Oh, so that way apps that automatically scale up to the size of your display, and expect "edge swipe" gestures don't work very good because the edge is a dead zone... but only on a few phones that have hardware designed for marketing rather than function.
Which is more, $5/mo from all cable and satellite providers in the Southern California area, or $5/mo opt in from Dodgers fans in the Southern California area?
The almighty dollar runs the show when it comes to channel carriage, and especially professional sports properties.
This is the kind of question that was asked after the Apollo 1 fire. Yes, there are risks in spaceflight. That doesn't mean that there isn't way to mitigate risks, or that undue risk has to be taken.
I imagine the benefit to fueling with the people already in the capsule is that you'll have less liquid oxygen boil off before launch, if you can launch as soon as the thing is fueled. Is that worth risking the lives of people? No, especially as you can likely fix it with a procedural change.
While technically correct, they will still win that battle in anything but an actual legal judgement. They have lawyers and plenty of resources and time, while you have a broken car, a job you have to go to, and rent / mortgage to pay. And you probably don't have the retainer for a lawyer's time to fight the kind of asshat company that would have used the DMCA to sue you yesterday, but can't today.
It depends on your agreement with Microsoft. A large company I worked for had "n-1" version regression rights in their EA. So if they are buying systems stickered with Windows 8, they can legally install Windows 7 on it.
As with all things Microsoft, it's a matter of how much you pay them.
There won't be any challenges, they just won't sue you under the DMCA.
They'll still void any warranty you may have and either refuse to work on it, or just fuck you bigtime if anything goes wrong that's even remotely connected to the "hack".
And who ultimately loses? The people that don't subscribe to Comcast or DirecTV, because they get a double-helping from the bowl of shit. That is, until their provider buys / merges with some other content house in order to have a seat at the big boys' table too.
ABC/Disney has a shitload of money, and holds the ESPN properties. Now taking bets on who they buy and when, should this TWAT&T deal happen.
This merger is only about one thing: sticking it to Comcast the way that Comcast has been sticking it to everyone else for NBC/Universal content. With AT&T owning Turner Broadcasting / Warner Brothers Home Entertainment / HBO, they've got a way to negotiate lower rates from Comcast in trade for continued lower rates for TimeWarner crap.
Who loses? People that subscribe to companies that aren't AT&T or Comcast, as they're going to discover the length, width, and breadth of the shaft until their service gets their own content production company and buy their way into the new game.
The next penny I expect to drop: Disney/ABC buying either Dish Networks or Charter Communications in order to play this game.
I'd bet a box of donuts that the chip China came up with is a direct rip of a Xeon. It's not like China would say "oh, that's Intel's patented IP, we can't use it."
They are far more likely to say "Oh, the US wants to ban us from being able to buy this chip? We'll just reverse engineer it and make our own factory to crank them out." It's not like they'd have a hard time getting their hands on working Xeon chips, what with Lenovo being a Chinese company...
I get the draw of adopting an existing standard, but what if Apple comes up with something better? I'm not saying that they will, and I don't know what it would be if they did, but adopting standards for the sake of adopting standards could have a stagnating effect.
If they do it right, they could support Qi, and do their own thing. That way you can interoperate with all the Qi stuff out there (airports, 3rd party integrations) for a base functionality, and then do your "Magic Charger (tm)" that has some extra special Apple-y shit.
Unfortunately, that's the path that Apple hardly takes.
Yeah, because no other company starts working on the next version of a successful product after launching the current one. They just sit back and enjoy the breeze.
Did I say that I need almost 2TB of RAM? Did anyone say that? And, by the way, the only way you can get one of those X1 class instances right now is by doing a reserved instance, which means you're agreeing to pay $4/hr for the next three years whether you use it or not. That's over $100k. Sounds like a fucking bargain in comparison to just buying several laptops with more RAM in it for $300 more each, especially since for the same cost we could buy 20 developers the top spec'd MacBook Pro every single year and save money compared to that single X1 host, which is still backed by slow EBS storage; any MacBook Pro from the last 3 years has better I/O than you will see from EBS, and probably by an order of magnitude.
