No third-party apps, lots of Apple-provided apps. It was pretty obvious from the start that this was just an excuse because they hadn't got the dev tools and signing infrastructure working properly: there was a weird disconnect between stuff that Apple-provided apps could do and stuff that third-party apps could do.
If it launched the other app correctly, then there will be a back thing, including the name of the app that you'll go back to, at the top left of the screen.
Option 1: Go to the Finder, hit the 'Applications' icon in the side bar, double-click on Safari (this is how you've launched applications on every Mac since 1984 - how many other operating systems can say that for any GUI?).
Option 2: Use spotlight by either hitting the shortcut keys (command-space) or by clicking on the search icon at the top right of the screen. Type 'Safari' (it will probably autocomplete after 'Sa'). Hit enter. This mechanism is relatively new and has only been the same for about 10 years.
I've filed a bug report, provided a trace, sent them mechanisms for reproducing, and pointed to the pile of Apple user forum posts of people with the same issue. No fix yet.
Yes and no. Those._foobar files don't appear on HFS+ either, they're only there for things like NFS or SMB shares or FAT filesystems that don't have the ability to store some of the metadata that Mac apps expect to work. The VFS layer transparently maps the metadata to and from these dot files when using a filesystem without the relevant metadata support. There are basically three solutions to this problem: silently lose the extra metadata (probably a bad idea), report an error to the program (which probably doesn't have handling for it) or store it somewhere else (which is ugly, but at least something that you can hide in the GUI). The ideal solution is for every (local and network) filesystem to support storing arbitrary metadata, but I don't know how we get from here to there.
My problem is that I can't help but assume that the Danish screamo band gets 2/3rds of a cut, and Apple gets the other third
If that were true, then this would a a much better deal for the artists than pretty much any distribution mechanism other than CDBaby. I suspect that Apple's cut is actually a lot higher.
What do they have? The only thing that had in the H.264 patent pool was a token patent relating to the QuickTime container format (which is why that became the standard container format for MPEG-4). That's probably expiring soon, if it hasn't already, and I haven't seen much CODEC R&D from Apple in the last decade.
They've already deployed APFS for all iOS devices. These are a nice place for FS testing, because they all tend to have regular backups and they have a debug interface that connects to Apple-controlled software for collecting problem reports. On top of that, the macOS beta testers probably include a lot of people who do weird things with their FS. I still wouldn't entirely trust it on release (the general rule of thumb is that it takes 10 years for a new FS to become stable), but it's probably had a better stress test than any other new FS. I'd be interested to know if more data is stored on APFS or ZFS at this point.
I've only noticed one difference since the 'upgrade'. The 'mark as junk' button in Mail.app has been replaced with a 'move to junk folder' button, which completely breaks if you have more than one email account, because you can no longer have a single spam folder, but have to instead use the junk folder on each mail server. It also breaks the custom rules that you can configure to run when an email is marked as junk. It provides no benefit at all (previously, hitting the 'mark as junk' button would move it to the junk folder if that was your default action for emails marked as junk) and breaks things. Apparently the Mail.app team is a deeply unhappy place. They seem to be taking every crappy misfeature from the awful iOS mail app and porting them, rather than actually improving anything.
What do you mean 'still'? Upgrading to 10.4 if you had enabled File Vault (home directory was an encrypted disk image, mounted on login) worked fine, right up until the first reboot (typically a month later) when your home directory became unreadable (eventually I discovered that this was only true for 10.4 - 10.3 could read it fine). No OS X / macOS has gone as badly since then - apparently their QA department didn't test the configuration of people upgrading from 10.3 and using one of the flagship features of 10.3.
It's a translation problem. In English, a liberal is someone who subscribes to Enlightenment ideas relating to equality and freedom. In American, it means 'someone I don't like' (confusingly for English speakers, American grammar, like Japanese, has different forms depending on the speaker, and the word 'conservative' has the same meaning when spoken by a different set of people).
