It's not because they forget to fill up, it's that they did fill up, but attempt to stretch it. All it takes is a sudden traffic jam, a road closure, diversion, or something else that idles them by the driveway and boom, they're out of gas.
And the first time somebody gets stuck on the side of the road for lack of wiper fluid, they'll learn to buy a $5 bottle and keep it at home. We're not talking about something you have to fill up twice a week here....
They can develop warnings all they want, but unless the sensors can operate when they are dirty, they aren't going to be much use in the winter in New England or Eastern Canada.
Obviously, the sensors have to operate when not entirely clean, but there are limits, and there's no way to avoid requiring that they be at least somewhat clean.
The hot new feature is the dual aperture camera for better low light photos.
Half a stop better. IMO, it's still about 5 1/2 stops too slow for low-light photography, thanks to the crop factor/tiny sensor, but I guess every little bit helps.
I'm kind of surprised that there's a noticeable improvement in depth of field between f/1.5 and f/2.4 on a sensor that small. I'd have expected diffraction limiting to make such dual-aperture designs not worth bothering with. But maybe not.
As long as there's plastic lenses in use, there's not really much manufacturers can do about this, other than have a secondary, cheap, external covering that can be unclipped and swapped out quickly.
Why the heck would someone put a plastic lens on a sensor intended for use in a high-dirt environment? Even glass isn't hard enough. They should be using sapphire crystal, at an absolute minimum. That's a pretty big design mistake.
That's easily solved. Most people never run out of fuel, because there's a gauge to tell them that they need to fill it up. If cars had an idiot light that says "Refill wiper fluid now" that goes on when it is at the 25% mark, nobody would run out of wiper fluid, either. Car companies don't add that sensor that because wiper fluid isn't a critical safety feature. If it becomes one, they presumably will do so.
They absolutely can prevent transfer of a license. The license is obtained by entering the code into a website, and that code is usable only once. That's not where Disney's argument falls apart from a legal perspective.
The problem with Disney's argument is that their reasoning, when applied to slightly different situations, results in a legal interpretation that fails the common sense test. Consider three scenarios:
Case 1: You buy a DVD for $10. You use the code, and keep it, because it is nontransferable. However, you sell me the DVD for $5 because you don't need it, according to your right of first sale.
Case 2: I buy a DVD for $10. I sell you the code from that DVD for $5. The company claims that I'm not allowed to sell the code to someone who is not the original buyer.
Case 3: I buy a DVD for $10 and sell it unopened to you for $10, which is my right (first sale again). You open it, use the code, and sell back the disc to me for $5.
In all three cases, the result is the same: I have a DVD and you have an electronic copy, and we each spent $5.
Nothing in Title 17 could plausibly explain why only one of these three transactions is legal, given that the end result and process are effectively identical except for trivial accounting differences. Such a requirement completely fails to stand up to the slightest bit of logical scrutiny, and any legal code that would result in such an outcome would have to be patently absurd.
Now the question of whether the second person (being not the original buyer) has the right to *use* the code is another question, but if you apply the same reductio ad absurdum to that, you get the same results.
So the correct question is not whether the judge's decision is correct — it very clearly is — but whether the particular path to that conclusion will survive appeal or will be replaced with a different path to the same inevitable conclusion.
You really have to wonder what Disney's lawyers were smoking to have believed that they could pull this off. IMO, it isn't just clear-cut legally; it is *laughably* so.
The frequency at which a law is violated has no bearing on whether something is illegal or not. Don't believe me? Drive 10 over the speed limit past a cop, and then tell a judge that because the overwhelming majority of Americans speed, you shouldn't have to pay a fine for doing it.
We've been saying ever since the iPhone 6 came out that putting the power button directly across from the volume buttons made it hard to turn off the phone without accidentally hitting the volume buttons, which cancels out the power button press.
So what did Apple do about it? They made holding down power + either volume button call freaking 911. It seems pretty likely that they didn't run any tests beforehand to determine how often those buttons were pressed accidentally, because if they had, the entire concept should have been shot down very, very quickly (along with, to be blunt, the completely asinine placement of the power button directly opposite the volume buttons, but I digress).
(Yeah, yeah, they wanted the emergency triggering to be able to be done with one hand. Whatever. You can either avoid false triggering or make it easy to trigger. You can't do both.)
