Obviously you don't know what you are talking about. It is one thing to get the ADC account, get the iPhone SDK and all the doc and develop the application. Then if you actually want to try it on your own device, you are out of luck and have to fork out $99 a year for this to happen. This is what the original post complained about. Add my vote, it is strange business model by Apple.
A whole $99 a year so you can sell your app on the most successful mobile app platform. The only thing strange is that Apple the other mobile developers while offering a richer SDK when thy could have charged more. f you don?t want to sell your app on the App Store then don?t pay the $99. The only issue here is trying to find a reason to raise your pitchfork and torch.
Even more outrageous, still, are the cost to develop for handheld gaming devices. This has kept plenty of talented developers off these devices do to excessive start up costs. The iPhone doesn?t have this issue while still having an SDk that allows for APIs that can utilize the HW, unless Java apps on other mobile phone platforms. There is a reason Nintendo and Sony are having to change up their strategy.
That's more of a case of Windows allowing it rather Apple. There is huge toolchain that will not function 100% correctly on Windows.
I don’t even see how that would work correctly. iPhone apps are compiled, not interpreted. I am sure MS uses Macs for their Mac OS X related development just as Apple uses Windows for their Windows-related development.
That is something that has nothing to do with this thread. The article is about the automation in checking for unapproved calls, not content.
No you comment is off thread , we were discussing censorship of apps which this developer obviously was a victim of. I was responding to another post regarding censorship of apps. Give it up.
A whole $99 a year so you can sell your app on the most successful mobile app platform. The only thing strange is that Apple the other mobile developers while offering a richer SDK when thy could have charged more. f you don?t want to sell your app on the App Store then don?t pay the $99. The only issue here is trying to find a reason to raise your pitchfork and torch.
Even more outrageous, still, are the cost to develop for handheld gaming devices. This has kept plenty of talented developers off these devices do to excessive start up costs. The iPhone doesn?t have this issue while still having an SDk that allows for APIs that can utilize the HW, unless Java apps on other mobile phone platforms. There is a reason Nintendo and Sony are having to change up their strategy.
Please read the original post. It was about the fact, that if you want to install an application you developed on your device (not selling it to the others), you have to pay $99 year to Apple. Which is either true and very odd.
Please read the original post. It was about the fact, that if you want to install an application you developed on your device (not selling it to the others), you have to pay $99 year to Apple. Which is either true and very odd.
And if read my post you?ll see that I clearly state that you don?t need to pay Apple a dime to create and use your own app.
And if read my post you?ll see that I clearly state that you don?t need to pay Apple a dime to create and use your own app.
Where did you get that notion from?
If by "use" you mean on your iPhone/iPod (otherwise what's the point) then you absolutely do have to pay Apple $99. It's the only way to run apps on an non-jailbroken phone.
If by "use" you mean on your iPhone/iPod (otherwise what's the point) then you absolutely do have to pay Apple $99. It's the only way to run apps on an non-jailbroken phone.
Jailbreak it. It was down before Apple had an App Store it can be done now. If a developer can code in C then they surely can figure out how to run the very simple jailbreaking apps and get there app on their iDevice. Look for solutions, not problems.
Jailbreak it. It was down before Apple had an App Store it can be done now. If a developer can code in C then they surely can figure out how to run the very simple jailbreaking apps and get there app on their iDevice. Look for solutions, not problems.
To be honest I'd rather run a phone that hasn't been hacked to make it work. I mean jailbreaking works by replacing system libraries and files with patched and sometimes older versions which weren't intended to work with version of OS being run. This will general re-introduce bugs/security vulnerabilities that were fixed by Apple, or worse still introduces new bugs that Apple would never have tested for.
I really fail to see how jailbreaking is a good thing. The official SDK is easier to work with than the jailbroken toolset (I did try once it was painful).
Most developers have no interest at all in anything other than official development. So jailBREAKING doesn't really count, it certainly isn't something a casual developer would care about ever.
To be honest I'd rather run a phone that hasn't been hacked to make it work. I mean jailbreaking works by replacing system libraries and files with patched and sometimes older versions which weren't intended to work with version of OS being run. This will general re-introduce bugs/security vulnerabilities that were fixed by Apple, or worse still introduces new bugs that Apple would never have tested for.
I really fail to see how jailbreaking is a good thing. The official SDK is easier to work with than the jailbroken toolset (I did try once it was painful).
Most developers have no interest at all in anything other than official development. So jailBREAKING doesn't really count, it certainly isn't something a casual developer would care about ever.
Jailbreaking counts plenty.
