fortran includes the ability to declare pure functions and (I think) verify their purity, which sounds useful.
Though I imagine the problem for larger computations could be that you're limited to writing functional-style code, at which point you might be better off using a language designed for it? What optimizations does gcc actually do on pure functions, having identified them? I think it would be very interesting to tie this with something like Apple's Grand Central Dispatch (in an automated way, of course) to get computations parallelised magically to whatever degree the hardware can support.
Because there are tons of vested interests. SAP is based in the EU, so there's the possibility they're lobbying the EC on this one. One assumes that Oracle / Sun are lobbying US senators (and politicians in the EU for that matter?). The EU, as the article points out, works under different rules and with a different viewpoint - Oracle and Sun agreed to be bound by local laws when they entered the European markets. The EU probably has a political interest in seeming to stand up to the US, though you'd hope the regulators wouldn't be swayed into unprofessional behaviour by that. The US has an interest in avoiding a precedent where the EU has power over one of their companies. Sun and Oracle are probably trying to dodge awkward questions and hope for the EU to cave. Really, there's no reason to believe 100% that anyone is acting entirely in good faith here, especially given we don't have access to all the information.
We're seeing an interesting consequence of the increasingly interconnected world, though, in that we're reaping business advantages from setting up shop in multiple large markets but in turn companies are then subject to multiple jurisdictions regardless of their country of origin. It seems like the EU and US regulators working together on a decision might be more appropriate, given neither of them has absolute authority to give the go ahead. A co-operative solution to regulation decisions would make a certain amount of sense since it's de facto what we have now. It's surely in nobody's interests for the decision to be left hanging.
Guys, I'm disappointed you haven't got here already. http://xkcd.com/178/
Heh, to be honest I'm mostly in the same boat as you - how can Haskell do side-effect stuff whilst still being purely functional. And what on earth is a Monad!?
A programmer friend of mine used to say that using monads to emulate state would be a bit like the following: imagine you want to emulate state but you're programming functionally. You want variables but you can't have them. Instead you have a big tuple of values representing the contents of all the "variables" and you pass this tuple to every function call and get back a new version with the variables updated. You're still programming purely functionally but you've got this emulation of mutable state.
But I mentioned this to another friend who has a PhD in compsci, specialising in compiler design and he looked at me like I was insane - so apparently that's not a good representation of how they behave. Still, it at least gets you a feel for how you could have imperative constructs like variables without breaking the rules of pure functional programming. Maybe that makes the idea of monads seem a little less impossible, even if it doesn't help explain how they work?
Other "functional" languages actually have an imperative subset (e.g. ML has variables, loop constructs, etc). Haskell is distinguished from those partly by being purely functional, so I presume that it really *is* purely functional by some definition, including the monads. But I can't say I understand how.
I had to study the type system of ML as an undergrad and it had this property of looking really opaque to me until I'd stared at it for hours and hours, then suddenly I went "Ooooh, I see!". Monads have a similar feel to me but I've never taken the time to stare them down. That said I imagine that, like the ML type system, they're much easier to use than to fully understand.
Well, pure functional languages are (potentially) good for concurrency in general. Because they have no mutable variables in the usual sense, it doesn't actually matter what order functions are evaluated in (other than the fact that callers cannot continue until their callees return). You can't do this in C or Java because it might be necessary for one function to see a variable modified by another. In a functional language, any dependencies are explicit call-return relationships (well, ish, they typically do have some non-functional constructs otherwise it's hard to do IO!), so in principle it's quite easy to split up a program's work across multiple CPUs (or machines) and not worry about whether they need to talk to each other.
Haskell, along with the ML family of languages, also has an amazing type checker that is waaay more sophisticated than most other languages. I think most people who've played around with these languages do start to feel that often "If it compiles, it's bug-free". Obviously that's not something you can rely on, since the compiler can't know what you meant to do. But it is true that the type system is *way* more useful at detecting bugs at compile-time than for any conventional language I've used.
It's bizarre but there still seems to be this perception that the police are a fine bunch of chaps who will universally do their best to apply the rules sensibly and fairly. There are plenty of police officers who that description applies to, I'm sure - but that's not an excuse for lawmakers and the justice system to assume it holds universally true.
