I'm pretty sure Steam has considerably more and will be shocked if Apple TV doesn't pass that in a few weeks.
I'm shocked if it hasn't hit that number already. After all, most games don't use any of the frameworks that are missing in tvOS (e.g. WebKit), so as I understand it, they should mostly be a straight recompile from iOS with just a handful of tiny tweaks. If it doesn't top a hundred thousand game titles within the first year, I'll be surprised.
Amazon isn't using Flash for Prime Video DRM on iPhone, iPad, or iPod Touch. There's no good reason that they can't provide an app on the new Apple TV.
My guess is that Amazon cut corners when they built their iOS app, and instead of building a real native UI, they used a web UI, and because Apple TV doesn't provide WebKit or UIWebView on tvOS, they found themselves having to spend way more work than all the other providers whose engineering teams didn't cut corners. And they're pitching a fit now to try to force Apple to make WebKit available on tvOS. But that's just a guess; I haven't torn apart their app to analyze it.
Which is why it makes perfect sense for them to do this now, after Apple opened up their TV platform to Amazon and others—something about stabbing someone with the olive branch....
No, because then they would look completely incompetent for being pretty much the only major paid streaming content provider that didn't have an app ready on or around the day that the first Apple TV shipped.
Easy solution: Require either user interaction (foreground task) or communication from the main app if you want a background task to run for more than an hour. Don't have either one? Your app gets suspended until it does. That means those background updaters stop wasting power, and basically stop working until the user runs the app.:-)
The Windows HIG specifies that they should be [OK] [Cancel] and OK should be the default action on the form.
If so, then the people who wrote it need to rethink that. Apple got that right when they specified that only safe actions should be the default.
The Apple HIG specifies those buttons in the opposite order.
Not necessarily. It specifies that the most common button is on the right side, and that the default should always be a safe choice, if you specify a default. So you very well might come up with something like this:
"Would you like to reformat your hard drive?" (Yes) ((Cancel))
Because the most common choice (and the safe choice) would be to click "Cancel".
While most replies assume it's okay for local responders to triage, it's just a waste of time.
Careful, there. I know what you meant (I think), but the first time I read that, I was going to disagree with you. It is dangerous to have the local responders determine whether it is worth calling an ambulance before doing so, because it wastes critical minutes. However, it isn't a waste of time to call the local first responders first so that they can get there as quickly as possible.
IMO, calling security first is always the right thing to do. Calling 911 first can cost critical minutes, and can be fatal. Remember, you're one person, and you can only realistically call one person at a time. Security is a group of people, and thus can simultaneously start the local first responder dispatch and the ambulance dispatch while they are still on the phone with you, talking you through determining whether to administer CPR and then beginning CPR, if needed.
Remember that about 60-80% of the time, a heart attack is caused by ventricular fibrillation or ventricular tachycardia. In both cases, an AED can potentially get the person back into normal sinus rhythm; although this may or may not a permanent solution, depending on the cause, it makes the situation less critical. The in-house first responders know where to find an AED and are trained in how to use it properly. Ideally, the magic time you're trying to hit is under three minutes from cardiac event to first shock. For every minute beyond that, the odds of survival decrease by anywhere from 3% if someone is administering CPR to 10% without bystander CPR. And even in non-shockable cases, having a second person to help with CPR is likely to improve the survival rate by reducing fatigue.
By contrast, the speed of ambulance arrival is only critical if the first responders aren't able to get the heart back into rhythm through a combination of CPR followed by an AED shock. With that said, both asystolic heart attacks (flatline) and PEA rhythm (pulseless electrical activity, i.e. normal electrical activity, but no pulse) are associated with low single-digit survival rate, even in the best of circumstances (witnessed by bystanders, immediate CPR, and under 8-minute response time), and in some studies, asystolic heart attacks were as low as 0.2% survival rate. So statistically, you are much more likely to save a life by getting the in-house first responders there sooner than by getting an ambulance there sooner.
As an aside, I've often wondered if use of an epipen (or several, or even vasopressin) by in-house first responders in asystolic cardiac arrests at the onset of CPR would give a better chance of getting the heart back into v-fib faster, and make the heart shockable sooner, resulting in better outcomes. If so, getting the first responders there sooner would have an even bigger impact. For that matter, to a lesser degree, the same treatment works for PEA, so as long as there's independent manual verification that the person has no pulse before you pull out the AED and it says "no shock indicated", there's probably little increased risk from doing it, and significant increased potential for a successful outcome. But that's a subject for another study....