It's not just the Japanese, it's basically every other country. When I first played SimCity, I found it horribly unrealistic: how on Earth could you build a city that segregated places people live, places people work, and places people go for recreation and required you to drive between them? Then I visited the USA and discovered that real city planners were doing this. Outside of the densest areas, US cities seem to combine the worst aspects of ubran and rural living: high population density and long driving distances to everywhere you'd want to go to regularly.
Which is why I find it odd that instead of trying to introduce backdoors, governments don't just try to keep encryption to a level where only nation states have the resources to crack it
Because that's impossible, for very many reasons.
First, the nation state doesn't need to break just one use of encryption, it needs to break several. If you need to be able to break 1,000 people's encrypted communications, then the criminal needs 1,000th of the computational power that the state can throw at it. If the state needs to be able to break them in a day, then a criminal happy to wait a month needs 1/30,000th the computational power, which is about the difference between a data centre and a mobile phone.
Second, because breaking encryption is an embarrassingly parallel workload and so is ideal for botnets of GPUs. Guess who controls those? It's not the governments.
Third, because encryption is time sensitive. For some things, you care that no one can decrypt or spoof it this week (e.g. 'fire missiles at North Korea now' or 'buy stocks in Whole Foods now, I just learned that Amazon is going to try to buy them'). For some things, you care about whether someone can decrypt them in the next 20 years or longer. These things factor in the growth of computational power in the work factor for the attackers (GPUs and cheap FPGAs caused problems for some users of these) and so expect not to be broken by your non-government attacker with 20 years of growth in available processing power. Given the polynomial growth rate of cheap computational power, there's absolutely no way that you could expect a government to have now, more computational power than a gang of criminals could have in 20 years. To put this in perspective, 20 years ago, a 300MHz Pentium II was a really expensive CPU and the world's fastest supercomputer had a peak of 1,453.0GFLOPS, using 7,264 cores, i.e. about 6 times as fast as the iPhone 7's GPU, or about a quarter the speed of an nVidia GPGPU card. For encryption, integer ops matter more than floating point, but the trajectories are in the same direction.
Everyone EXCEPT Remoaners realises that the UK can issues the same grants without the expensive middle-man taking his cut.
Take a look at your tax statement this year (you do pay tax in the UK, right?). It will have a breakdown of the amount that you're paying into the EU. Notice how tiny a percentage of the total tax revenue that is. Notice how it's far less than you're paying in council tax (assuming that you're not a student or otherwise exempt from council tax). Now, compare the proportion of that money that is spent improving the quality of poorer regions in the UK and investing in UK infrastructure than the proportion of the remainder. Now tell me who you'd rather have spending your taxes.
Or, if you don't want to do this on a purely financial basis, compare the EU data protection office with the actions of the UK's regulator ('oh, you just gave loads of medical records to Google / Deep Mind without consent of the patients? I'm sure that's fine') and tell me which you'd prefer having control over privacy. Or compare Theresa 'Encryption bad, must backdoor everything' May's attitude with TFA and tell me who you'd prefer.
The problem with the leave arguments is that almost all of the negative things about the EU (concentrating power in the Commission rather than the Parliament, pushing pro-corporation trade treaties, and so on) were pushed hard by the UK government's representatives in the Council of Ministers and over the objection of other EU countries. You don't like these things, so you'd rather give more power to the people responsible for them.
Why do pro-European advocates always talk about the amount of money Regions in the UK receive from the European bodies ? Heads-up : It's the UK's money so start with. We are a net contributor to the EU so we pay in more than comes out.
So you'd rather trust the London government to invest that money in the poorer regions? The reason that people bring it up is that, between Brussels and London, Brussels has a far better track record of investing in the poorer areas of the UK.
There's also suddenly a lot less public appetite for throwing away a load of regulations after a tower block burned down killing a load of people because the UK allows the use of an insulating material that Germany and the US have banned as having a too-high fire risk.
Amusing fact if you were canvasing for Labour: of all of the key policies in their manifesto, leaving the EU had the lowest approval rating (though apparently there's a mandate for that, but not for properly funding the NHS or nationalising the railways, for example).