Regardless, if you want to install an app you?e built and are too cheap to pay $99 for the official method then there is jailbreaking. This entitlement that a company should bend over to give you exactly what you want when you want without you having to think of how you can do it yourself is asinine.
Apple assumes that iPhone OS devs want to push their wares so they charge $99 for this. It?s not much but if you don?t want to pay it then jailbreak your phone to install your app. It?s not a big deal. life is full of choices that are not ideal, you just man man up and pick the best option for yourself.
Regarding the actual topic of the article, this story shows a horrible weakness in Apple's development technology. If a software object has been marked Private, there should be no way for any developer to see it, never mind use it. These analysis tools are using the Microsoft anti-virus approach: we didn't secure the system properly, so now create programs to chase all the security infringements around.
Regarding the actual topic of the article, this story shows a horrible weakness in Apple's development technology. If a software object has been marked Private, there should be no way for any developer to see it, never mind use it. These analysis tools are using the Microsoft anti-virus approach: we didn't secure the system properly, so now create programs to chase all the security infringements around.
Apple can do better.
Marking private is the equivalent of sticking a "Do Not Enter" sign on a door. It doesn't stop people from entering does it?
Private as we see in languages like C# is nothing more than a compile time restriction. Even in C# you can call a private method using reflection, I have done this myself.
Private is just a concept, in reality a private method is just a public method that may or may not have compiler flags to throw errors or warnings.
In all sort of different programming languages you can see examples where they say private is just a piece of metadata, anyone can call this function regardless.
The way Objective-C works (correctly as far as object oriented programming goes) and due to the fact it is a dynamically typed language, the restrictions required for a c# style private "error" cannot work. If you look at "Object Messaging" an example would behttp://www.inf.ufsc.br/poo/smalltalk...orial/oop.html then you can see that if an unsupported method is called, it MUST be passed up the object hierarchy until it responded to. If nothing responds then a runtime error occurs.
This principle is FUNDAMENTAL to PROPER (i.e. not C++) object-oriented programming. There is no way at all that it can ever be implemented.
These analysis tools are not doing what you suggest at all. They are standard static analysis tools used (most likely clang based) to ensure safer, better quality code. This is an industry recognised practice.
Just to re-cap:
public = public
internal = public
protected = public
private = public
It's only the compiler/compiler flags that enforce errors/warnings.
Marking private is the equivalent of sticking a "Do Not Enter" sign on a door. It doesn't stop people from entering does it?
Private as we see in languages like C# is nothing more than a compile time restriction. Even in C# you can call a private method using reflection, I have done this myself.
Private is just a concept, in reality a private method is just a public method that may or may not have compiler flags to throw errors or warnings.
In all sort of different programming languages you can see examples where they say private is just a piece of metadata, anyone can call this function regardless.
The way Objective-C works (correctly as far as object oriented programming goes) and due to the fact it is a dynamically typed language, the restrictions required for a c# style private "error" cannot work. If you look at "Object Messaging" an example would behttp://www.inf.ufsc.br/poo/smalltalk...orial/oop.html then you can see that if an unsupported method is called, it MUST be passed up the object hierarchy until it responded to. If nothing responds then a runtime error occurs.
This principle is FUNDAMENTAL to PROPER (i.e. not C++) object-oriented programming. There is no way at all that it can ever be implemented.
These analysis tools are not doing what you suggest at all. They are standard static analysis tools used (most likely clang based) to ensure safer, better quality code. This is an industry recognised practice.
Just to re-cap:
public = public
internal = public
protected = public
private = public
It's only the compiler/compiler flags that enforce errors/warnings.
Java solved this by OSGi. It can be done. That is the message of the original post. Those are just technicalities.
I do not know about dot net platform but in priciple as this is what they call "managed" the private == public must not be the truth!
What I do know is that the Java architecture hade security closure designed into it, simply and brillianty from day one. It was used for applets and later for the widely used j2ME platform. Java Midlets (phone apps) can safely be run no matter who wrote them or how sinister the author is because the program is run inside a virtual env, 'sandbox' and can never escape in theory. I can never access private apis, never access other Midlets data. The worst that can happen is that the program does not do what it said it should. This, of course relies on correct implementation of the sanbox, and a number of APIs which is the responsibility of the platform provider, but it is logically a truth that it is safe if these implementations are correct. Still hard to prove but a lot better that just leaving anyting open. Apple's "stone age" Obj C-language can never prove any like properties so this inferior manual/automatic process is just what to expect from at 20 year old technology without any firm base.
I do not know about dot net platform but in priciple as this is what they call "managed" the private == public must not be the truth!