Most folks are saying, "Blame Amazon for taking nine minutes between when he was discovered and when someone actually called 911 to get an ambulance dispatched." If the only reason for such a policy is to ensure that the ambulance gets dispatched to the right place, to have security ready to meet them, and to get the internal emergency personnel there while you wait for the ambulance, then a "call security first" policy is fine. If the reason behind it is to save money on ambulance calls by not calling the ambulance unless you're absolutely certain that the person needs it, it isn't fine, and it costs lives. And a "call only our security people, and never call 911" policy is wrong, period.
Most companies' policies say "Call security first, and then, if you feel that it is urgent to do so, you can call 911 second, but you don't have to." That's the right policy. If corporate security calls them also, the 911 dispatchers will recognize that two people are calling about the same incident, and they'll handle it appropriately by telling you that corporate security has already called them, and an ambulance has been dispatched.
I think the GP is saying that we need a minimum wage law that permanently ties the minimum wage to the CPI, adjusted annually. If we had that, then salaries would have kept up with inflation (or at least would have been much closer to doing so), because the minimum wage would have kept up, and people working for more than minimum wage prior to the increase would begin to gripe that they're now getting minimum wage, so they would get a raise, and so on down the line.:-)
This. I used to work with a couple of folks who were once part of the emergency response team at Apple. Here's what I learned from conversations with them.
If you call 911 from any Apple corporate phone in Santa Clara Valley, unless they've changed things in the last few years, the address that comes up on the 911 switchboard is 1 Infinite Loop, Cupertino. That's definitely true for the entire main campus, and I think it is also true for the satellite offices. If you're at one of their Sunnyvale offices, you might end up directing the ambulance to a site that is twenty minutes' drive away from the actual location of the emergency. Even if you're in one of the Cupertino offices, it could delay the ambulance by as much as ten minutes.
The minimum arrival time for first responders is likely to be four or five minutes, from the fire station down on Stevens Creek Blvd, by which time there had better be someone at the front door to let them in and lead them to the person in crisis (usually security and/or the building receptionist). Otherwise, you've delayed them further.
By contrast, IIRC, there are at least two ERTs with CPR training in every building, and there's an AED on either every floor or every other floor. So if you call security, it takes maybe a minute to get someone there with an AED and begin defibrillation and/or manual CPR if needed. The difference between one minute response and five minutes is enough to make the difference between making a complete recovery and serious brain damage. So unless you are CPR-trained and know how to use an AED (which may or may not be an fully automatic AED—I have no idea), you darn well better call corporate security and get the ERTs to your location as quickly as possible.
But their training always says, "You can call 911, but call us first." There's no rule that says you can't call 911, and if you're at all concerned about whether security will do it in a timely manner, you should absolutely do so. But call security first, because they're going to be able to respond much more quickly, and if they deem it necessary, they can also quickly deliver people to help get the person onto a stretcher, down to the lobby, and curbside, ready to go into the ambulance, saving additional critical minutes. Calling corporate security first can easily make the difference between life and death.
Your situation was somewhat exceptional, because the immediate threat to life was likely over by the time you called. Normally, for anything that might actually be an active life-or-death emergency, 911 asks you to stay on the line until emergency responders arrive, so that if they need additional directions, they don't have to try to call you back, and maybe fail to do so. So calling 911 pretty much precludes calling corporate security unless you have ready access to more than one phone. And because the emergency responders can't get to you without help from corporate security, calling them without calling corporate security is likely to result in someone dying.
You don't know if a person on the floor hit their head on something, passed out from exhaustion or low blood sugar, or (worst case) having a cardiac emergency. Rest and a candy bar doesn't require the waste of EMS resources. A bumped head can be a low priority dispatch.
When someone loses consciousness, you cannot assume that it is a low-priority problem. It is always better to have an EMS response, and have them examine the person and determine that the person will be okay than to not have an EMS response and have the person die while waiting for the security people to call for EMS.
The employees followed a reasonable procedure. If the numbers posted above are correct, then Amazon's security staff did not. The moment they heard that there was a medical emergency with someone collapsed, they should have called on their radio to all internal first responders, but afterwards, they should have been on the phone to 911 within thirty seconds. Any delay longer than that is inexcusable.
You don't wait until your first responders get there. You stay on the line with 911 until your first responders get there, and if they determine that no ambulance is needed, you inform the 911 operator that they can cancel the call (which they may or may not do, depending on what your first responders said, but at that point, you've done your job either way).
