Why would I bring my phone to the US? It most likely won't work
Coming from the UK, not only does my phone work in the US, text messages are cheaper for me to send than back home, calls and data are 10% more expensive, and it's cheaper than most US mobiles to use. And it's good to have it, because the fraud rates on credit cards in the US are high (my card has been cloned twice out there) so my bank often flags US transactions as fraudulent and sends me an SMS to confirm them. Without my phone working, I'd have to go and look up their fraud department number and make an international call to confirm them.
My next "must" beer trip is to Ireland and go through the Guinness factory tour. Last time I was there I simply did not get time.
It's well worth doing, but ironically it's not a great place to drink Guinness. It needs to be left to settle for a little bit in the barrel before pouring and the bar on top of the museum (great views - do go there) doesn't leave it long enough. When I was there, the best place for guinness was a 'pre-club bar' on the river called the White Horse, but I think it's closed now. Ask a local for advice, it's well worth it!
Actually, the Daily WTF article is not particularly educational when it comes to EFL. It covers the obvious surface detail of what the developers do dangerously wrong. There are far worse things under the surface.
I had a chat to some of the Enlightenment devs at FOSDEM a few years ago. They were very proud of their new object system and IDL, which they thought would make it easy to bridge higher-level languages with their libraries. Unfortunately, their IDL exposed C types and nothing but C types as arguments. Their example had a char* parameter and a char* return. I asked them a few questions:
How do I know if it's and input or output (or both) parameter?
Is its length another argument (and, if so, in what units) or is it NULL-terminated?
Is there ownership transfer involved (i.e. is the caller still responsible for freeing the argument or does the callee take that responsibility? Is the caller responsible for freeing the return value and if so must they call free() or some other cleanup function)?
Is this an array of bytes or a string (i.e. should I map it to a string or data object in another language), if it's a string, what encoding does it expect and is that a global property or specified explicitly?
Apparently none of these questions had occurred to them and they didn't even understand why you'd want to know the answers to about half of them. The worst thing for me is that not only are these all important for bridging with higher-level languages, you need to know most of this information to be able to correctly use a C API, and they weren't putting it in the documentation and didn't even have consistent conventions (and therefore only need to document the exceptions). That was when I learned to avoid EFL like the plague. It may have improved since then, but I doubt it - good developers only reinvent the wheel after they've looked at existing ones and understood their flaws. The EFL developers are vaguely aware of square wheels and decided to try triangular ones as a replacement.
I didn't RTFA, but is Tizen still using Enlightenment? There's a great Daily WTF about that code which, unfortunately, doesn't scratch the surface of what it does dangerously wrong. If it does, I'd be very surprised at only 40 zero-days.
The other things were more interesting. Facebook is, apparently, not very cool (to the extent that even Uber is more cool, as are Walmart and Samsung). Facebook Messenger is even less cool than Facebook but, ironically, WhatsApp is even less cool (maybe buying it wasn't such a good plan for Facebook after all).
As for money, they're printing it. 86 billion USD revenue last year. 95% of that is from ads
And that's something that should really be concerning them. The ad market is a huge bubble - companies are increasingly seeing reduced returns from ads and most of the things that people have tried to increase this have had the opposite effect. Google has recently made the news for advertising top brands (which pay a huge amount for adverts) next to ISIS beheading videos and racist rants, which doesn't help build a positive brand image for any of these companies. It wouldn't take much for some of their biggest customers to see advertising with Google as a negative, at which point Google's revenue dries up overnight.
Unless it's changed in the last few years, that's not how Google works (or most tech companies, in fact). They know that it's vastly more expensive to hire a bad employee than to hire no one, so they will throw out candidates with a single strongly negative score. They have enough applicants that they can afford to do this.
Try booking a flight to the US and with and without a Saturday night if you don't believe me. When you get to the price breakdown, the taxes will be a lot higher for one.
So what you're saying is that PC-BSD is more Linux than Android is Linux? That's a very convoluted rationale you have. All those years of Stallman saying that you should say GNU/Linux because the core utilities, C library, and so on are at least as important as the kernel, and now people are saying that a system without those components but with Linux is not Linux.
I suspect that most relevant patents are in the mobile space. PowerVR was the first GPU to use a tile-based model and mobile GPUs use later iterations of that technology - I'd be very surprised if they didn't have patents later than 1996 on something that they've been actively working on for so long. ARM isn't on that list, but I think Mali has enough similarities to the PowerVR designs that they've almost certainly got a cross-licensing deal with them. They may well do with nVidia as well - there were a bunch of lawsuits with 3dfx and both companies that I think ended up with cross-licensing deals. That list is probably only of the licensees that are paying them money.