The best priced smartphone I've had was a Moto G. Incredible battery life, no bloat, almost vanilla Android. Roughly $180 - $200.
And still getting regular updates from LineageOS, with the latest Android version. The only reason I'm considering upgrading mine is that the map data in OSMAnd~ is a lot more detailed than it was a few years ago and there is now quite a noticeable lag when rendering it. For everything else, it's perfectly adequate.
This is true for first-party updates, but I'm still getting regular updates on my Moto G (released November 2013) from LineageOS (thanks to the Slashdot AC who pointed me to the process for installing it).
You can even modify the architecture enough that it is no longer compatible with other ARM based products if you choose
No you can't. ARM will not sell a license that permits you to change the ISA for any price: they saw what happened with fragmentation with the MIPS market and have no desire to commit suicide in the same way. You can add custom interrupt controllers, UARTs and DMA engines, which makes porting an OS a bit harder but is invisible to userspace (and to the compiler). You can also add custom coprocessors, which every SoC vendor does (for example, a lot have face detection as a dedicated logic block: you DMA an image from the camera to the block and then read back a list of rectangles from an I/O register). Apple's chips also have a smaller ARM core that has private memory that is not readable by the main core for storing encryption keys, so that a compromised device can be forced to encrypt or decrypt things for the attacker, but can't leak the keys.
You've obviously not seen a lot of DARPA projects then. How about Clean-Slate Design of Resilient, Adaptive, Secure Hosts (CRASH), which was funded by DARPA's I2O (information innovation office - no, I'm not making this up).
The last 10 year of ISA slots on motherboards were connected to the CPU via a PCI to ISA bridge. Those same chips are now being used on ISA expansion cards. If your hardware worked with a Pentium's ISA slot, it's likely to work with one of these. USB serial devices worked fine for things that weren't timing sensitive, but the lack of isochronous transfer in USB 1.1 really screwed up things that were. Again, if you really want a serial port and you don't have to use a laptop, you can still buy 16550 UARTs on PCI cards and they'll work with anything that worked with a Pentium's serial port because they're connected in exactly the same way (well, unless you had a Pentium that put the UART on the other side of the PCI to ISA bridge).
Going the alternative firmware way is simply not an option for non-techs.
Switching to alternative firmware isn't. The upgrade step is not that complicated, but it is pretty scary (lots of 'this may destroy your phone' steps). Once it's done; however, the phone auto-updates each week with minimal intervention and, aside from not being filled with crapware (at least, until the user installs some from the Google store), is pretty much indistinguishable from one that's vendor supplied. I'd have no problem giving a phone installed with LineageOS to my mother to use, though there's absolutely no way I'd be able to persuade her to do the install herself (she'd probably manage it fine following the instructions, but persuading her to actually do the steps that tell her they may brick her phone would be pretty hard).
Open a terminal. Run otool -L on one of those binaries (you'll find it in the Contents/MacOS directory inside the bundle). Look at all of the libraries in/Library or/System/Library that they link to. Now look at the file sizes of those libraries. Now look at the Resources folder for each of the.framework bundles that contains those libraries. Now imagine if each of those disk images had contained a copy of those and if each running process had contained a separate copy of them, rather than a CoW shared version.
You can still buy PCI to ISA bridges, which will show up as ISA busses by anything on DOS once you configure them and FreeDOS still allows things that want to communicate with COM and LPT ports to do so directly.
When Facebook bought WhatsApp, it was free for the first year (long enough for network effects to lock users in) and then $1/year after that. $1/year is small enough that you can pay it without thinking, but the cost of hosting is tiny. I don't know exactly how many they users were hosting per machine (FreeBSD + Erlang is a pretty efficient combination), but I do remember them having over a million open sockets on a single node. If each one of those is to a separate user, then that's $1m/year/server, which pays for quite a lot of VM time.