At the end of the day, the police are there - in practice - *to catch potential criminals*. Sorting out who is and isn't guilty is not their job, that's the job of the courts (as it should be). So the police don't really have an incentive to be especially fair or reasonable; that's not what we've tasked them with doing. What lawmakers sometimes seem to fail to understand is that if we pressure them to achieve "catch all the terrorists / criminals" then they'll try to do that, even if they "catch" many innocent people too. If we give them new tools to do that then *they will use them*. If the tools we give them are extremely blunt instruments, like the ability to hold innocent people's data on the DNA database, they're going to use them to their fullest extent. If we want them to behave sensibly, the laws need to be more focused and less open to abuse.
It's the same issue with various "anti-terror" laws. Allegedly local councils in the UK have used these to put people under surveillance for reasons unrelated to terrorism (like whether they're using their rubbish bins correctly and whether they live in the locality of a school they have applied to). We gave them overly broad legislation and assumed that they wouldn't use it, even though it helps them to do what they see as their job. None of these organisations can be relied upon to act in the best interests in society because each of them only sees part of the big picture - our politicians are *supposed* to maintain the balance of power with targeted legislation that results in society's best interests being served overall. That goal can't be reached by handing out disproportionate powers indiscriminately.
>>>when I pay taxes I don't consider a state healthcare system to be a theft of my labour.
Even if you never get sick? i.e. You get hit by a car and die, or have a massive heart attack at age 80 and fall over dead. Then you've paid ~$500,000 in Euros towards the hospitals and got (virtually) nothing in return. That's almost as bad as the thousands we Americans were forced to give to AIG - welfare for the corporations.
Well, if I pay into a private health insurance plan and I never get sick, my money gets used to subsidise the treatment of those who did, right? And the other people in the private scheme will only be people who were wealthy enough to afford private insurance, so I'm subsidising relatively wealthy people.
I've still paid money for nothing in return and it's still been used to treat other people. Or do you just mean I should have the right to not pay into *any* health plan and just hope not to get sick?
In that case, I would see your point but I still wouldn't like to live in a country with that system, since I think it would result in a lot of down-and-out people who would be a burden on the economy in other ways (my life is better overall if the reasons for crime, bankruptcies, etc are minimised). It's like paying taxes for the police force - it may never help me directly but if it helps society function well then that benefits everyone. Yes, I actually think it's worth my subsidising other people's healthcare up to a point. Moreover, if the system is run right there are economies of scale to be had so that the cost is not prohibitive.
Personally, if there wasn't state healthcare here I'd be paying into a private insurance plan *anyhow*, since I can't guarantee I'll never be sick or injured. If I pay into a private insurance plan and don't get sick then only richer people benefit from my payments. If I pay into a state healthcare system then all of society benefits and I'll likely get indirect value out of that even if I'm never ill myself.
Consider trying OpenStreetMap for some of your navigation. http://www.openstreetmap.org/ It's a Wikipedia-like effort to create an Open Source-ish map of the world. The project was started in the UK and although coverage is still patchy in some areas it does already have excellent coverage of large parts of London. It includes some data that Google Maps does not have. Due to the Free as in Freedom nature of the project, the website also makes it easy to export PDFs, pngs, etc of areas of the map which you can then store for reference offline (though obviously you can't then search them).
There's a navigation service at http://yournavigation.org/ which has some neat features, like the ability to plot a route for pedestrians or cyclists that will take you down cycleways and footpaths where possible (there was also an elevation profile of your route, so you'll know how hilly it is! That was a GSoC project, I think but it seems not to be available for now).
Disclaimer: I am plugging a project I've contributed to here but it really is pretty good.
PS. There are internet kiosks but you'll need to ask your butler to bring some coal to power them. ;-)
At least Dr Dre has an academic title! ;-)
I can't speak for other Europeans but: when I pay taxes I don't consider a state healthcare system to be a theft of my labour. Leaving aside moral issues, I think it is maybe preferable to take the up-front financial cost of funding a state healthcare system than to take the social and financial penalties of creating a social and financial underclass who still have "costs" attached for society, even if they're not being paid welfare directly.
Secondly I don't mind because I'm investing money in a system I may need to use - and that system only works because a large enough body of people are paying into it. Insurance doesn't (just) work on the idea that you pay in enough to make your account profitable over time, it attempts to average out the relatively few unprofitable but nonetheless covered individuals across all of the clients with policies. So if you pay into an insurer, it's entirely possible that more sickly people will effectively benefit from your money. It's worth paying in because it gives you the ability to have treatment paid (effectively) out of the other client's fees, should you need to.