The main use for them is an early filter. They'll give you a rough list of hotels in a particular area, sorted by price or rating. Then you can go and look at the hotels' own web sites.
For international flights, it's often a lot cheaper to book the round trip together because the vast majority of the fare is often the taxes. If you book a round trip flight to the US that includes a Saturday night, then you pay a significantly smaller amount of tax than if you book flights that don't include staying for a Saturday night. This tax was set up to encourage people travelling on business to stay an extra night or two and spend money doing tourist things. If you book individual legs, then you'll often be taxed at the higher rate because they don't know when you're leaving. The US isn't the only country to do this, but it's the one where I've noticed it the most (it's great when you're travelling for work, because it's easy to persuade the bean counters to let you spend a couple of extra days as a tourist when it's cheaper for them to pay for a couple of nights of hotel than an earlier flight).
They will advertise cheap hotel rooms but those have the worst views of the garbage dump.
Most of the time, I don't care. Whether I'm on holiday or on business, if I'm spending a lot of time in my hotel room then I'm doing something wrong. Give me a comfortable bed, a clean and quiet room, and I'm happy. One hotel I stayed in gave you a $5 drink token for their bar for every day that you didn't use the housekeeping service: great, because I don't make the room messy enough to want someone to tidy it in a week-long stay and I don't want someone moving things in the room.
On the flip side of this is the fact that many hotels don't expect anyone to pay full price. They'll give discounts to events, people in their loyalty program, and so on. They think the sticker price is for suckers and anyone silly enough to turn up and want a room at that price should expect to be fleeced.
It often doesn't matter if there are cookies. When you visit a site, often it will place a hold on whatever you're looking for, so that if you choose to buy then it will be able to sell it to you at that price. Airlines will adjust their price based on the popularity of the flights, so if there are a load of holds placed on a particular flight, then it will be very expensive. There used to be a nice attack that would work to get cheap flights, where you'd go to a dozen or so different travel agents' sites and look at the same flight. The price would spike, because of the perceived demand. You then come back a day later and as all of them release their holds the price tanks as the system detects a sudden drop in popularity and projects that a load of existing tickets will be cancelled, so must sell more for the flight to break even.
The one doesn't preclude the other. There are a few things that mobile GPUs do to favour compute over off-chip data transfer because it saves power, but generally phone, tablet, and laptop GPUs are not that different other than in the number of pipelines that they support. That said, the numbers aren't really there for the larger parts. The iPhone and iPad between them make a sufficiently large chunk of the high-end mobile market that it's worth developing a chip that's used solely by them. The Mac lines are a sufficiently small part of their overall markets that it's difficult to compete with the economies of scale of companies like AMD and nVidia.
Once wanted to apply for a job in the US. There was an option to enter my race. (Optional, I know). That would not be allowed to be asked.
I'm not sure where you are, but in the UK it's fairly common to ask for that information on a separate form. It isn't allowed to be used in the hiring process (and there are penalties if you collect it and don't have a process in place to prevent it from being accessed by people involved with the hiring process), but is allowed to be used for statistical checks on the backgrounds of your applicants, so that you can spot things like '90% of applicants are in group X, but all hires are in group Y' and so on. I can't imagine a system not permitting this and somehow avoiding implicit bias being a huge part of the hiring process.
It's similar for the Welsh if the go to university in Wales, it's only the English or the Welsh in English or Scottish universities that have to pay. That said, the loans don't work like normal loans. The interest is at the rate of inflation and the repayments are based on your income and they're written off after a certain age, so it's effectively a graduate tax with a cap. It's not a great situation, but if your degree makes you even vaguely employable it's basically an extra percent or two on income tax for a while until it's paid off.
It is a large language, but you should re-read what he said about subsetting. 90% of programming problems can be solved with 10% of C++. The other 90% of the language is for the remaining 10%. If you learn the core of the language, it doesn't take much longer than learning the core of C, and you can then pick up the other features as you need to - many you never will unless you're implementing something like the standard library.