Remember when SJ said that his most important job was saying, "No?" This is why it matters. And yes, they should be fined.
A large percentage of the rest are people calling about non-emergency stuff. There's a homeless camp outside my house, someone broke into my car last night, my car was towed, and the like. Since you have these folks numbers how about a three strike policy. You get three strikes over two years, then you get fined $200.
On Sunday, I called the Santa Cruz police department because my car got broken into. Their voicemail system said that their phone number is for administrative calls, and that if you have an emergency or if you need an officer to respond, call 911. I hung up and called 911.
The problem is not that people are calling 911 for non-emergencies. The problem is that the 911 system is underbuilt for its real-world utilization. That's a problem that simply can't be solved by punishing the users for using it as it was intended.
No, a partial ban on drinking straws that allows restaurants to provide them for the handicapped is simply not a viable approach. A partial ban would make this reasonable accommodation increasingly hard to provide, and thus will make it more and more rare. The problem is, if you ban them for the rest of the population, then restaurants won't have any straws to provide, so the people who need them won't be able to get them even if they ask. It would take a *lot* of ADA lawsuits against individual businesses before the average restaurant owner will hear about it, so a lawsuit against the law itself is really the only approach that stands a chance of preventing serious harm.
As for the percentage, about 2.5% of the population have either Essential Tremor or Parkinson's Disease, and possibly as high as 4% if you expand that to include all movement disorders. However, not all of those people have tremors severe enough to make such accommodations necessary. At some point (assuming they all live long enough), they probably all will, but not at the same time.
So the number is not a small number, but it's also not a big enough number to avoid a high rate of accidental discrimination by restaurants. Such accidental discrimination occurs with alarming regularity even without this law. My parents actually carry a disposable plastic cup in their car at all times, because so many restaurants don't have cups with lids. When they're traveling, it can cause significant problems. Laws like this will make that problem much worse unless lawmakers balance it with an explicit legal requirement for straws and cups with lids to be available at all restaurants for people who have movement disorders—not a "may provide", but rather a "must provide". Otherwise, most restaurant owners will assume that most patrons won't sue over a lack of straws, and they will simply not bother to buy straws. And at some point, it will become impossible for them to buy straws even for the restaurant owners who want to do so, because their suppliers will stop carrying them. And prices will increase as demand decreases, resulting in restaurants having to charge for straws, which becomes another form of discrimination.
But, as I said, a regulation requiring disposable straws to be made out of compostable material would be fine. It would have almost the same ecological impact (at least over the long term), would have minimal impact on the cost for restaurants, and would not have any negative impact on patrons with movement disorders. The only thing it wouldn't do is eliminate the negligible environmental cost of transporting those compostable drinking straws from restaurants to the landfills, but eliminating that impact just isn't worth the harm it would cause to our nation's elderly.
Senator: What do you mean there are no licensed drivers in the State of California? Elon Musk: Well, you see, the government asked us to start a program in which we automatically reported every significant driving mistake, and it counted as a point on your license. After a month, we ran out of licensed drivers. Senator: That's terrible. Can't you fix it so that it only counts the mistakes that got close to causing a crash? Elon Musk: This is the fixed version. We started out with a version that included speeding tickets, and ran a 1% study. We went back to the drawing board when the number of licensed drivers dropped to zero within the first hour of driving.
For people with motion disorders, such as Parkinson's Disease or Essential Tremors, straws make the difference between being able to drink and being unable to do so. Banning all straws would be a VERY BAD thing, and if anyone tries it in the U.S., I guarantee there will be an ADA suit overturning the law within the first week.
Of course, a ban on non-compostable plastic straws would be fine. There are plenty of more environmentally friendly alternatives that work just as well. But a ban on all plastic straws would be a disaster.
Huh? Reading comprehension fail, dude. The goalposts are precisely where they were before. The hotel example explains why dropping HDMI was at least a decade premature. The fact that no hotel rooms have 5K or higher displays proves my point.
Displayport doesn't do everything one needs unless you never have to connect to a monitor or TV that's owned by someone else, and you replace all your TVs.
I guess I should have been more precise. Some monitors provide HDMI for backwards compatibility, but you'll need a faster connection to use their highest resolution modes. And many of them don't offer HDMI at all.