Slashdot Mirror
No third-party apps, lots of Apple-provided apps. It was pretty obvious from the start that this was just an excuse because they hadn't got the dev tools and signing infrastructure working properly: there was a weird disconnect between stuff that Apple-provided apps could do and stuff that third-party apps could do.
If it launched the other app correctly, then there will be a back thing, including the name of the app that you'll go back to, at the top left of the screen.
How the FUCK do I launch Safari?
Option 1: Go to the Finder, hit the 'Applications' icon in the side bar, double-click on Safari (this is how you've launched applications on every Mac since 1984 - how many other operating systems can say that for any GUI?).
Option 2: Use spotlight by either hitting the shortcut keys (command-space) or by clicking on the search icon at the top right of the screen. Type 'Safari' (it will probably autocomplete after 'Sa'). Hit enter. This mechanism is relatively new and has only been the same for about 10 years.
I've filed a bug report, provided a trace, sent them mechanisms for reproducing, and pointed to the pile of Apple user forum posts of people with the same issue. No fix yet.
Yes and no. Those ._foobar files don't appear on HFS+ either, they're only there for things like NFS or SMB shares or FAT filesystems that don't have the ability to store some of the metadata that Mac apps expect to work. The VFS layer transparently maps the metadata to and from these dot files when using a filesystem without the relevant metadata support. There are basically three solutions to this problem: silently lose the extra metadata (probably a bad idea), report an error to the program (which probably doesn't have handling for it) or store it somewhere else (which is ugly, but at least something that you can hide in the GUI). The ideal solution is for every (local and network) filesystem to support storing arbitrary metadata, but I don't know how we get from here to there.
My problem is that I can't help but assume that the Danish screamo band gets 2/3rds of a cut, and Apple gets the other third
If that were true, then this would a a much better deal for the artists than pretty much any distribution mechanism other than CDBaby. I suspect that Apple's cut is actually a lot higher.
Apple owns some essential HEVC patents
What do they have? The only thing that had in the H.264 patent pool was a token patent relating to the QuickTime container format (which is why that became the standard container format for MPEG-4). That's probably expiring soon, if it hasn't already, and I haven't seen much CODEC R&D from Apple in the last decade.
They've already deployed APFS for all iOS devices. These are a nice place for FS testing, because they all tend to have regular backups and they have a debug interface that connects to Apple-controlled software for collecting problem reports. On top of that, the macOS beta testers probably include a lot of people who do weird things with their FS. I still wouldn't entirely trust it on release (the general rule of thumb is that it takes 10 years for a new FS to become stable), but it's probably had a better stress test than any other new FS. I'd be interested to know if more data is stored on APFS or ZFS at this point.
I've only noticed one difference since the 'upgrade'. The 'mark as junk' button in Mail.app has been replaced with a 'move to junk folder' button, which completely breaks if you have more than one email account, because you can no longer have a single spam folder, but have to instead use the junk folder on each mail server. It also breaks the custom rules that you can configure to run when an email is marked as junk. It provides no benefit at all (previously, hitting the 'mark as junk' button would move it to the junk folder if that was your default action for emails marked as junk) and breaks things. Apparently the Mail.app team is a deeply unhappy place. They seem to be taking every crappy misfeature from the awful iOS mail app and porting them, rather than actually improving anything.
What do you mean 'still'? Upgrading to 10.4 if you had enabled File Vault (home directory was an encrypted disk image, mounted on login) worked fine, right up until the first reboot (typically a month later) when your home directory became unreadable (eventually I discovered that this was only true for 10.4 - 10.3 could read it fine). No OS X / macOS has gone as badly since then - apparently their QA department didn't test the configuration of people upgrading from 10.3 and using one of the flagship features of 10.3.
It's a translation problem. In English, a liberal is someone who subscribes to Enlightenment ideas relating to equality and freedom. In American, it means 'someone I don't like' (confusingly for English speakers, American grammar, like Japanese, has different forms depending on the speaker, and the word 'conservative' has the same meaning when spoken by a different set of people).