This sounds like the comment of someone crazy enough to actually do so alone. Police are not supposed to be an army of one. They are supposed to be a unified force, acting collectively to protect the public. This isn't a movie, and cops aren't supposed to act like John McClane.
This is approximately like saying that the Holocaust was caused by someone who bullied Neville Chamberlain when he was a kid. The effect is so distant from the supposed cause that it is laughable.
To the extent that video scrutiny leads to a rise in violent crime, it is principally because otherwise non-violent criminals have become scared of excessive police brutality, and thus more frequently choose to arm themselves for their own safety. You cannot foment peace at the tip of a sword. Violence begets violence.
Why? The main problem, as presented, is having a single place to find content, with links directly to the video, without having to worry about what app it is in. With minimal changes, the content distributors (Netflix et al) can make that possible. With a little effort, they can probably even provide a way to provide structured data for things like TV shows, with seasons containing episodes, though that may require enhancements to the API. (I haven't looked at the API very closely; if it doesn't provide sufficiently rich structure, that's what enhancement requests are for.)
By contrast, there's vanishingly little benefit to having a single player UI, because most of the time, users never even touch the actual player UI except to pause playback occasionally. Solving the search problem (and, potentially, the browsing problem) is trivial. By contrast, a unified player is unlikely to ever happen because of DRM requirements from the content providers.
And this is why members of Congress should be required by law to spend at least 48 weeks per year in the districts that they serve. Meet physically in D.C. three times per year—for the opening session, for the State of the Union address, and for the closing session. Use a multi-way video conferencing system for everything else. Make the lobbyists work for it by flying to 535 different places instead of being able to talk to everybody in one place.
Last I checked, iOS is not open. It is the most closed and locked-down mobile OS out there.
It is open in the "anyone can write software for it" sense of the term, and in the "all of these streaming companies already support these platforms" sense of the term. And I think Google has a similar API for Android.
My point is that the last thing we need is some new web-based open platform. We already have all the platforms we need. In ten years, I doubt that a significant percentage of video streaming will still involve the web. It's a terrible experience compared with native apps. And the major mobile OSes provide app search APIs already.
No, it's actually the exact opposite of what we asked for. Channels are mini-bundles of content, true. However:
If I buy access to three individual TV channels in an a la carte fashion, although there might be some overlap of older, syndicated shows, I can be reasonably assured that there will be zero or near-zero overlap of new content (unless two of those channels are Disney and ABC).
Netflix and Amazon Prime and Hulu, by contrast, are more like entire cable companies. They each try to carry as much new content as possible. So there's a huge overlap of new content. It is mostly the older, back catalog content where these providers differ greatly.
In fact, the new companies are effectively doing exactly what the old companies did—creating relatively expensive packages of content from dozens of different networks, in an all-or-nothing fashion. The only real difference is that now, each of the individual cable packages comes from a different company, and there's no "give me everything" option without paying for the same content twice.
Yes it is. Different channels, sometimes with overlapping content, are offering their own services for separate prices, all of which can be ultimately displayed by the same end hardware.
That's exactly what a la carte pricing is.
No, not really. True a la carte pricing doesn't generally allow for overlapping content. As soon as you have overlapping content, you aren't a la carte. When I go in a restaurant and order a la carte, I can order a burrito, or I can order a taco, or I can order both. As soon as there's an option that includes both a burrito and a taco, if there's another option that includes a taco and a chimichanga, that's not a la carte. It's bundled.
Similarly, what we have now are basically glorified cable companies, just over the Internet instead of over a fixed physical wire. One might provide shows from two networks, and another might offer shows from three networks, and both might overlap by one. As a result, you're forced to pay for two separate instances of one network in order to get content from all four. That's not a la carte; it is very much bundled.
Worse, unlike cable companies (which rarely originated their own content), these companies create their own original content in addition to aggregating content from other sources, which means the only way to get all the content is to essentially have more than one "cable" company, with massive overlap between them. That's pure insanity.
If Amazon provided a Subprime streaming service that provided only their original content, that would be true a la carte pricing. They would never do that, though, because then they would have to actually compete with Netflix instead of trying to convince everyone that they absolutely need both, even though the two services have something like sixty or seventy percent overlap.
Similarly, if Netflix let you choose which TV networks' shows to include, so that you could exclude the ones that overlap with Hulu, that would be true a la carte pricing. Of course, the networks don't want this, because they like getting paid twice for the same content, so they'll fight any attempts at true a la carte pricing with predatory licensing agreements.