What I do know is that the Java architecture hade security closure designed into it, simply and brillianty from day one. It was used for applets and later for the widely used j2ME platform. Java Midlets (phone apps) can safely be run no matter who wrote them or how sinister the author is because the program is run inside a virtual env, 'sandbox' and can never escape in theory. I can never access private apis, never access other Midlets data. The worst that can happen is that the program does not do what it said it should. This, of course relies on correct implementation of the sanbox, and a number of APIs which is the responsibility of the platform provider, but it is logically a truth that it is safe if these implementations are correct. Still hard to prove but a lot better that just leaving anyting open. Apple's "stone age" Obj C-language can never prove any like properties so this inferior manual/automatic process is just what to expect from at 20 year old technology without any firm base.
You really haven't a clue what you are talking about do you?
Reflective programming (reflection) is an extension to object-oriented programming. Java too allows the calling of private methods via reflection. A little snippet from Wikipedia:
Quote:
Reflection is most commonly used in high-level virtual machine programming languages like Smalltalk and scripting languages, and less commonly used in manifestly typed and/or statically typed programming languages such as Java and C.
As for your whole sandboxing thing, what the hell has that got to do with anything? It certainly has nothing to do with public/private accessors.
A class, at its very base level, is just a grouping of related functions.
Again, the public/protected/internal/private keywords are really just compiler directives to tell the compiler to error or throw a warning should a developer try and call the code. If a method has an address (which all of them have to have) you can call it by some means, it's just a case of whether or not you can easily glean much information about the implementation (which Reflection allows).
There is nothing inferior about Objective-C, least of all because of its age, where did you get that idea from?
By that reasoning, Assembly is inferior to Java, which is obviously not the case, least of all because of performance.
And also you say all dynamically typed languages (including C) are inferior to Java, because you "can never prove any like properties". Again, utter rubbish.
Objective-C is a very elegant programming language and is closer to the idea of object-oriented programming envisioned by its creator than C++ or Java will ever be.
Comments
Obviously you don't know what you are talking about. It is one thing to get the ADC account, get the iPhone SDK and all the doc and develop the application. Then if you actually want to try it on your own device, you are out of luck and have to fork out $99 a year for this to happen. This is what the original post complained about. Add my vote, it is strange business model by Apple.
A whole $99 a year so you can sell your app on the most successful mobile app platform. The only thing strange is that Apple the other mobile developers while offering a richer SDK when thy could have charged more. f you don?t want to sell your app on the App Store then don?t pay the $99. The only issue here is trying to find a reason to raise your pitchfork and torch.
Even more outrageous, still, are the cost to develop for handheld gaming devices. This has kept plenty of talented developers off these devices do to excessive start up costs. The iPhone doesn?t have this issue while still having an SDk that allows for APIs that can utilize the HW, unless Java apps on other mobile phone platforms. There is a reason Nintendo and Sony are having to change up their strategy.
Really? Then what is this? http://www.macobserver.com/tmo/artic...litical_caric/
That is something that has nothing to do with this thread. The article is about the automation in checking for unapproved calls, not content.
My biggest hope is for them to allow iphone App development on windows. I would be on that quick!
That's more of a case of Windows allowing it rather Apple. There is huge toolchain that will not function 100% correctly on Windows.
That's more of a case of Windows allowing it rather Apple. There is huge toolchain that will not function 100% correctly on Windows.
I don’t even see how that would work correctly. iPhone apps are compiled, not interpreted. I am sure MS uses Macs for their Mac OS X related development just as Apple uses Windows for their Windows-related development.
You can burn songs from iTunes onto a CD if you want.
My CD's joined the boxes of video tapes, cassette tapes and vinyl stored in my garage.
The world changes and nothing is going to stop that.
So have you bought a Blu-ray player yet- the most cutiing edge video technology in this ever changing world you speak of?
That is something that has nothing to do with this thread. The article is about the automation in checking for unapproved calls, not content.
No you comment is off thread , we were discussing censorship of apps which this developer obviously was a victim of. I was responding to another post regarding censorship of apps. Give it up.
A whole $99 a year so you can sell your app on the most successful mobile app platform. The only thing strange is that Apple the other mobile developers while offering a richer SDK when thy could have charged more. f you don?t want to sell your app on the App Store then don?t pay the $99. The only issue here is trying to find a reason to raise your pitchfork and torch.
Even more outrageous, still, are the cost to develop for handheld gaming devices. This has kept plenty of talented developers off these devices do to excessive start up costs. The iPhone doesn?t have this issue while still having an SDk that allows for APIs that can utilize the HW, unless Java apps on other mobile phone platforms. There is a reason Nintendo and Sony are having to change up their strategy.