Currently, HDMI-based displays are limited to 4K resolutions in HDMI mode. We've had 5K monitors since 2014, and 8K monitors are available on store shelves today. At the current pace, by the time actual HDMI 2.1 hardware (8K/10K support, finalized last November) arrives in stores, we'll probably have a Thunderbolt 4 standard that leaves it in the dust again.
In a world dominated by Facebook and Google, Apple is hands-down the most privacy and security-focused major vendor. There's something important to be said for that.
Privacy, maybe. Security? Not so much. For many corporate users, new major OS releases have to be delayed by a number of months while internal tools are updated to support them. This means there are a *lot* of high-value computers running at least the immediate previous version of the OS. When the Spectre/Meltdown patches came out, Apple took more than two *weeks* after the High Sierra update before they released the fixes for the previous two versions of macOS, during which time attackers were actively working on creating functioning exploits. Leaving some of your highest-value targets vulnerable is really not the way you do security right.
And they shipped High Sierra with a bug that let anybody get root privileges without a password. And I think I remember reading that they had to pull the OS X update after that and rerelease it because somebody failed to merge those patches in.
I'm not saying their security is a disaster, mind you. Mistakes happen. But in my experience, Apple doesn't have a culture of learning from mistakes, but rather a culture of punishing the people who made the mistakes. The result is that instead of seeing failures as a driving force to improve their infrastructure to ensure that critical bug fixes don't get stomped on, improve their QA and testing to ensure that the bugs don't appear in the first place, etc., I would expect a few low-level engineers to get punished in their annual reviews, and for nothing to improve. And that serious flaw in Apple's culture ensures that these sorts of mistakes will keep happening over and over, just as they always have. Maybe eventually they'll happen one too many times, and someone will say, "Hey, this looks like a systemic problem," but it will take a lot longer than it would take in companies that have a no-fault postmortem culture.
It keeps getting worse, too. The average software update in 10.6 took something like five minutes to install. My last software update (not upgrade) took half an hour. I think that's a strong hint that the OS is simply too darn big.
Remember when all of OS X could fit on a single CD-ROM? Pepperidge Farm remembers.
Call me when even 1% of all hotel TV sets have a USB-C port. Go ahead. I'll wait. I expect to hear from you by 2030 or so.
The problem is, the people making the decisions about these things are assuming that people are buying computers to use with monitors, and dropped the feature because most recent monitors don't even have HDMI. What they forgot is that most people who use the port do so for hooking up to TVs while traveling, and that the Apple TV is useless on most hotel wireless networks because it has no captive portal support.
TVs have a much longer replacement cycle, so it will be a very long time before USB-C is ubiquitous, mostly happening through attrition as old TVs die after 20+ years or new hotels get built. Couple that with the inconvenience of having to remember to pack a pricey specialty adapter that you probably won't be able to find locally, and you have all the necessary pieces for generating some serious long-term ill-will towards Apple. It's like dropping the headphone jacks on iPhones. It won't cause a huge drop in sales, because most people won't think too much about it. But as you use it, you'll keep running into corner cases where you can't do something because of that decision. And every time you do, it erodes the brand in your mind—and all because somebody at Apple thought it was more important to save a couple of bucks per machine on HDMI/HDCP hardware than to give customers a good experience.
Limited Features? Like FOUR USB-C Ports on a Laptop, for an aggregate 80 Gb/s I/O bandwidth, and which can be easily and inexpensively broken-out into a MYRIAD of different configurations, up to FIFTY-TWO SIMULTANEOUS "Legacy" Ports?
So that you can carry a dongle for everything that your device ought to do built-in, like the $70 dongle just to get HDMI output for watching movies in your hotel room. Limited features.
Highly Controlling? Like for example, the fact that, since iOS 8, Apple has officially allowed "Sideloading" of Apps on iOS Devices, both through Open Source XCode Application-Building, and through the loading of precompiled.ipa files using Cydia Impactor, which runs on every desktop platform?
Like the fact that we had to scream for an entire decade to get that capability.
Abandoned Product Lines? Every OEM drops products and sometimes whole product-lines. So?
Every vendor doesn't build the only products compatible with their OS, or require that all iOS apps be compiled on Macs. Ever try to set up a build/test farm now that the XServe is discontinued? See also "Highly Controlling".