It's not just the Japanese, it's basically every other country. When I first played SimCity, I found it horribly unrealistic: how on Earth could you build a city that segregated places people live, places people work, and places people go for recreation and required you to drive between them? Then I visited the USA and discovered that real city planners were doing this. Outside of the densest areas, US cities seem to combine the worst aspects of ubran and rural living: high population density and long driving distances to everywhere you'd want to go to regularly.
Which is why I find it odd that instead of trying to introduce backdoors, governments don't just try to keep encryption to a level where only nation states have the resources to crack it
Because that's impossible, for very many reasons.
First, the nation state doesn't need to break just one use of encryption, it needs to break several. If you need to be able to break 1,000 people's encrypted communications, then the criminal needs 1,000th of the computational power that the state can throw at it. If the state needs to be able to break them in a day, then a criminal happy to wait a month needs 1/30,000th the computational power, which is about the difference between a data centre and a mobile phone.
Second, because breaking encryption is an embarrassingly parallel workload and so is ideal for botnets of GPUs. Guess who controls those? It's not the governments.
Third, because encryption is time sensitive. For some things, you care that no one can decrypt or spoof it this week (e.g. 'fire missiles at North Korea now' or 'buy stocks in Whole Foods now, I just learned that Amazon is going to try to buy them'). For some things, you care about whether someone can decrypt them in the next 20 years or longer. These things factor in the growth of computational power in the work factor for the attackers (GPUs and cheap FPGAs caused problems for some users of these) and so expect not to be broken by your non-government attacker with 20 years of growth in available processing power. Given the polynomial growth rate of cheap computational power, there's absolutely no way that you could expect a government to have now, more computational power than a gang of criminals could have in 20 years. To put this in perspective, 20 years ago, a 300MHz Pentium II was a really expensive CPU and the world's fastest supercomputer had a peak of 1,453.0GFLOPS, using 7,264 cores, i.e. about 6 times as fast as the iPhone 7's GPU, or about a quarter the speed of an nVidia GPGPU card. For encryption, integer ops matter more than floating point, but the trajectories are in the same direction.
Everyone EXCEPT Remoaners realises that the UK can issues the same grants without the expensive middle-man taking his cut.
Take a look at your tax statement this year (you do pay tax in the UK, right?). It will have a breakdown of the amount that you're paying into the EU. Notice how tiny a percentage of the total tax revenue that is. Notice how it's far less than you're paying in council tax (assuming that you're not a student or otherwise exempt from council tax). Now, compare the proportion of that money that is spent improving the quality of poorer regions in the UK and investing in UK infrastructure than the proportion of the remainder. Now tell me who you'd rather have spending your taxes.
Or, if you don't want to do this on a purely financial basis, compare the EU data protection office with the actions of the UK's regulator ('oh, you just gave loads of medical records to Google / Deep Mind without consent of the patients? I'm sure that's fine') and tell me which you'd prefer having control over privacy. Or compare Theresa 'Encryption bad, must backdoor everything' May's attitude with TFA and tell me who you'd prefer.
The problem with the leave arguments is that almost all of the negative things about the EU (concentrating power in the Commission rather than the Parliament, pushing pro-corporation trade treaties, and so on) were pushed hard by the UK government's representatives in the Council of Ministers and over the objection of other EU countries. You don't like these things, so you'd rather give more power to the people responsible for them.
Why do pro-European advocates always talk about the amount of money Regions in the UK receive from the European bodies ? Heads-up : It's the UK's money so start with. We are a net contributor to the EU so we pay in more than comes out.
So you'd rather trust the London government to invest that money in the poorer regions? The reason that people bring it up is that, between Brussels and London, Brussels has a far better track record of investing in the poorer areas of the UK.
There's also suddenly a lot less public appetite for throwing away a load of regulations after a tower block burned down killing a load of people because the UK allows the use of an insulating material that Germany and the US have banned as having a too-high fire risk.