So no, we do not have a la carte pricing in any meaningful sense of the word. What we have is market fragmentation with massive overlap and major gaps in coverage from each of the various providers. That's not the same thing; in fact, they're pretty much polar opposites.:-)
Slashdot Mirror
I'm shocked if it hasn't hit that number already. After all, most games don't use any of the frameworks that are missing in tvOS (e.g. WebKit), so as I understand it, they should mostly be a straight recompile from iOS with just a handful of tiny tweaks. If it doesn't top a hundred thousand game titles within the first year, I'll be surprised.
To be fair, I didn't say it was a good idea. I just said it would be easy to stop the background updaters from using CPU. :-D
Amazon isn't using Flash for Prime Video DRM on iPhone, iPad, or iPod Touch. There's no good reason that they can't provide an app on the new Apple TV.
My guess is that Amazon cut corners when they built their iOS app, and instead of building a real native UI, they used a web UI, and because Apple TV doesn't provide WebKit or UIWebView on tvOS, they found themselves having to spend way more work than all the other providers whose engineering teams didn't cut corners. And they're pitching a fit now to try to force Apple to make WebKit available on tvOS. But that's just a guess; I haven't torn apart their app to analyze it.
Which is why it makes perfect sense for them to do this now, after Apple opened up their TV platform to Amazon and others—something about stabbing someone with the olive branch....
Since a single retailer gained a near monopoly on online product sales. That's when.
No, because then they would look completely incompetent for being pretty much the only major paid streaming content provider that didn't have an app ready on or around the day that the first Apple TV shipped.
Easy solution: Require either user interaction (foreground task) or communication from the main app if you want a background task to run for more than an hour. Don't have either one? Your app gets suspended until it does. That means those background updaters stop wasting power, and basically stop working until the user runs the app. :-)
If so, then the people who wrote it need to rethink that. Apple got that right when they specified that only safe actions should be the default.
Not necessarily. It specifies that the most common button is on the right side, and that the default should always be a safe choice, if you specify a default. So you very well might come up with something like this:
"Would you like to reformat your hard drive?" (Yes) ((Cancel))
Because the most common choice (and the safe choice) would be to click "Cancel".
Careful, there. I know what you meant (I think), but the first time I read that, I was going to disagree with you. It is dangerous to have the local responders determine whether it is worth calling an ambulance before doing so, because it wastes critical minutes. However, it isn't a waste of time to call the local first responders first so that they can get there as quickly as possible.
IMO, calling security first is always the right thing to do. Calling 911 first can cost critical minutes, and can be fatal. Remember, you're one person, and you can only realistically call one person at a time. Security is a group of people, and thus can simultaneously start the local first responder dispatch and the ambulance dispatch while they are still on the phone with you, talking you through determining whether to administer CPR and then beginning CPR, if needed.
Remember that about 60-80% of the time, a heart attack is caused by ventricular fibrillation or ventricular tachycardia. In both cases, an AED can potentially get the person back into normal sinus rhythm; although this may or may not a permanent solution, depending on the cause, it makes the situation less critical. The in-house first responders know where to find an AED and are trained in how to use it properly. Ideally, the magic time you're trying to hit is under three minutes from cardiac event to first shock. For every minute beyond that, the odds of survival decrease by anywhere from 3% if someone is administering CPR to 10% without bystander CPR. And even in non-shockable cases, having a second person to help with CPR is likely to improve the survival rate by reducing fatigue.
By contrast, the speed of ambulance arrival is only critical if the first responders aren't able to get the heart back into rhythm through a combination of CPR followed by an AED shock. With that said, both asystolic heart attacks (flatline) and PEA rhythm (pulseless electrical activity, i.e. normal electrical activity, but no pulse) are associated with low single-digit survival rate, even in the best of circumstances (witnessed by bystanders, immediate CPR, and under 8-minute response time), and in some studies, asystolic heart attacks were as low as 0.2% survival rate. So statistically, you are much more likely to save a life by getting the in-house first responders there sooner than by getting an ambulance there sooner.