Erratic Decision-Making? As compared with, say, Microsoft? Yeahrightsure...
I'm not sure what the GP was thinking about here. Apple's decision-making is pretty self-consistent. As of late, it has resulted in some rather bizarre outcomes, but the logic resulting in those bizarre outcomes was self-consistent, and thus not erratic.
Pros: A much more sensible application packaging model than Windows or Linux, resulting in fewer conflicts and other surprises; a robust, extensible driver architecture (Mac only) that again has fewer conflicts and other surprises; generally reliable hardware (with a fair number of notable exceptions).
Cons: frequent software updates and an inadequate bug fix rate.
Getting regular software updates is useful in that it is always improving, but it is also annoying, because updates involve not using the device for potentially an extended period of time (sometimes as much as half an hour on spinning-rust Macs). I'd rather have fewer, larger updates, with the exception of security updates, which should be tiny and should install quickly.
The major updates make the problem even worse. Apple provides security updates for the last two OS releases. That used to mean you could go four to six years without doing a major OS upgrade, so if something is broken, you had half a decade to deal with it. Now, if something gets broken by a major update, you have two years to find a replacement. And when support for your hardware gets dropped, you have two years to buy a replacement.
And it feels like the bug fix rate really isn't keeping up with the bug creation rate lately. Yesterday, I ran into a bug where some test code wouldn't compile, and there was no obvious reason why. It turns out Apple left out a couple of very important parentheses in a number of their XCTAssert macros. Somebody filed a bug about it (rdar://14504007) in 2013, and almost five years later, they still haven't fixed it, even though the fix should be zero-risk and would literally take seconds to fix. Checking the change into their build system would take longer than the fix.
One of the things I consider important when it comes to judging the quality of software is whether the manufacturer fixes the bugs that I care about. Obviously, bugs that affect the most users must have the highest priority, but that doesn't mean the other bugs shouldn't eventually get fixed. Unfortunately, at Apple, it is common for projects to gets cancelled with crazy numbers of bugs still open (and then closed as NTBF). I'm not sure if their bug triaging processes are simply inadequate, if they just don't have enough people to fix bugs, or if they are just introducing too many new bugs (and thus running out of time to fix old ones), but either way, when something can be fixed in minutes and it still hasn't happened after five years, something is very, very wrong.
And it isn't just developer-facing bugs, either. If you've ever used CarPlay and sworn when your car starts playing music as soon as you get into the car, there's an open bug asking for a switch to turn that off. That's a pretty big annoyance for a *lot* of people, but the bug is still unfixed after almost two years.
Up until 2005—as in not including 2005. Look at a report from Q4 2004. They had sales figures for iMac, iBook, PowerBook, and Power Mac individually, plus iPod. They didn't break down individual models of PowerBook or iBook, because in each line, they basically only had the one model (just in different configurations). They should have broken down iPod vs. iPod Mini sales, but didn't. But still, that's a *lot* more detail than "Desktops" and "Laptops". In 2005, the product categories changed to the level of detail that we have now.
Slashdot Mirror
And the first time somebody gets stuck on the side of the road for lack of wiper fluid, they'll learn to buy a $5 bottle and keep it at home. We're not talking about something you have to fill up twice a week here....
Obviously, the sensors have to operate when not entirely clean, but there are limits, and there's no way to avoid requiring that they be at least somewhat clean.
Half a stop better. IMO, it's still about 5 1/2 stops too slow for low-light photography, thanks to the crop factor/tiny sensor, but I guess every little bit helps.
I'm kind of surprised that there's a noticeable improvement in depth of field between f/1.5 and f/2.4 on a sensor that small. I'd have expected diffraction limiting to make such dual-aperture designs not worth bothering with. But maybe not.
*shrugs*
Why the heck would someone put a plastic lens on a sensor intended for use in a high-dirt environment? Even glass isn't hard enough. They should be using sapphire crystal, at an absolute minimum. That's a pretty big design mistake.
That's easily solved. Most people never run out of fuel, because there's a gauge to tell them that they need to fill it up. If cars had an idiot light that says "Refill wiper fluid now" that goes on when it is at the 25% mark, nobody would run out of wiper fluid, either. Car companies don't add that sensor that because wiper fluid isn't a critical safety feature. If it becomes one, they presumably will do so.