It's a more valid criticism of C++ to say that it is too explicit what is going on behind the scenes. For example, C++ provides virtual functions and templates as completely orthogonal features. One performs run-time specialisation, the other compile-time specialisation. You know that using one will increase dynamic instruction count, using the other will increase i-cache usage. In a language like Lisp, it's trivial to switch between the two (and you can't tell which you're using unless you read the definition of the code). In a language like Smalltalk, Java or C#, the abstract machine only provides run-time specialisation, but the implementation is free to prefer compile-time specialisation if profiling indicates that it would be faster. In C++, you make this decision early and then you pay for it if you made the wrong choice.
From a maintaining or improving existing code point of view, it is a huge clusterfuck, especially on systems that don't make ensuring slotting or abi versioning a priority (microsoft has gotten good about this since the win9x days, although they still constantly break API/ABI functionality with each new version or service pack, sometimes even in minor number releases, like the Win9x game breakage when DX8.1 came out on Windows 2000/XP...)
Odd to pick Microsoft there. Microsoft still has a policy of no stable C++ ABI across visual studio revisions, unlike the *NIX world that standardised on the Itanium ABI around 15 years ago. In contrast, Microsoft's policy is to tell you to use C or COM across library boundaries.
You're also conflating three issues:
Language ABI.
Standard library ABI.
Third-party library ABIs.
Outside of Windows, the first has been a solved problem for well over a decade. The second was a little bit problematic in the move to C++11 because that change fixed a lot of fundamental problems with the language, but C++14 and C++17 have both been explicitly designed to permit a stable standard library ABI and libc++ supports a single ABI for all versions of the language. Neither C nor C++ is particularly good for third-party ABIs, unless you put a lot of care into the design. C++ at least makes it possible to use namespaces for versioning and so you can add compat functions in the language and have old code fall back to using them while new code uses the newer versions: C requires you to use symbol versioning in the linker to do the same thing.
But system level software and operating system kernels are DECIDEDLY NOT the place for it, unless a *LOT* of features are pared back and warnings increased for documenting how and where API/ABI changes happen between versions of code/compiled libraries and binaries.
The microkernel that runs the secure element in an iPhone and the baseband in any phone is written in C++. Windows and macOS device drivers are written in C++. Google's new kernel is written in C++. Symbian (which held 80% of the smartphone market until Nokia mismanaged it into oblivion) was written in C++. About the only operating systems that prefer C to C++ are ones that predate C++.
Slashdot Mirror
Why would I bring my phone to the US? It most likely won't work
Coming from the UK, not only does my phone work in the US, text messages are cheaper for me to send than back home, calls and data are 10% more expensive, and it's cheaper than most US mobiles to use. And it's good to have it, because the fraud rates on credit cards in the US are high (my card has been cloned twice out there) so my bank often flags US transactions as fraudulent and sends me an SMS to confirm them. Without my phone working, I'd have to go and look up their fraud department number and make an international call to confirm them.
My next "must" beer trip is to Ireland and go through the Guinness factory tour. Last time I was there I simply did not get time.
It's well worth doing, but ironically it's not a great place to drink Guinness. It needs to be left to settle for a little bit in the barrel before pouring and the bar on top of the museum (great views - do go there) doesn't leave it long enough. When I was there, the best place for guinness was a 'pre-club bar' on the river called the White Horse, but I think it's closed now. Ask a local for advice, it's well worth it!
I had a chat to some of the Enlightenment devs at FOSDEM a few years ago. They were very proud of their new object system and IDL, which they thought would make it easy to bridge higher-level languages with their libraries. Unfortunately, their IDL exposed C types and nothing but C types as arguments. Their example had a char* parameter and a char* return. I asked them a few questions:
How do I know if it's and input or output (or both) parameter?
Is its length another argument (and, if so, in what units) or is it NULL-terminated?
Is there ownership transfer involved (i.e. is the caller still responsible for freeing the argument or does the callee take that responsibility? Is the caller responsible for freeing the return value and if so must they call free() or some other cleanup function)?
Is this an array of bytes or a string (i.e. should I map it to a string or data object in another language), if it's a string, what encoding does it expect and is that a global property or specified explicitly?
Apparently none of these questions had occurred to them and they didn't even understand why you'd want to know the answers to about half of them. The worst thing for me is that not only are these all important for bridging with higher-level languages, you need to know most of this information to be able to correctly use a C API, and they weren't putting it in the documentation and didn't even have consistent conventions (and therefore only need to document the exceptions). That was when I learned to avoid EFL like the plague. It may have improved since then, but I doubt it - good developers only reinvent the wheel after they've looked at existing ones and understood their flaws. The EFL developers are vaguely aware of square wheels and decided to try triangular ones as a replacement.
