The US pays a whack of money to support the airlines as it is. Did you ever wonder why most countries have a single airline? Its because the government pays a crap load of money so that they aren't at the whim of some other countries airline system. The US loves being different and one of the problems with propping up the diversity of so many carriers that can't fail is that there's less incentive to increase efficiency. The airlines may have tried to slash their spending over the years, but they're still horribly inefficient beasts. What's more, they're pegged so closely to Oil, that an inevitably dramatic rise in petroleum costs will directly affect flight costs. If the US let a few get merged or die, the US airline sector as a whole make look a little rosier.
"The pollution per person must be awful." Sorry, is this pollution per person over the net of the service, or are you looking at times that require cars to run with its capacity?
I would doubt that if you looked at the 'systems pollution output / persons serviced per day' that it would be worse than 'Car pollution / one or two people in the same car commuting per day'. Who cares if most of the cars are totally empty. If the system is a net loss of pollution, does it mean the system's broken? Worst case scenario, they run two or three sets of schedules to handle peak travel times and get the best of inconvenience / pollution&waste.
Um, why not put a train beside every interstate and then have either regular bus or light rail services into the suburbs? You already have the arteries, all you need to do is push through higher density conduits. Maybe it'll breath some much needed energy into the mostly dead night life in downtown America (some cities still have reasonable night life, but most that I've driven through are just depressing to see at night).
Obviously its not as crime riddled as LA, but up in Vancouver,BC we have cameras in trains. If there's a crime, they're on camera doing it. Say what you will about privacy, but the only times I've remotely felt insecure was around drunks that were too blitzed to know what they were doing. Speaking of which, another benefit to public transit is less drunk drivers assuming they run late (which Vancouver thankfully does).
Buses the end all long haul solution, but they are a great 'last mile or two' solution that a well engineered transit system has to use in order to thrive. In computer terms, I wouldn't cobble together DSL lines all over a city, but combined with fiber trunks and all of a sudden you have a viable, economical solution. Why doesn't everyone go out and buy a Cessna and drop the higher pollution, less flexible commercial airlines? With your own plane, you can go wherever you want whenever you want.
You answered faster than I did! I mean DUH, that's how these contests work. That is why companies release them, and that is why there will always be a niche software market for them. If anything, it really tells us that there is an over supply of talent just wasting away in the market if they all have time to join these contests and get recognition. I don't really know how big this market is, but I can't imagine that the rewards are much above table scraps when you calculate time invested.
Aw, that makes me sad. When I was developing for Windows in school, the one joy I had developing for Microsoft tech was their MSDN documentation. I hope it hasn't degenerated that much over the years.
Checked arithmetic (on overflow, they throw an exception) It would be handy, but if I really want to do high importance math in Java, I'd use a framework that was a little more robust than stock anyways.
Support for tail calls (for Lisp, F# and other functional languages) Coming to Java 7, but I generally hate the whole functional programing paradigm outside of a very few use cases, so no biggie for me
Value types, these are structs that are not wrapped in an object Don't see the benefit of being more procedural in an object oriented language, but good for you I suppose
VM-level support for generics Not missing it. Code level generally catches 99.99999% of any generic-type problems that I've ever run into
Platform-invoke allows developers to call native code without having to write any glue in C++ using JNI, it can all be done in the managed language. JNA?