Or they could just add red seaweed to their feed.
They absolutely can prevent transfer of a license. The license is obtained by entering the code into a website, and that code is usable only once. That's not where Disney's argument falls apart from a legal perspective.
The problem with Disney's argument is that their reasoning, when applied to slightly different situations, results in a legal interpretation that fails the common sense test. Consider three scenarios:
In all three cases, the result is the same: I have a DVD and you have an electronic copy, and we each spent $5.
Nothing in Title 17 could plausibly explain why only one of these three transactions is legal, given that the end result and process are effectively identical except for trivial accounting differences. Such a requirement completely fails to stand up to the slightest bit of logical scrutiny, and any legal code that would result in such an outcome would have to be patently absurd.
Now the question of whether the second person (being not the original buyer) has the right to *use* the code is another question, but if you apply the same reductio ad absurdum to that, you get the same results.
So the correct question is not whether the judge's decision is correct — it very clearly is — but whether the particular path to that conclusion will survive appeal or will be replaced with a different path to the same inevitable conclusion.
You really have to wonder what Disney's lawyers were smoking to have believed that they could pull this off. IMO, it isn't just clear-cut legally; it is *laughably* so.
The frequency at which a law is violated has no bearing on whether something is illegal or not. Don't believe me? Drive 10 over the speed limit past a cop, and then tell a judge that because the overwhelming majority of Americans speed, you shouldn't have to pay a fine for doing it.
How about the first 18 digits of pi? Stay in school, kids. It could save your life someday.
We've been saying ever since the iPhone 6 came out that putting the power button directly across from the volume buttons made it hard to turn off the phone without accidentally hitting the volume buttons, which cancels out the power button press.
So what did Apple do about it? They made holding down power + either volume button call freaking 911. It seems pretty likely that they didn't run any tests beforehand to determine how often those buttons were pressed accidentally, because if they had, the entire concept should have been shot down very, very quickly (along with, to be blunt, the completely asinine placement of the power button directly opposite the volume buttons, but I digress).
(Yeah, yeah, they wanted the emergency triggering to be able to be done with one hand. Whatever. You can either avoid false triggering or make it easy to trigger. You can't do both.)
Remember when SJ said that his most important job was saying, "No?" This is why it matters. And yes, they should be fined.
On Sunday, I called the Santa Cruz police department because my car got broken into. Their voicemail system said that their phone number is for administrative calls, and that if you have an emergency or if you need an officer to respond, call 911. I hung up and called 911.
The problem is not that people are calling 911 for non-emergencies. The problem is that the 911 system is underbuilt for its real-world utilization. That's a problem that simply can't be solved by punishing the users for using it as it was intended.
No, a partial ban on drinking straws that allows restaurants to provide them for the handicapped is simply not a viable approach. A partial ban would make this reasonable accommodation increasingly hard to provide, and thus will make it more and more rare. The problem is, if you ban them for the rest of the population, then restaurants won't have any straws to provide, so the people who need them won't be able to get them even if they ask. It would take a *lot* of ADA lawsuits against individual businesses before the average restaurant owner will hear about it, so a lawsuit against the law itself is really the only approach that stands a chance of preventing serious harm.
As for the percentage, about 2.5% of the population have either Essential Tremor or Parkinson's Disease, and possibly as high as 4% if you expand that to include all movement disorders. However, not all of those people have tremors severe enough to make such accommodations necessary. At some point (assuming they all live long enough), they probably all will, but not at the same time.
So the number is not a small number, but it's also not a big enough number to avoid a high rate of accidental discrimination by restaurants. Such accidental discrimination occurs with alarming regularity even without this law. My parents actually carry a disposable plastic cup in their car at all times, because so many restaurants don't have cups with lids. When they're traveling, it can cause significant problems. Laws like this will make that problem much worse unless lawmakers balance it with an explicit legal requirement for straws and cups with lids to be available at all restaurants for people who have movement disorders—not a "may provide", but rather a "must provide". Otherwise, most restaurant owners will assume that most patrons won't sue over a lack of straws, and they will simply not bother to buy straws. And at some point, it will become impossible for them to buy straws even for the restaurant owners who want to do so, because their suppliers will stop carrying them. And prices will increase as demand decreases, resulting in restaurants having to charge for straws, which becomes another form of discrimination.