I didn't RTFA, but is Tizen still using Enlightenment? There's a great Daily WTF about that code which, unfortunately, doesn't scratch the surface of what it does dangerously wrong. If it does, I'd be very surprised at only 40 zero-days.
The other things were more interesting. Facebook is, apparently, not very cool (to the extent that even Uber is more cool, as are Walmart and Samsung). Facebook Messenger is even less cool than Facebook but, ironically, WhatsApp is even less cool (maybe buying it wasn't such a good plan for Facebook after all).
As for money, they're printing it. 86 billion USD revenue last year. 95% of that is from ads
And that's something that should really be concerning them. The ad market is a huge bubble - companies are increasingly seeing reduced returns from ads and most of the things that people have tried to increase this have had the opposite effect. Google has recently made the news for advertising top brands (which pay a huge amount for adverts) next to ISIS beheading videos and racist rants, which doesn't help build a positive brand image for any of these companies. It wouldn't take much for some of their biggest customers to see advertising with Google as a negative, at which point Google's revenue dries up overnight.
Unless it's changed in the last few years, that's not how Google works (or most tech companies, in fact). They know that it's vastly more expensive to hire a bad employee than to hire no one, so they will throw out candidates with a single strongly negative score. They have enough applicants that they can afford to do this.
Try booking a flight to the US and with and without a Saturday night if you don't believe me. When you get to the price breakdown, the taxes will be a lot higher for one.
Is app short for Advert Presentation Program?
So what you're saying is that PC-BSD is more Linux than Android is Linux? That's a very convoluted rationale you have. All those years of Stallman saying that you should say GNU/Linux because the core utilities, C library, and so on are at least as important as the kernel, and now people are saying that a system without those components but with Linux is not Linux.
I suspect that most relevant patents are in the mobile space. PowerVR was the first GPU to use a tile-based model and mobile GPUs use later iterations of that technology - I'd be very surprised if they didn't have patents later than 1996 on something that they've been actively working on for so long. ARM isn't on that list, but I think Mali has enough similarities to the PowerVR designs that they've almost certainly got a cross-licensing deal with them. They may well do with nVidia as well - there were a bunch of lawsuits with 3dfx and both companies that I think ended up with cross-licensing deals. That list is probably only of the licensees that are paying them money.
The main use for them is an early filter. They'll give you a rough list of hotels in a particular area, sorted by price or rating. Then you can go and look at the hotels' own web sites.
For international flights, it's often a lot cheaper to book the round trip together because the vast majority of the fare is often the taxes. If you book a round trip flight to the US that includes a Saturday night, then you pay a significantly smaller amount of tax than if you book flights that don't include staying for a Saturday night. This tax was set up to encourage people travelling on business to stay an extra night or two and spend money doing tourist things. If you book individual legs, then you'll often be taxed at the higher rate because they don't know when you're leaving. The US isn't the only country to do this, but it's the one where I've noticed it the most (it's great when you're travelling for work, because it's easy to persuade the bean counters to let you spend a couple of extra days as a tourist when it's cheaper for them to pay for a couple of nights of hotel than an earlier flight).
They will advertise cheap hotel rooms but those have the worst views of the garbage dump.
Most of the time, I don't care. Whether I'm on holiday or on business, if I'm spending a lot of time in my hotel room then I'm doing something wrong. Give me a comfortable bed, a clean and quiet room, and I'm happy. One hotel I stayed in gave you a $5 drink token for their bar for every day that you didn't use the housekeeping service: great, because I don't make the room messy enough to want someone to tidy it in a week-long stay and I don't want someone moving things in the room.
On the flip side of this is the fact that many hotels don't expect anyone to pay full price. They'll give discounts to events, people in their loyalty program, and so on. They think the sticker price is for suckers and anyone silly enough to turn up and want a room at that price should expect to be fleeced.
It often doesn't matter if there are cookies. When you visit a site, often it will place a hold on whatever you're looking for, so that if you choose to buy then it will be able to sell it to you at that price. Airlines will adjust their price based on the popularity of the flights, so if there are a load of holds placed on a particular flight, then it will be very expensive. There used to be a nice attack that would work to get cheap flights, where you'd go to a dozen or so different travel agents' sites and look at the same flight. The price would spike, because of the perceived demand. You then come back a day later and as all of them release their holds the price tanks as the system detects a sudden drop in popularity and projects that a load of existing tickets will be cancelled, so must sell more for the flight to break even.