Please read the original post. It was about the fact, that if you want to install an application you developed on your device (not selling it to the others), you have to pay $99 year to Apple. Which is either true and very odd.
Please read the original post. It was about the fact, that if you want to install an application you developed on your device (not selling it to the others), you have to pay $99 year to Apple. Which is either true and very odd.
And if read my post you?ll see that I clearly state that you don?t need to pay Apple a dime to create and use your own app.
And if read my post you?ll see that I clearly state that you don?t need to pay Apple a dime to create and use your own app.
Where did you get that notion from?
If by "use" you mean on your iPhone/iPod (otherwise what's the point) then you absolutely do have to pay Apple $99. It's the only way to run apps on an non-jailbroken phone.
Where did you get that notion from?
If by "use" you mean on your iPhone/iPod (otherwise what's the point) then you absolutely do have to pay Apple $99. It's the only way to run apps on an non-jailbroken phone.
Jailbreak it. It was down before Apple had an App Store it can be done now. If a developer can code in C then they surely can figure out how to run the very simple jailbreaking apps and get there app on their iDevice. Look for solutions, not problems.
iPhone apps are interpreted, not compiled.
Where on earth did you get that idea from?
I can assure with 100% certainty that they are compiled.
Mainly because the Objective-C language they are written in is a compiled language.
Where on earth did you get that idea from?
I can assure with 100% certainty that they are compiled.
Mainly because the Objective-C language they are written in is a compiled language.
It?s obvious I flipped them by mistake.
Jailbreak it. It was down before Apple had an App Store it can be done now. If a developer can code in C then they surely can figure out how to run the very simple jailbreaking apps and get there app on their iDevice. Look for solutions, not problems.
To be honest I'd rather run a phone that hasn't been hacked to make it work. I mean jailbreaking works by replacing system libraries and files with patched and sometimes older versions which weren't intended to work with version of OS being run. This will general re-introduce bugs/security vulnerabilities that were fixed by Apple, or worse still introduces new bugs that Apple would never have tested for.
I really fail to see how jailbreaking is a good thing. The official SDK is easier to work with than the jailbroken toolset (I did try once it was painful).
Most developers have no interest at all in anything other than official development. So jailBREAKING doesn't really count, it certainly isn't something a casual developer would care about ever.
To be honest I'd rather run a phone that hasn't been hacked to make it work. I mean jailbreaking works by replacing system libraries and files with patched and sometimes older versions which weren't intended to work with version of OS being run. This will general re-introduce bugs/security vulnerabilities that were fixed by Apple, or worse still introduces new bugs that Apple would never have tested for.
I really fail to see how jailbreaking is a good thing. The official SDK is easier to work with than the jailbroken toolset (I did try once it was painful).
Most developers have no interest at all in anything other than official development. So jailBREAKING doesn't really count, it certainly isn't something a casual developer would care about ever.
Jailbreaking counts plenty.
Regardless, if you want to install an app you?e built and are too cheap to pay $99 for the official method then there is jailbreaking. This entitlement that a company should bend over to give you exactly what you want when you want without you having to think of how you can do it yourself is asinine.
Apple assumes that iPhone OS devs want to push their wares so they charge $99 for this. It?s not much but if you don?t want to pay it then jailbreak your phone to install your app. It?s not a big deal. life is full of choices that are not ideal, you just man man up and pick the best option for yourself.
Apple can do better.
Regarding the actual topic of the article, this story shows a horrible weakness in Apple's development technology. If a software object has been marked Private, there should be no way for any developer to see it, never mind use it. These analysis tools are using the Microsoft anti-virus approach: we didn't secure the system properly, so now create programs to chase all the security infringements around.
Apple can do better.
Marking private is the equivalent of sticking a "Do Not Enter" sign on a door. It doesn't stop people from entering does it?
Private as we see in languages like C# is nothing more than a compile time restriction. Even in C# you can call a private method using reflection, I have done this myself.
Private is just a concept, in reality a private method is just a public method that may or may not have compiler flags to throw errors or warnings.
In all sort of different programming languages you can see examples where they say private is just a piece of metadata, anyone can call this function regardless.
The way Objective-C works (correctly as far as object oriented programming goes) and due to the fact it is a dynamically typed language, the restrictions required for a c# style private "error" cannot work. If you look at "Object Messaging" an example would behttp://www.inf.ufsc.br/poo/smalltalk...orial/oop.html then you can see that if an unsupported method is called, it MUST be passed up the object hierarchy until it responded to. If nothing responds then a runtime error occurs.
This principle is FUNDAMENTAL to PROPER (i.e. not C++) object-oriented programming. There is no way at all that it can ever be implemented.