As an aside, I've often wondered if use of an epipen (or several, or even vasopressin) by in-house first responders in asystolic cardiac arrests at the onset of CPR would give a better chance of getting the heart back into v-fib faster, and make the heart shockable sooner, resulting in better outcomes. If so, getting the first responders there sooner would have an even bigger impact. For that matter, to a lesser degree, the same treatment works for PEA, so as long as there's independent manual verification that the person has no pulse before you pull out the AED and it says "no shock indicated", there's probably little increased risk from doing it, and significant increased potential for a successful outcome. But that's a subject for another study....
Most folks are saying, "Blame Amazon for taking nine minutes between when he was discovered and when someone actually called 911 to get an ambulance dispatched." If the only reason for such a policy is to ensure that the ambulance gets dispatched to the right place, to have security ready to meet them, and to get the internal emergency personnel there while you wait for the ambulance, then a "call security first" policy is fine. If the reason behind it is to save money on ambulance calls by not calling the ambulance unless you're absolutely certain that the person needs it, it isn't fine, and it costs lives. And a "call only our security people, and never call 911" policy is wrong, period.
Most companies' policies say "Call security first, and then, if you feel that it is urgent to do so, you can call 911 second, but you don't have to." That's the right policy. If corporate security calls them also, the 911 dispatchers will recognize that two people are calling about the same incident, and they'll handle it appropriately by telling you that corporate security has already called them, and an ambulance has been dispatched.
I think the GP is saying that we need a minimum wage law that permanently ties the minimum wage to the CPI, adjusted annually. If we had that, then salaries would have kept up with inflation (or at least would have been much closer to doing so), because the minimum wage would have kept up, and people working for more than minimum wage prior to the increase would begin to gripe that they're now getting minimum wage, so they would get a raise, and so on down the line. :-)
This. I used to work with a couple of folks who were once part of the emergency response team at Apple. Here's what I learned from conversations with them.
If you call 911 from any Apple corporate phone in Santa Clara Valley, unless they've changed things in the last few years, the address that comes up on the 911 switchboard is 1 Infinite Loop, Cupertino. That's definitely true for the entire main campus, and I think it is also true for the satellite offices. If you're at one of their Sunnyvale offices, you might end up directing the ambulance to a site that is twenty minutes' drive away from the actual location of the emergency. Even if you're in one of the Cupertino offices, it could delay the ambulance by as much as ten minutes.
The minimum arrival time for first responders is likely to be four or five minutes, from the fire station down on Stevens Creek Blvd, by which time there had better be someone at the front door to let them in and lead them to the person in crisis (usually security and/or the building receptionist). Otherwise, you've delayed them further.
By contrast, IIRC, there are at least two ERTs with CPR training in every building, and there's an AED on either every floor or every other floor. So if you call security, it takes maybe a minute to get someone there with an AED and begin defibrillation and/or manual CPR if needed. The difference between one minute response and five minutes is enough to make the difference between making a complete recovery and serious brain damage. So unless you are CPR-trained and know how to use an AED (which may or may not be an fully automatic AED—I have no idea), you darn well better call corporate security and get the ERTs to your location as quickly as possible.
But their training always says, "You can call 911, but call us first." There's no rule that says you can't call 911, and if you're at all concerned about whether security will do it in a timely manner, you should absolutely do so. But call security first, because they're going to be able to respond much more quickly, and if they deem it necessary, they can also quickly deliver people to help get the person onto a stretcher, down to the lobby, and curbside, ready to go into the ambulance, saving additional critical minutes. Calling corporate security first can easily make the difference between life and death.
Your situation was somewhat exceptional, because the immediate threat to life was likely over by the time you called. Normally, for anything that might actually be an active life-or-death emergency, 911 asks you to stay on the line until emergency responders arrive, so that if they need additional directions, they don't have to try to call you back, and maybe fail to do so. So calling 911 pretty much precludes calling corporate security unless you have ready access to more than one phone. And because the emergency responders can't get to you without help from corporate security, calling them without calling corporate security is likely to result in someone dying.
The difference is that AFAIK, Apple calls 911 in parallel, not in series (presumably after security gets there and fails to resuscitate the person).
When someone loses consciousness, you cannot assume that it is a low-priority problem. It is always better to have an EMS response, and have them examine the person and determine that the person will be okay than to not have an EMS response and have the person die while waiting for the security people to call for EMS.
The employees followed a reasonable procedure. If the numbers posted above are correct, then Amazon's security staff did not. The moment they heard that there was a medical emergency with someone collapsed, they should have called on their radio to all internal first responders, but afterwards, they should have been on the phone to 911 within thirty seconds. Any delay longer than that is inexcusable.