The Common Language Specification: a spec that set the rules for interoperability across programming languages (for example: the rules for public identifier casing, handling of int vs uint and so on). Dynamic languages on top of the JVM apparently want to work together in integrating the higher level abstractions that java itself doesn't expose, but the rules in the basic java language generally carry forward to all derivative ones
Delegates allow user to keep a reference to a method or an instance method and invoke it. The VM also can turn any delegate invocation into an asynchronous invocation, so you can turn any method into an async method, like this: mySortFunc.BeginInvoke (array) Isn't this along the lines of Java reflection? java.lang.reflect? The only thing about reflection in java is that it has to be derived from a class originally. You can't just make up methods on the fly that are naked. You can use class instrumentation to build a brand new one method class and use that anonymously during runtime I suppose, but thats generally overkill. If and when I work on this sort of problem, I use java's Proxy mechanism which works pretty darn well in allowing for delegate handling of all interface methods (caveat is that it doesn't support concrete method signatures, just interface ones)
Support for dynamic code generation through the Reflection.Emit API. Java has class instrumentation that does similar if not the same thing. Look into ASM if you're at all interested. Unless you have specific non-trivial use cases, this is not relevant to 99% of all developers
A database file format allows for efficient reading of the metadata from assemblies. It does not require decompression and the database format is suitable for lazy loading. Is this a language or runtime feature? Although Java is almost always deployed using Jar's, the java spec doesn't mandate it. The whole classpath implementation in java is 'just the way it is', not necessarily a spec criteria. That said, I don't know the benefits, but I assume they're yummy...
Attributes were special objects that could annotate most elements in the virtual machine, you could annotate classes, methods, parameters, local variables, and at runtime the VM or the runtime could do interesting things with it. Like Java annotations?
Unsafe code (pointers) to support C++, Cobol and Fortran compilers running on the CLI. Good. I'd rather use nice safe code.
Native support for talking to COM-based APIs. Although mostly Windows-specific, its used to talk to Mozilla, VirtualBox and OpenOffice APIs in Linux. Good. Really, dealing with COM the few times that I did, there was no love lost. If you really need or love COM in java, there are products out
If (and possibly) when the source is leaked, there is 0 reprieve for MS besides suing the offender. The binaries change from release to release, so unless the copyright for a 'product' can expand for every possible forever derivative of that original work then MS could say that everything's a derivative or windows. All windows developers could just say that their products are derivatives of windows, so they shouldn't have to pay either. Publishers of new works could claim that since there are a few lines that are similar to those of book previously, so they don't have to pay either... A re-mastering of an existing copyrighted material has been explicitly stated as a new work justifying a new clock on the copyright clock, so do they get lumped together with the original work in your fees or separate?
Where do you draw the magic copyright for what is derivative (and included in the initial $5000 fee) and what is a new work of creation? Trust me, this entire solution is not tenable.
This could be: a. Yet another part of the ever growing, ever large conspiracy to cover up the existence of seeming observational aliens b. A maneuver by Churchill to silence a few pilots who didn't want to fly the channel anymore and made up a story to get out of getting themselves blown up. If the story had gotten out (true or not), it would've caused possible panic and more importantly a good reasons for pilots to refuse to fly, this with the backdrop of the a truly catastrophic war.
Read the article and find that the story is told by the grandson of a guard who overheard a conversation. Wow, that is just brilliant.
What about that *sobs* poor schmuck who who was brilliant and died young leaving his frail wife and kids on the streets with nothing but scraps to eat from. They can't -possibly- support themselves now, can they?
Joking aside, just have a fixed period of time since first creation, say 60 years. If I made one book when I was 16 and the copyright expired before I died, big woop, I should have done more throughout my life. Only the most brilliant of artists can support themselves through a single release and only a rare few of these artists just end it with one release. Allow immediate relatives (not corporations, etc..) receive the same protections if I do die before the expiration expires. If I was single and I died without a will then my works go public domain right away (or maybe after a -short- grace period, whatever).
The most harmful influence on the way copyright has fallen out has been in that corporations can acquire works now. Its no longer matters in supporting family and relatives long after you've died. It now has to do with getting your rights sold off to corporations that sit back and amass large collections of works that they will suck as much value out of as possible. Sometimes that means letting good works sit idle in 'the vault' for decades of miss use before doing something with it.
Easy to do with books, but imagine identifying and renewing code for all software you've ever released to the public. If MS releases a security patch every month, does that mean each patch has to pay 5000 every 10 years or be outed as public domain? If Linux doesn't renew the 5000 / 10 years fee for every single release of his codebase, does that mean his works go into the public domain as well? The costs are prohibitively expensive for works with natural evolution like software or mashups (though arguably mashups have a limited window of interest anyway). The idea of charging to 'remain in the system' simply cannot work. Maybe it had a chance 100 years ago when all one cared about was books and audio recordings.