These analysis tools are not doing what you suggest at all. They are standard static analysis tools used (most likely clang based) to ensure safer, better quality code. This is an industry recognised practice.
Just to re-cap:
public = public
internal = public
protected = public
private = public
It's only the compiler/compiler flags that enforce errors/warnings.
Marking private is the equivalent of sticking a "Do Not Enter" sign on a door. It doesn't stop people from entering does it?
Private as we see in languages like C# is nothing more than a compile time restriction. Even in C# you can call a private method using reflection, I have done this myself.
Private is just a concept, in reality a private method is just a public method that may or may not have compiler flags to throw errors or warnings.
In all sort of different programming languages you can see examples where they say private is just a piece of metadata, anyone can call this function regardless.
The way Objective-C works (correctly as far as object oriented programming goes) and due to the fact it is a dynamically typed language, the restrictions required for a c# style private "error" cannot work. If you look at "Object Messaging" an example would behttp://www.inf.ufsc.br/poo/smalltalk...orial/oop.html then you can see that if an unsupported method is called, it MUST be passed up the object hierarchy until it responded to. If nothing responds then a runtime error occurs.
This principle is FUNDAMENTAL to PROPER (i.e. not C++) object-oriented programming. There is no way at all that it can ever be implemented.
These analysis tools are not doing what you suggest at all. They are standard static analysis tools used (most likely clang based) to ensure safer, better quality code. This is an industry recognised practice.
Just to re-cap:
public = public
internal = public
protected = public
private = public
It's only the compiler/compiler flags that enforce errors/warnings.
Java solved this by OSGi. It can be done. That is the message of the original post. Those are just technicalities.
What I do know is that the Java architecture hade security closure designed into it, simply and brillianty from day one. It was used for applets and later for the widely used j2ME platform. Java Midlets (phone apps) can safely be run no matter who wrote them or how sinister the author is because the program is run inside a virtual env, 'sandbox' and can never escape in theory. I can never access private apis, never access other Midlets data. The worst that can happen is that the program does not do what it said it should. This, of course relies on correct implementation of the sanbox, and a number of APIs which is the responsibility of the platform provider, but it is logically a truth that it is safe if these implementations are correct. Still hard to prove but a lot better that just leaving anyting open. Apple's "stone age" Obj C-language can never prove any like properties so this inferior manual/automatic process is just what to expect from at 20 year old technology without any firm base.
I do not know about dot net platform but in priciple as this is what they call "managed" the private == public must not be the truth!
What I do know is that the Java architecture hade security closure designed into it, simply and brillianty from day one. It was used for applets and later for the widely used j2ME platform. Java Midlets (phone apps) can safely be run no matter who wrote them or how sinister the author is because the program is run inside a virtual env, 'sandbox' and can never escape in theory. I can never access private apis, never access other Midlets data. The worst that can happen is that the program does not do what it said it should. This, of course relies on correct implementation of the sanbox, and a number of APIs which is the responsibility of the platform provider, but it is logically a truth that it is safe if these implementations are correct. Still hard to prove but a lot better that just leaving anyting open. Apple's "stone age" Obj C-language can never prove any like properties so this inferior manual/automatic process is just what to expect from at 20 year old technology without any firm base.
You really haven't a clue what you are talking about do you?
Reflective programming (reflection) is an extension to object-oriented programming. Java too allows the calling of private methods via reflection. A little snippet from Wikipedia:
Reflection is most commonly used in high-level virtual machine programming languages like Smalltalk and scripting languages, and less commonly used in manifestly typed and/or statically typed programming languages such as Java and C.
As for your whole sandboxing thing, what the hell has that got to do with anything? It certainly has nothing to do with public/private accessors.
A class, at its very base level, is just a grouping of related functions.
Again, the public/protected/internal/private keywords are really just compiler directives to tell the compiler to error or throw a warning should a developer try and call the code. If a method has an address (which all of them have to have) you can call it by some means, it's just a case of whether or not you can easily glean much information about the implementation (which Reflection allows).
There is nothing inferior about Objective-C, least of all because of its age, where did you get that idea from?
By that reasoning, Assembly is inferior to Java, which is obviously not the case, least of all because of performance.
And also you say all dynamically typed languages (including C) are inferior to Java, because you "can never prove any like properties". Again, utter rubbish.
Objective-C is a very elegant programming language and is closer to the idea of object-oriented programming envisioned by its creator than C++ or Java will ever be.
Java solved this by OSGi. It can be done. That is the message of the original post. Those are just technicalities.
Java still allows reflection, so my point still stands.