Amusing fact if you were canvasing for Labour: of all of the key policies in their manifesto, leaving the EU had the lowest approval rating (though apparently there's a mandate for that, but not for properly funding the NHS or nationalising the railways, for example).
The best priced smartphone I've had was a Moto G. Incredible battery life, no bloat, almost vanilla Android. Roughly $180 - $200.
And still getting regular updates from LineageOS, with the latest Android version. The only reason I'm considering upgrading mine is that the map data in OSMAnd~ is a lot more detailed than it was a few years ago and there is now quite a noticeable lag when rendering it. For everything else, it's perfectly adequate.
This is true for first-party updates, but I'm still getting regular updates on my Moto G (released November 2013) from LineageOS (thanks to the Slashdot AC who pointed me to the process for installing it).
You can even modify the architecture enough that it is no longer compatible with other ARM based products if you choose
No you can't. ARM will not sell a license that permits you to change the ISA for any price: they saw what happened with fragmentation with the MIPS market and have no desire to commit suicide in the same way. You can add custom interrupt controllers, UARTs and DMA engines, which makes porting an OS a bit harder but is invisible to userspace (and to the compiler). You can also add custom coprocessors, which every SoC vendor does (for example, a lot have face detection as a dedicated logic block: you DMA an image from the camera to the block and then read back a list of rectangles from an I/O register). Apple's chips also have a smaller ARM core that has private memory that is not readable by the main core for storing encryption keys, so that a compromised device can be forced to encrypt or decrypt things for the attacker, but can't leak the keys.
You've obviously not seen a lot of DARPA projects then. How about Clean-Slate Design of Resilient, Adaptive, Secure Hosts (CRASH), which was funded by DARPA's I2O (information innovation office - no, I'm not making this up).
The last 10 year of ISA slots on motherboards were connected to the CPU via a PCI to ISA bridge. Those same chips are now being used on ISA expansion cards. If your hardware worked with a Pentium's ISA slot, it's likely to work with one of these. USB serial devices worked fine for things that weren't timing sensitive, but the lack of isochronous transfer in USB 1.1 really screwed up things that were. Again, if you really want a serial port and you don't have to use a laptop, you can still buy 16550 UARTs on PCI cards and they'll work with anything that worked with a Pentium's serial port because they're connected in exactly the same way (well, unless you had a Pentium that put the UART on the other side of the PCI to ISA bridge).
Going the alternative firmware way is simply not an option for non-techs.
Switching to alternative firmware isn't. The upgrade step is not that complicated, but it is pretty scary (lots of 'this may destroy your phone' steps). Once it's done; however, the phone auto-updates each week with minimal intervention and, aside from not being filled with crapware (at least, until the user installs some from the Google store), is pretty much indistinguishable from one that's vendor supplied. I'd have no problem giving a phone installed with LineageOS to my mother to use, though there's absolutely no way I'd be able to persuade her to do the install herself (she'd probably manage it fine following the instructions, but persuading her to actually do the steps that tell her they may brick her phone would be pretty hard).
Open a terminal. Run otool -L on one of those binaries (you'll find it in the Contents/MacOS directory inside the bundle). Look at all of the libraries in /Library or /System/Library that they link to. Now look at the file sizes of those libraries. Now look at the Resources folder for each of the .framework bundles that contains those libraries. Now imagine if each of those disk images had contained a copy of those and if each running process had contained a separate copy of them, rather than a CoW shared version.
You can still buy PCI to ISA bridges, which will show up as ISA busses by anything on DOS once you configure them and FreeDOS still allows things that want to communicate with COM and LPT ports to do so directly.
When Facebook bought WhatsApp, it was free for the first year (long enough for network effects to lock users in) and then $1/year after that. $1/year is small enough that you can pay it without thinking, but the cost of hosting is tiny. I don't know exactly how many they users were hosting per machine (FreeBSD + Erlang is a pretty efficient combination), but I do remember them having over a million open sockets on a single node. If each one of those is to a separate user, then that's $1m/year/server, which pays for quite a lot of VM time.