I was more speculating on a possible convergence - Android going the Chrome way, you might say. Although it'd be a bit surprising if they dropped compatibility completely, I guess. Good to see that the native support for Android has improved, although if they're allowing native code I'm even more perplexed as to why they went the Java route in the first place.
True. The short answer is that I do personally care, so it does affect my purchase decisions. The longer, wider ranging answer is that I don't think making a platform more limited for programmers is necessarily going to result in a better range of apps for the user, so it's not a strategy I want to see spread.
I wonder if this means Android will converge towards a more standard Linux, or if Chrome will converge to become less standard. Or if they'll keep the unique aspects of each and just try to unify stuff like browser code. I don't really fancy a phone that can only run web apps, or a "PC" that can only run Java apps compiled to a weird byte code! I don't really like the way Android has reinvented all of userspace, whereas at least Chrome builds on existing code a bit more. But they are solving different problems, which perhaps explains *some* of the differences...
A man recovering from spinal injuries goes to his hundredth physio session in just a few months. With the physio's help and all his strength he hauls himself upright and takes a step - his wife takes a photograph - before falling painfully to the ground. The insurer looks at the photograph "Congratulations, you're cured because you're standing and you couldn't do that before.".
The medical profession seems to generally agree that depression is a real condition, there's research to demonstrate that there's a real underlying chemical difference in the brains of sufferers. Are you saying that they're all wrong? If you are asserting that, it would be most helpful for you to tell us where you received your medical degree and provide a link to your list of peer-reviewed research papers in the area. If you don't have those then your claim is similar to the people who don't know anything of physics but are utterly convinced that the LHC will destroy the world and that they know better than the scientists who built it. If you do have those then you are, presumably, revolutionizing the state of medical science with your argument?
I'm not really clear whether you're saying depression doesn't exist or just that most people who claim benefits due to depression are lying. I think I've addressed the former. But if you meant the latter - is it really fair behaviour to accuse every single depression claimant of being a fraud because you heard that some people have been frauds? Do you apply the same logic (many X are Y, therefore I assume that all X are Y) in other areas of life?
You make it sound like the insurance company is forced to do this due to business realities, which I don't think is really fair. There are a lot of circumstances in which a business might like to get out of fulfilling the obligation to do what you paid them for - that's a reason for them to weasel out of stuff but it's not an excuse. If the woman has got a diagnosis saying she has depression and the insurance company sees evidence that she's improved then they could ask her to come back for a medical assessment.
To continue using your analogy of someone who supposedly couldn't move but is photographed living it up, unless they put some more thorough effort (than just a few Facebook photos) into examining her case it could equally be like saying "Oh, we saw them standing up even though they were supposed to be crippled" having just found a picture of a physio session where 30 seconds after the photo was taken the person fell down because they weren't strong enough to stand.
It's right (and understandable) for insurers to be careful in what they pay out for and to verify that people's needs are genuine. I just think that due diligence in this case involves a bit more than looking at a couple of photos without context. The cover is a service that you *pay* for after all, so if they're going to bail out on giving you that service they'd better have a solid explanation.
I own an air guitar and it's actually pretty sweet - I can make like I'm rocking out wherever I am. Whereas an air car ... I don't see the market. You're at a party and the music's pumping and you just decide to "air drive" to the shops? Not cool.
Chrubuntu ftaghn? ;-)
Some comedian - I think it was Ed Byrne - on the UK satirical news quiz "Have I Got News For You" made a good point about nuclear submarines. Probably misquoting slightly it was:
"We need to spend £20bn on a giant metal sausage under the sea in case one day we decide to destroy the world"
I think that put the whole scheme into perspective a bit.
When Palm tried to use Apple's software in a way Apple didn't like (syncing the Pre with iTunes), it got broken pretty fast - more than once. It's a different situation - Pre is a competitor to the iPhone and Apple aren't obliged to support a competitor. But Apple was going out and changing iTunes to make it not sync with the Pre, it's not like changes Apple made in normal development just *happened* to break the Pre. Whereas nobody would have batted an eyelid if Apple had accidentally broken the Pre, a lot of people felt that deliberately breaking it was not something they were keen on.
Seeing how aggressively they've pursued that case, you can see how they might get accused of doing a similar thing for the iPhone. I can just about imagine that Apple would have the arrogance to take a hard line on iPhone jailbreaking but I'd be mightily surprised if they had the legal and PR foolhardiness to actually deliberately brick devices.