Um, pretty much every big Hollywood adaptation in the last oh, 20 years has been from books or comics that are well over 14 years old. For example, spiderman, xmen, superman, ironman, lord of the rings, i robot, watchmen... There aren't that many adaptions that happen within that time of the books for one reason or another.
Copyright is broken and should be retracted, but 14 years definitely doesn't support the creators enough to get them by. In reality, unless you're a AAA 'whatever' creating content, you're not making much. You can just hope that some good TV/movie/book/recording/etc..'s rev sharing keeps putting meals on the table. The world can't live with just AAA content no matter what. If anything, the way the world is going, we'll need more content as the coming decades go as we further reduce the amount of work we (as a society) need to directly contribute to. Only (a large) war, recession, famine, peak oil, etc.. will start to affect the amount of time we have to spend on free time.
Why are you battling semantics? People make the comparison with iPhone's because Apple exclusively uses iOS on their phones. If They licensed the OS to other companies, you know damn well they'd be comparing iOS vs. Android. Just like how we always compare phones with Symbian and WinMo against RIM phones. Since RIM doesn't license, we never really refer to the OS independently. The only real break from this trend was with Palm and their WebOS. The WebOS became such a large selling point for Palm that people weren't talking about palm phones, they were talking about WebOS phones.
Plus, if they really just talked about handsets, the Android universe is so diversified that their handhelds could have 70-80% market share and still have iPhone beating every Android handset individually.
1. Requiring users to pay though my site is an extra hurdle that should be unnecessary for users regardless of what merchant service I link in the back end. The marketplace makes purchases like 2 clicks.
If I wanted to charge through my site it would greatly complicate the process, though it would shave the 30% commission that Google collects. I opened up a PayPal merchant account just for this purpose before I jumped through the hoops in order to get Google Checkout working through the US.
2. Google Checkout is even more retarded outside of selling Android apps. They only support US/UK retailers.
Piracy is a big problem on Android in no small part due to the pathetically small number of supported payee countries in the world. China should be 100% piracy because you can't legitimately buy apps from China. Same with any other country not in this list:
Australia
Austria
Canada
France
Germany
Italy
Japan
Netherlands
New Zealand
Spain
Switzerland
United Kingdom
United States
You want to sell apps legitimately through Google? They only allow paid developers from:
Austria France Germany Italy Japan Netherlands Spain United Kingdom United States
For a platform that Google want's to take over the world with, they're sure doing a piss poor job of doing it. Their marketplace is a sham that has to be addressed. No amount of DRM will fix the fundamental problem that even if you want to sell / pay for apps, most regions of the world can't do it anyways.
I personally had to open US bank accounts just to sell through the marketplace (I live in Canada). This adds no end of annoyance and extra banking fees. Coming from a small independent developer, this is yet another obstacle for successfully working in the Android ecosystem.
Android may be in Java (for the non-NDK based code anyways) but there is still a lot that one can do to bleed information to other apps. My first post on this topic described a quite viable and very difficult security hole to find or fix. Google may have an easier time attempting this, but its still a LOT of work to get it done. BTW, Android does have inter-app RPC which is yet another conduit of communication beyond just internet / filesystem.
One could say that Google's easy to integrate android API encourages apps to reuse each-other's functionality to better lift the platform and its integration. This also means that there are many vectors where apps can leak data between one another. Take a stupid example: From my non-internet permission contact book app, I can ask the web browser to navigate to 'http://myhomepage/a/encryptedcontactdata' which could just redirect the browser user to Google, an about page, or whatever. The only issues with such a vector is: 1. The activation of the internet browser has to be legitimately integrated with the app in order for savy users to not get suspicious. An about page would be a great example. The user may have forgotten that the app in question didn't have internet access to begin with. 2. If there are multiple browsers on the system, the user will be prompted to select the browser they want to use. At this point the user could just hit the back button to cancel the activity all together.