But, as I said, a regulation requiring disposable straws to be made out of compostable material would be fine. It would have almost the same ecological impact (at least over the long term), would have minimal impact on the cost for restaurants, and would not have any negative impact on patrons with movement disorders. The only thing it wouldn't do is eliminate the negligible environmental cost of transporting those compostable drinking straws from restaurants to the landfills, but eliminating that impact just isn't worth the harm it would cause to our nation's elderly.
I can see it now.
Senator: What do you mean there are no licensed drivers in the State of California?
Elon Musk: Well, you see, the government asked us to start a program in which we automatically reported every significant driving mistake, and it counted as a point on your license. After a month, we ran out of licensed drivers.
Senator: That's terrible. Can't you fix it so that it only counts the mistakes that got close to causing a crash?
Elon Musk: This is the fixed version. We started out with a version that included speeding tickets, and ran a 1% study. We went back to the drawing board when the number of licensed drivers dropped to zero within the first hour of driving.
For people with motion disorders, such as Parkinson's Disease or Essential Tremors, straws make the difference between being able to drink and being unable to do so. Banning all straws would be a VERY BAD thing, and if anyone tries it in the U.S., I guarantee there will be an ADA suit overturning the law within the first week.
Of course, a ban on non-compostable plastic straws would be fine. There are plenty of more environmentally friendly alternatives that work just as well. But a ban on all plastic straws would be a disaster.
Huh? Reading comprehension fail, dude. The goalposts are precisely where they were before. The hotel example explains why dropping HDMI was at least a decade premature. The fact that no hotel rooms have 5K or higher displays proves my point.
Displayport doesn't do everything one needs unless you never have to connect to a monitor or TV that's owned by someone else, and you replace all your TVs.
I guess I should have been more precise. Some monitors provide HDMI for backwards compatibility, but you'll need a faster connection to use their highest resolution modes. And many of them don't offer HDMI at all.
Currently, HDMI-based displays are limited to 4K resolutions in HDMI mode. We've had 5K monitors since 2014, and 8K monitors are available on store shelves today. At the current pace, by the time actual HDMI 2.1 hardware (8K/10K support, finalized last November) arrives in stores, we'll probably have a Thunderbolt 4 standard that leaves it in the dust again.
Yeah, is it bad that the first thing I thought when I read the headline was, "What kind of cryptocurrency is Cobalt?"
Privacy, maybe. Security? Not so much. For many corporate users, new major OS releases have to be delayed by a number of months while internal tools are updated to support them. This means there are a *lot* of high-value computers running at least the immediate previous version of the OS. When the Spectre/Meltdown patches came out, Apple took more than two *weeks* after the High Sierra update before they released the fixes for the previous two versions of macOS, during which time attackers were actively working on creating functioning exploits. Leaving some of your highest-value targets vulnerable is really not the way you do security right.
And they shipped High Sierra with a bug that let anybody get root privileges without a password. And I think I remember reading that they had to pull the OS X update after that and rerelease it because somebody failed to merge those patches in.
I'm not saying their security is a disaster, mind you. Mistakes happen. But in my experience, Apple doesn't have a culture of learning from mistakes, but rather a culture of punishing the people who made the mistakes. The result is that instead of seeing failures as a driving force to improve their infrastructure to ensure that critical bug fixes don't get stomped on, improve their QA and testing to ensure that the bugs don't appear in the first place, etc., I would expect a few low-level engineers to get punished in their annual reviews, and for nothing to improve. And that serious flaw in Apple's culture ensures that these sorts of mistakes will keep happening over and over, just as they always have. Maybe eventually they'll happen one too many times, and someone will say, "Hey, this looks like a systemic problem," but it will take a lot longer than it would take in companies that have a no-fault postmortem culture.
It keeps getting worse, too. The average software update in 10.6 took something like five minutes to install. My last software update (not upgrade) took half an hour. I think that's a strong hint that the OS is simply too darn big.
Remember when all of OS X could fit on a single CD-ROM? Pepperidge Farm remembers.
Call me when even 1% of all hotel TV sets have a USB-C port. Go ahead. I'll wait. I expect to hear from you by 2030 or so.