The one doesn't preclude the other. There are a few things that mobile GPUs do to favour compute over off-chip data transfer because it saves power, but generally phone, tablet, and laptop GPUs are not that different other than in the number of pipelines that they support. That said, the numbers aren't really there for the larger parts. The iPhone and iPad between them make a sufficiently large chunk of the high-end mobile market that it's worth developing a chip that's used solely by them. The Mac lines are a sufficiently small part of their overall markets that it's difficult to compete with the economies of scale of companies like AMD and nVidia.
If ever a post deserved a +5 Troll, it's this one.
Once wanted to apply for a job in the US. There was an option to enter my race. (Optional, I know). That would not be allowed to be asked.
I'm not sure where you are, but in the UK it's fairly common to ask for that information on a separate form. It isn't allowed to be used in the hiring process (and there are penalties if you collect it and don't have a process in place to prevent it from being accessed by people involved with the hiring process), but is allowed to be used for statistical checks on the backgrounds of your applicants, so that you can spot things like '90% of applicants are in group X, but all hires are in group Y' and so on. I can't imagine a system not permitting this and somehow avoiding implicit bias being a huge part of the hiring process.
It's similar for the Welsh if the go to university in Wales, it's only the English or the Welsh in English or Scottish universities that have to pay. That said, the loans don't work like normal loans. The interest is at the rate of inflation and the repayments are based on your income and they're written off after a certain age, so it's effectively a graduate tax with a cap. It's not a great situation, but if your degree makes you even vaguely employable it's basically an extra percent or two on income tax for a while until it's paid off.
Strongly Christian areas also under-report suicides, to avoid the stigma, so you'll get a lot of 'accidents'.
and a many thousand dollar rainy day fund in case I stub my toe and need to go to a doctor.
The rest of the stuff on your list is universal, but that one is pretty much a US-only thing.
It is a large language, but you should re-read what he said about subsetting. 90% of programming problems can be solved with 10% of C++. The other 90% of the language is for the remaining 10%. If you learn the core of the language, it doesn't take much longer than learning the core of C, and you can then pick up the other features as you need to - many you never will unless you're implementing something like the standard library.
It's a more valid criticism of C++ to say that it is too explicit what is going on behind the scenes. For example, C++ provides virtual functions and templates as completely orthogonal features. One performs run-time specialisation, the other compile-time specialisation. You know that using one will increase dynamic instruction count, using the other will increase i-cache usage. In a language like Lisp, it's trivial to switch between the two (and you can't tell which you're using unless you read the definition of the code). In a language like Smalltalk, Java or C#, the abstract machine only provides run-time specialisation, but the implementation is free to prefer compile-time specialisation if profiling indicates that it would be faster. In C++, you make this decision early and then you pay for it if you made the wrong choice.
From a maintaining or improving existing code point of view, it is a huge clusterfuck, especially on systems that don't make ensuring slotting or abi versioning a priority (microsoft has gotten good about this since the win9x days, although they still constantly break API/ABI functionality with each new version or service pack, sometimes even in minor number releases, like the Win9x game breakage when DX8.1 came out on Windows 2000/XP...)
Odd to pick Microsoft there. Microsoft still has a policy of no stable C++ ABI across visual studio revisions, unlike the *NIX world that standardised on the Itanium ABI around 15 years ago. In contrast, Microsoft's policy is to tell you to use C or COM across library boundaries.
You're also conflating three issues:
Outside of Windows, the first has been a solved problem for well over a decade. The second was a little bit problematic in the move to C++11 because that change fixed a lot of fundamental problems with the language, but C++14 and C++17 have both been explicitly designed to permit a stable standard library ABI and libc++ supports a single ABI for all versions of the language. Neither C nor C++ is particularly good for third-party ABIs, unless you put a lot of care into the design. C++ at least makes it possible to use namespaces for versioning and so you can add compat functions in the language and have old code fall back to using them while new code uses the newer versions: C requires you to use symbol versioning in the linker to do the same thing.
But system level software and operating system kernels are DECIDEDLY NOT the place for it, unless a *LOT* of features are pared back and warnings increased for documenting how and where API/ABI changes happen between versions of code/compiled libraries and binaries.
The microkernel that runs the secure element in an iPhone and the baseband in any phone is written in C++. Windows and macOS device drivers are written in C++. Google's new kernel is written in C++. Symbian (which held 80% of the smartphone market until Nokia mismanaged it into oblivion) was written in C++. About the only operating systems that prefer C to C++ are ones that predate C++.