All of which can pretty well be extrapolated to a comfortable level of accuracy with IP geo tracking features. If not, there's always 'course' geo location which doesn't specifically know where you are, just the general area. I don't know the range of accuracy here, so it still may be pretty invasive.
Once these 'vetted' apps are gold star approved by Google and still rip you off for all your data, is Google now put in a liable state because they said it was OK?
No, it really doesn't matter how well the app store works. If you allow an app to run with elevated privileges (REGARDLESS of what it is/does) you will have the possibility of getting ripped off. Lets say some guy decides to write a phone book app in ruby or python. Is Google going to sift through the entire language runtime for every app that implements them? Your solution is impractical and unrealistic.
The better approach is to: 1. Educate users on what permissions are before they can download an applications. Deny them the ability to download until they are well aware of what these permissions are, and WHY they are important. Add this to the EULA so that they can't get sued for doing it. 2. Have Google consider sheltered hubs where apps with specific status cannot directly access public files, network, or other processes that may leak the data.
Eg. I could have a 100% legitimate address book that just happens to clone all private contact data non-app disk storage as an encrypted data 'cache' file. The app could be amazing and do everything people wanted and eventually become very popular. Later on, I could introduce other apps that serves another purposes, like an online news reader from the same company or as a 3rd party (doesn't matter for the exploit). This app just reads all the data that application #1 has now exposed through the file system and uses its own internet access permission to transmit the data to my collection server. In the end I have two successful apps with a privacy leak that's very difficult to find / detect for even pro dev's. This is not a solution that Google (or any company) can just wave a magic wand and have disappear.
Security is not easy, and by allowing that one malicious application on your system, you can basically throw away all privacy. Its the same on PC's, WinMo, iPhone, blackberry, etc...
"Things - we don't have any idea what he was doing before the movie started (or before the prequel comic, if you read that), everything just gets going. Several times in the movie they point out that one of the ways you can tell you're in a dream"..
I had the same first impression of the story while I was watching the film. I imagined that his kids were really dead and that he went into a long sleep to 'be' with them again. I could still hold the opinion, but the movie didn't really have that type of feel to it. You can wrap yourself up as many times as you like, but I think it was pretty well established that there were no cheats in the 'real' world to make one assume that it wasn't the real world (except for the end, but that's a result of his reawakening or not).
What really left me with a big question. The lead up as I gathered is: 1. In order to wake up from the dream, you have to: die (in the dream), subconsciously wake oneself up (described as a 'kick'), or die in reality (in which case you're really dead) 2. Dreamers inside the dream receive the same requirements 3. The first layer hack's wake up was the feeling of falling (hitting the truck against the rail then water) 4. The second layer hack was the same (hitting the elevator floor) 5. Due to chemicals used, dreamers that die drop into 'limbo' which is the left over cruft of all dreamer's subconscious 6. Cobb is VERY afraid of people falling into limbo, because they would have to wait a long time (weeks in 2nd level dreaming, years when in 3rd level dream) in limbo before being able to escape
It turns out that the way out of limbo is to kill yourself?? I thought the big thing with being in limbo was that there was no way out until you eventually woke up. The way it was portrayed in the film, you could just get out of limbo if you 'die(in your dream)'. Doesn't that just defeat the whole fear of going to limbo to begin with? I mean Cobb's first foray into limbo only ended when they killed themselves. How could he not care to share that fact with everyone else? Can anyone clear this one up?
I think that the CLR thing has more credence than some of the other suspicions. That type of chip is really something that only Microsoft would spearhead, and the fact is that I don't think MS would make a better chip than what it would cost to license it from others.
That said, a Desktop level performance ARM chip is something that hasn't been done yet, so maybe this is just an Xbox play.. It couldn't hurt their 'runs on all devices' initiative having two of their platforms sharing the same CPU architecture.