You don't wait until your first responders get there. You stay on the line with 911 until your first responders get there, and if they determine that no ambulance is needed, you inform the 911 operator that they can cancel the call (which they may or may not do, depending on what your first responders said, but at that point, you've done your job either way).
This sounds like the comment of someone crazy enough to actually do so alone. Police are not supposed to be an army of one. They are supposed to be a unified force, acting collectively to protect the public. This isn't a movie, and cops aren't supposed to act like John McClane.
This is approximately like saying that the Holocaust was caused by someone who bullied Neville Chamberlain when he was a kid. The effect is so distant from the supposed cause that it is laughable.
To the extent that video scrutiny leads to a rise in violent crime, it is principally because otherwise non-violent criminals have become scared of excessive police brutality, and thus more frequently choose to arm themselves for their own safety. You cannot foment peace at the tip of a sword. Violence begets violence.
Why? The main problem, as presented, is having a single place to find content, with links directly to the video, without having to worry about what app it is in. With minimal changes, the content distributors (Netflix et al) can make that possible. With a little effort, they can probably even provide a way to provide structured data for things like TV shows, with seasons containing episodes, though that may require enhancements to the API. (I haven't looked at the API very closely; if it doesn't provide sufficiently rich structure, that's what enhancement requests are for.)
By contrast, there's vanishingly little benefit to having a single player UI, because most of the time, users never even touch the actual player UI except to pause playback occasionally. Solving the search problem (and, potentially, the browsing problem) is trivial. By contrast, a unified player is unlikely to ever happen because of DRM requirements from the content providers.
And this is why members of Congress should be required by law to spend at least 48 weeks per year in the districts that they serve. Meet physically in D.C. three times per year—for the opening session, for the State of the Union address, and for the closing session. Use a multi-way video conferencing system for everything else. Make the lobbyists work for it by flying to 535 different places instead of being able to talk to everybody in one place.
It is open in the "anyone can write software for it" sense of the term, and in the "all of these streaming companies already support these platforms" sense of the term. And I think Google has a similar API for Android.
My point is that the last thing we need is some new web-based open platform. We already have all the platforms we need. In ten years, I doubt that a significant percentage of video streaming will still involve the web. It's a terrible experience compared with native apps. And the major mobile OSes provide app search APIs already.
We have that now. It's called iOS 9. Developers like Netflix and Amazon and Hulu just have to make their content searchable using Core Spotlight.
Yup. This is the difference between doing the right thing [for their bottom line] and not being evil.
No, it's actually the exact opposite of what we asked for. Channels are mini-bundles of content, true. However:
In fact, the new companies are effectively doing exactly what the old companies did—creating relatively expensive packages of content from dozens of different networks, in an all-or-nothing fashion. The only real difference is that now, each of the individual cable packages comes from a different company, and there's no "give me everything" option without paying for the same content twice.
No, not really. True a la carte pricing doesn't generally allow for overlapping content. As soon as you have overlapping content, you aren't a la carte. When I go in a restaurant and order a la carte, I can order a burrito, or I can order a taco, or I can order both. As soon as there's an option that includes both a burrito and a taco, if there's another option that includes a taco and a chimichanga, that's not a la carte. It's bundled.
Similarly, what we have now are basically glorified cable companies, just over the Internet instead of over a fixed physical wire. One might provide shows from two networks, and another might offer shows from three networks, and both might overlap by one. As a result, you're forced to pay for two separate instances of one network in order to get content from all four. That's not a la carte; it is very much bundled.
Worse, unlike cable companies (which rarely originated their own content), these companies create their own original content in addition to aggregating content from other sources, which means the only way to get all the content is to essentially have more than one "cable" company, with massive overlap between them. That's pure insanity.
If Amazon provided a Subprime streaming service that provided only their original content, that would be true a la carte pricing. They would never do that, though, because then they would have to actually compete with Netflix instead of trying to convince everyone that they absolutely need both, even though the two services have something like sixty or seventy percent overlap.
Similarly, if Netflix let you choose which TV networks' shows to include, so that you could exclude the ones that overlap with Hulu, that would be true a la carte pricing. Of course, the networks don't want this, because they like getting paid twice for the same content, so they'll fight any attempts at true a la carte pricing with predatory licensing agreements.
So no, we do not have a la carte pricing in any meaningful sense of the word. What we have is market fragmentation with massive overlap and major gaps in coverage from each of the various providers. That's not the same thing; in fact, they're pretty much polar opposites. :-)