The problem is, the people making the decisions about these things are assuming that people are buying computers to use with monitors, and dropped the feature because most recent monitors don't even have HDMI. What they forgot is that most people who use the port do so for hooking up to TVs while traveling, and that the Apple TV is useless on most hotel wireless networks because it has no captive portal support.
TVs have a much longer replacement cycle, so it will be a very long time before USB-C is ubiquitous, mostly happening through attrition as old TVs die after 20+ years or new hotels get built. Couple that with the inconvenience of having to remember to pack a pricey specialty adapter that you probably won't be able to find locally, and you have all the necessary pieces for generating some serious long-term ill-will towards Apple. It's like dropping the headphone jacks on iPhones. It won't cause a huge drop in sales, because most people won't think too much about it. But as you use it, you'll keep running into corner cases where you can't do something because of that decision. And every time you do, it erodes the brand in your mind—and all because somebody at Apple thought it was more important to save a couple of bucks per machine on HDMI/HDCP hardware than to give customers a good experience.
So that you can carry a dongle for everything that your device ought to do built-in, like the $70 dongle just to get HDMI output for watching movies in your hotel room. Limited features.
Like the fact that we had to scream for an entire decade to get that capability.
Every vendor doesn't build the only products compatible with their OS, or require that all iOS apps be compiled on Macs. Ever try to set up a build/test farm now that the XServe is discontinued? See also "Highly Controlling".
I'm not sure what the GP was thinking about here. Apple's decision-making is pretty self-consistent. As of late, it has resulted in some rather bizarre outcomes, but the logic resulting in those bizarre outcomes was self-consistent, and thus not erratic.
Pros: A much more sensible application packaging model than Windows or Linux, resulting in fewer conflicts and other surprises; a robust, extensible driver architecture (Mac only) that again has fewer conflicts and other surprises; generally reliable hardware (with a fair number of notable exceptions).
Cons: frequent software updates and an inadequate bug fix rate.
Getting regular software updates is useful in that it is always improving, but it is also annoying, because updates involve not using the device for potentially an extended period of time (sometimes as much as half an hour on spinning-rust Macs). I'd rather have fewer, larger updates, with the exception of security updates, which should be tiny and should install quickly.
The major updates make the problem even worse. Apple provides security updates for the last two OS releases. That used to mean you could go four to six years without doing a major OS upgrade, so if something is broken, you had half a decade to deal with it. Now, if something gets broken by a major update, you have two years to find a replacement. And when support for your hardware gets dropped, you have two years to buy a replacement.
And it feels like the bug fix rate really isn't keeping up with the bug creation rate lately. Yesterday, I ran into a bug where some test code wouldn't compile, and there was no obvious reason why. It turns out Apple left out a couple of very important parentheses in a number of their XCTAssert macros. Somebody filed a bug about it (rdar://14504007) in 2013, and almost five years later, they still haven't fixed it, even though the fix should be zero-risk and would literally take seconds to fix. Checking the change into their build system would take longer than the fix.
One of the things I consider important when it comes to judging the quality of software is whether the manufacturer fixes the bugs that I care about. Obviously, bugs that affect the most users must have the highest priority, but that doesn't mean the other bugs shouldn't eventually get fixed. Unfortunately, at Apple, it is common for projects to gets cancelled with crazy numbers of bugs still open (and then closed as NTBF). I'm not sure if their bug triaging processes are simply inadequate, if they just don't have enough people to fix bugs, or if they are just introducing too many new bugs (and thus running out of time to fix old ones), but either way, when something can be fixed in minutes and it still hasn't happened after five years, something is very, very wrong.
And it isn't just developer-facing bugs, either. If you've ever used CarPlay and sworn when your car starts playing music as soon as you get into the car, there's an open bug asking for a switch to turn that off. That's a pretty big annoyance for a *lot* of people, but the bug is still unfixed after almost two years.
Up until 2005—as in not including 2005. Look at a report from Q4 2004. They had sales figures for iMac, iBook, PowerBook, and Power Mac individually, plus iPod. They didn't break down individual models of PowerBook or iBook, because in each line, they basically only had the one model (just in different configurations). They should have broken down iPod vs. iPod Mini sales, but didn't. But still, that's a *lot* more detail than "Desktops" and "Laptops". In 2005, the product categories changed to the level of detail that we have now.
I think we can prove that it's the right action, anyway. We just can't prove that it's for the wrong reason. :-)