Slashdot Mirror
The US pays a whack of money to support the airlines as it is. Did you ever wonder why most countries have a single airline? Its because the government pays a crap load of money so that they aren't at the whim of some other countries airline system. The US loves being different and one of the problems with propping up the diversity of so many carriers that can't fail is that there's less incentive to increase efficiency. The airlines may have tried to slash their spending over the years, but they're still horribly inefficient beasts. What's more, they're pegged so closely to Oil, that an inevitably dramatic rise in petroleum costs will directly affect flight costs. If the US let a few get merged or die, the US airline sector as a whole make look a little rosier.
"The pollution per person must be awful."
Sorry, is this pollution per person over the net of the service, or are you looking at times that require cars to run with its capacity?
I would doubt that if you looked at the 'systems pollution output / persons serviced per day' that it would be worse than 'Car pollution / one or two people in the same car commuting per day'. Who cares if most of the cars are totally empty. If the system is a net loss of pollution, does it mean the system's broken? Worst case scenario, they run two or three sets of schedules to handle peak travel times and get the best of inconvenience / pollution&waste.
Um, why not put a train beside every interstate and then have either regular bus or light rail services into the suburbs? You already have the arteries, all you need to do is push through higher density conduits. Maybe it'll breath some much needed energy into the mostly dead night life in downtown America (some cities still have reasonable night life, but most that I've driven through are just depressing to see at night).
Obviously its not as crime riddled as LA, but up in Vancouver,BC we have cameras in trains. If there's a crime, they're on camera doing it. Say what you will about privacy, but the only times I've remotely felt insecure was around drunks that were too blitzed to know what they were doing. Speaking of which, another benefit to public transit is less drunk drivers assuming they run late (which Vancouver thankfully does).
Buses the end all long haul solution, but they are a great 'last mile or two' solution that a well engineered transit system has to use in order to thrive. In computer terms, I wouldn't cobble together DSL lines all over a city, but combined with fiber trunks and all of a sudden you have a viable, economical solution. Why doesn't everyone go out and buy a Cessna and drop the higher pollution, less flexible commercial airlines? With your own plane, you can go wherever you want whenever you want.
You answered faster than I did! I mean DUH, that's how these contests work. That is why companies release them, and that is why there will always be a niche software market for them. If anything, it really tells us that there is an over supply of talent just wasting away in the market if they all have time to join these contests and get recognition. I don't really know how big this market is, but I can't imagine that the rewards are much above table scraps when you calculate time invested.
Aw, that makes me sad. When I was developing for Windows in school, the one joy I had developing for Microsoft tech was their MSDN documentation. I hope it hasn't degenerated that much over the years.
Unsigned data types
Wow, I'm sold
Checked arithmetic (on overflow, they throw an exception)
It would be handy, but if I really want to do high importance math in Java, I'd use a framework that was a little more robust than stock anyways.
Support for tail calls (for Lisp, F# and other functional languages)
Coming to Java 7, but I generally hate the whole functional programing paradigm outside of a very few use cases, so no biggie for me
Value types, these are structs that are not wrapped in an object
Don't see the benefit of being more procedural in an object oriented language, but good for you I suppose
VM-level support for generics
Not missing it. Code level generally catches 99.99999% of any generic-type problems that I've ever run into
Platform-invoke allows developers to call native code without having to write any glue in C++ using JNI, it can all be done in the managed language.
JNA?
The Common Language Specification: a spec that set the rules for interoperability across programming languages (for example: the rules for public identifier casing, handling of int vs uint and so on).
Dynamic languages on top of the JVM apparently want to work together in integrating the higher level abstractions that java itself doesn't expose, but the rules in the basic java language generally carry forward to all derivative ones
Delegates allow user to keep a reference to a method or an instance method and invoke it. The VM also can turn any delegate invocation into an asynchronous invocation, so you can turn any method into an async method, like this: mySortFunc.BeginInvoke (array)
Isn't this along the lines of Java reflection? java.lang.reflect? The only thing about reflection in java is that it has to be derived from a class originally. You can't just make up methods on the fly that are naked. You can use class instrumentation to build a brand new one method class and use that anonymously during runtime I suppose, but thats generally overkill. If and when I work on this sort of problem, I use java's Proxy mechanism which works pretty darn well in allowing for delegate handling of all interface methods (caveat is that it doesn't support concrete method signatures, just interface ones)
Support for dynamic code generation through the Reflection.Emit API.
Java has class instrumentation that does similar if not the same thing. Look into ASM if you're at all interested. Unless you have specific non-trivial use cases, this is not relevant to 99% of all developers
A database file format allows for efficient reading of the metadata from assemblies. It does not require decompression and the database format is suitable for lazy loading.
Is this a language or runtime feature? Although Java is almost always deployed using Jar's, the java spec doesn't mandate it. The whole classpath implementation in java is 'just the way it is', not necessarily a spec criteria. That said, I don't know the benefits, but I assume they're yummy...
Attributes were special objects that could annotate most elements in the virtual machine, you could annotate classes, methods, parameters, local variables, and at runtime the VM or the runtime could do interesting things with it.
Like Java annotations?
Unsafe code (pointers) to support C++, Cobol and Fortran compilers running on the CLI.
Good. I'd rather use nice safe code.
Native support for talking to COM-based APIs. Although mostly Windows-specific, its used to talk to Mozilla, VirtualBox and OpenOffice APIs in Linux.
Good. Really, dealing with COM the few times that I did, there was no love lost. If you really need or love COM in java, there are products out
If only OPEC could be held to the standards of everyone else...
If (and possibly) when the source is leaked, there is 0 reprieve for MS besides suing the offender. The binaries change from release to release, so unless the copyright for a 'product' can expand for every possible forever derivative of that original work then MS could say that everything's a derivative or windows. All windows developers could just say that their products are derivatives of windows, so they shouldn't have to pay either. Publishers of new works could claim that since there are a few lines that are similar to those of book previously, so they don't have to pay either... A re-mastering of an existing copyrighted material has been explicitly stated as a new work justifying a new clock on the copyright clock, so do they get lumped together with the original work in your fees or separate?
Where do you draw the magic copyright for what is derivative (and included in the initial $5000 fee) and what is a new work of creation? Trust me, this entire solution is not tenable.
You should watch Path of Glory.. good old Kubrick..
This could be:
a. Yet another part of the ever growing, ever large conspiracy to cover up the existence of seeming observational aliens
b. A maneuver by Churchill to silence a few pilots who didn't want to fly the channel anymore and made up a story to get out of getting themselves blown up. If the story had gotten out (true or not), it would've caused possible panic and more importantly a good reasons for pilots to refuse to fly, this with the backdrop of the a truly catastrophic war.
Read the article and find that the story is told by the grandson of a guard who overheard a conversation. Wow, that is just brilliant.
What about that *sobs* poor schmuck who who was brilliant and died young leaving his frail wife and kids on the streets with nothing but scraps to eat from. They can't -possibly- support themselves now, can they?
Joking aside, just have a fixed period of time since first creation, say 60 years. If I made one book when I was 16 and the copyright expired before I died, big woop, I should have done more throughout my life. Only the most brilliant of artists can support themselves through a single release and only a rare few of these artists just end it with one release. Allow immediate relatives (not corporations, etc..) receive the same protections if I do die before the expiration expires. If I was single and I died without a will then my works go public domain right away (or maybe after a -short- grace period, whatever).
The most harmful influence on the way copyright has fallen out has been in that corporations can acquire works now. Its no longer matters in supporting family and relatives long after you've died. It now has to do with getting your rights sold off to corporations that sit back and amass large collections of works that they will suck as much value out of as possible. Sometimes that means letting good works sit idle in 'the vault' for decades of miss use before doing something with it.
Easy to do with books, but imagine identifying and renewing code for all software you've ever released to the public. If MS releases a security patch every month, does that mean each patch has to pay 5000 every 10 years or be outed as public domain? If Linux doesn't renew the 5000 / 10 years fee for every single release of his codebase, does that mean his works go into the public domain as well? The costs are prohibitively expensive for works with natural evolution like software or mashups (though arguably mashups have a limited window of interest anyway). The idea of charging to 'remain in the system' simply cannot work. Maybe it had a chance 100 years ago when all one cared about was books and audio recordings.
Um, pretty much every big Hollywood adaptation in the last oh, 20 years has been from books or comics that are well over 14 years old. For example, spiderman, xmen, superman, ironman, lord of the rings, i robot, watchmen... There aren't that many adaptions that happen within that time of the books for one reason or another.
Copyright is broken and should be retracted, but 14 years definitely doesn't support the creators enough to get them by. In reality, unless you're a AAA 'whatever' creating content, you're not making much. You can just hope that some good TV/movie/book/recording/etc..'s rev sharing keeps putting meals on the table. The world can't live with just AAA content no matter what. If anything, the way the world is going, we'll need more content as the coming decades go as we further reduce the amount of work we (as a society) need to directly contribute to. Only (a large) war, recession, famine, peak oil, etc.. will start to affect the amount of time we have to spend on free time.
http://en.wikipedia.org/wiki/File:IPhone_sales_per_quarter_simple.svg
Yep, the market really cooled off in Q1/Q2 in bated anticipation...
Why are you battling semantics? People make the comparison with iPhone's because Apple exclusively uses iOS on their phones. If They licensed the OS to other companies, you know damn well they'd be comparing iOS vs. Android. Just like how we always compare phones with Symbian and WinMo against RIM phones. Since RIM doesn't license, we never really refer to the OS independently. The only real break from this trend was with Palm and their WebOS. The WebOS became such a large selling point for Palm that people weren't talking about palm phones, they were talking about WebOS phones.
Plus, if they really just talked about handsets, the Android universe is so diversified that their handhelds could have 70-80% market share and still have iPhone beating every Android handset individually.
1. Requiring users to pay though my site is an extra hurdle that should be unnecessary for users regardless of what merchant service I link in the back end. The marketplace makes purchases like 2 clicks.
If I wanted to charge through my site it would greatly complicate the process, though it would shave the 30% commission that Google collects. I opened up a PayPal merchant account just for this purpose before I jumped through the hoops in order to get Google Checkout working through the US.
2. Google Checkout is even more retarded outside of selling Android apps. They only support US/UK retailers.
Piracy is a big problem on Android in no small part due to the pathetically small number of supported payee countries in the world. China should be 100% piracy because you can't legitimately buy apps from China. Same with any other country not in this list:
Australia
Austria
Canada
France
Germany
Italy
Japan
Netherlands
New Zealand
Spain
Switzerland
United Kingdom
United States
You want to sell apps legitimately through Google? They only allow paid developers from:
Austria
France
Germany
Italy
Japan
Netherlands
Spain
United Kingdom
United States
For a platform that Google want's to take over the world with, they're sure doing a piss poor job of doing it. Their marketplace is a sham that has to be addressed. No amount of DRM will fix the fundamental problem that even if you want to sell / pay for apps, most regions of the world can't do it anyways.
I personally had to open US bank accounts just to sell through the marketplace (I live in Canada). This adds no end of annoyance and extra banking fees. Coming from a small independent developer, this is yet another obstacle for successfully working in the Android ecosystem.
Android may be in Java (for the non-NDK based code anyways) but there is still a lot that one can do to bleed information to other apps. My first post on this topic described a quite viable and very difficult security hole to find or fix. Google may have an easier time attempting this, but its still a LOT of work to get it done. BTW, Android does have inter-app RPC which is yet another conduit of communication beyond just internet / filesystem.
One could say that Google's easy to integrate android API encourages apps to reuse each-other's functionality to better lift the platform and its integration. This also means that there are many vectors where apps can leak data between one another. Take a stupid example: From my non-internet permission contact book app, I can ask the web browser to navigate to 'http://myhomepage/a/encryptedcontactdata' which could just redirect the browser user to Google, an about page, or whatever. The only issues with such a vector is:
1. The activation of the internet browser has to be legitimately integrated with the app in order for savy users to not get suspicious. An about page would be a great example. The user may have forgotten that the app in question didn't have internet access to begin with.
2. If there are multiple browsers on the system, the user will be prompted to select the browser they want to use. At this point the user could just hit the back button to cancel the activity all together.
All of which can pretty well be extrapolated to a comfortable level of accuracy with IP geo tracking features. If not, there's always 'course' geo location which doesn't specifically know where you are, just the general area. I don't know the range of accuracy here, so it still may be pretty invasive.
Once these 'vetted' apps are gold star approved by Google and still rip you off for all your data, is Google now put in a liable state because they said it was OK?
No, it really doesn't matter how well the app store works. If you allow an app to run with elevated privileges (REGARDLESS of what it is/does) you will have the possibility of getting ripped off. Lets say some guy decides to write a phone book app in ruby or python. Is Google going to sift through the entire language runtime for every app that implements them? Your solution is impractical and unrealistic.
The better approach is to:
1. Educate users on what permissions are before they can download an applications. Deny them the ability to download until they are well aware of what these permissions are, and WHY they are important. Add this to the EULA so that they can't get sued for doing it.
2. Have Google consider sheltered hubs where apps with specific status cannot directly access public files, network, or other processes that may leak the data.
Eg.
I could have a 100% legitimate address book that just happens to clone all private contact data non-app disk storage as an encrypted data 'cache' file. The app could be amazing and do everything people wanted and eventually become very popular.
Later on, I could introduce other apps that serves another purposes, like an online news reader from the same company or as a 3rd party (doesn't matter for the exploit). This app just reads all the data that application #1 has now exposed through the file system and uses its own internet access permission to transmit the data to my collection server. In the end I have two successful apps with a privacy leak that's very difficult to find / detect for even pro dev's. This is not a solution that Google (or any company) can just wave a magic wand and have disappear.
Security is not easy, and by allowing that one malicious application on your system, you can basically throw away all privacy. Its the same on PC's, WinMo, iPhone, blackberry, etc...
"Things - we don't have any idea what he was doing before the movie started (or before the prequel comic, if you read that), everything just gets going. Several times in the movie they point out that one of the ways you can tell you're in a dream"..
I had the same first impression of the story while I was watching the film. I imagined that his kids were really dead and that he went into a long sleep to 'be' with them again. I could still hold the opinion, but the movie didn't really have that type of feel to it. You can wrap yourself up as many times as you like, but I think it was pretty well established that there were no cheats in the 'real' world to make one assume that it wasn't the real world (except for the end, but that's a result of his reawakening or not).
What really left me with a big question. The lead up as I gathered is:
1. In order to wake up from the dream, you have to: die (in the dream), subconsciously wake oneself up (described as a 'kick'), or die in reality (in which case you're really dead)
2. Dreamers inside the dream receive the same requirements
3. The first layer hack's wake up was the feeling of falling (hitting the truck against the rail then water)
4. The second layer hack was the same (hitting the elevator floor)
5. Due to chemicals used, dreamers that die drop into 'limbo' which is the left over cruft of all dreamer's subconscious
6. Cobb is VERY afraid of people falling into limbo, because they would have to wait a long time (weeks in 2nd level dreaming, years when in 3rd level dream) in limbo before being able to escape
It turns out that the way out of limbo is to kill yourself?? I thought the big thing with being in limbo was that there was no way out until you eventually woke up. The way it was portrayed in the film, you could just get out of limbo if you 'die(in your dream)'. Doesn't that just defeat the whole fear of going to limbo to begin with? I mean Cobb's first foray into limbo only ended when they killed themselves. How could he not care to share that fact with everyone else? Can anyone clear this one up?
I think that the CLR thing has more credence than some of the other suspicions. That type of chip is really something that only Microsoft would spearhead, and the fact is that I don't think MS would make a better chip than what it would cost to license it from others.
That said, a Desktop level performance ARM chip is something that hasn't been done yet, so maybe this is just an Xbox play.. It couldn't hurt their 'runs on all devices' initiative having two of their platforms sharing the same CPU architecture.
*cough* Xscale was sold to Marvell 4 years ago...