Good job knocking the hell out of a straw man though. Here's what my devs want:
4-core laptop, at least 32GB RAM. Why? So they can run an Ubuntu app server VM that mirrors the production stack config, a DB host VM or container, and still have ample room for the IDE, multiple web browsers for testing, bloated Outlook email, etc.
It's no secret that every single database engine runs better with more memory, and is massively I/O bound if you don't have the RAM to cache the whole thing. They can currently do the above with 16GB of RAM but the database runs like crap, because the data set they are working with is ~200GB, and you can't cache very good when you only have somewhere around 8GB to 10GB of RAM to give the whole VM, much less the DB engine.
I work in AWS every day, I know what the instances are capable of, and I know what the shortfalls are. Sometimes there is no substitute for having a self contained development environment that doesn't require network connectivity in order to function.
I work with software developers and "the cloud" every day - we have around 150 instances in AWS that are both production and development systems. Our developers still work off their local laptops whenever possible, and want to move their primary development environment locally using Vagrant and VMs because you never have connection issues, and you get to use the full bandwidth of the local NVMe SSD in your laptop rather than the slow EBS storage you get with EC2 or RDS. Not to mention that if you should not have an internet connection available (think: airplane), you can still work because your laptop isn't an expensive dumb terminal.
Some of the queries and reports they are running take 20+ minutes in "the cloud" and take far less time running locally. But you're saying that we should just move everything into AWS and the developers should just get over it? Because a laptop manufacturer can't figure out how to wedge a little more RAM into the computer?
Just on a cost perspective: $300 once for more RAM (probably less, but this is Apple we're talking about), or hourly charges forever on a cloud provider, and a developer twiddling their thumbs while waiting for a test to complete on a less efficient system. Let me think that one over real quick...
I'm actually ok with this except for #3, because it would be a bit at odds with #4. Your ideas go towards preventing my biggest issue with the current political structure - large stacks of cash buying politicians wholesale. The issue I have is that the corporations and unions paying for and running their own ads is indistinguishable from giving craptons of cash to the campaign for ad buys. It would just change the accounts ledger for the campaign a bit - they could spend the individual donation money on non-media expenses that these organizations cannot fund, and let the 3rd party entities pick up the ad buys. Fundamentally, nothing changes - large bags of cash traded during the campaign for political favors in the future once they are in whatever office they are running for.
Too bad you've got me pegged completely wrong. I'm not a "left winger" unless you are one of these guys who thinks that actually striking a compromise with "the other side of the aisle" in order to let the government actually function and serve the people is "selling out". And you know what? If you are that guy, that's cool for you - be your own guy; I don't have a problem with it the way that you seem to.
At the end of the day, the same problem that allows Soros to have as much influence as he does, is the same problem that allows the Koch brothers to have the undue influence that they do. Nobody's "speech" should have more weight than anyone else's just because they have a fatter stack of cash, and that's exactly what the system we have right now accomplishes.
Why are we afraid of letting ideas stand on their own merit, without having to buy undue influence? Remember when that was called "bribery" rather than "speech" ?
It's like they didn't even ask "professionals" what they really wanted. Thinner and only one type of albeit faster port? I highly doubt anybody was asking for that. It probably would have been "Faster, longer battery life, maybe add 1 or 2 USB-C ports, otherwise leave it alone".
The same could have been said about the Mac Pro in 2013. "Faster CPU and RAM, PCI-E 3.0, Throw some USB 3.0 and Thunderbolt on there, and give us a power supply that can fire up 4 GPUs."
Instead we got a little cylinder that left people unimpressed with specs at launch, and is just a sad joke 3 years later after being completely ignored. The thermal engineering of the thing is rather impressive, but it's only required because they seemed to think that size of a workstation is far more important than the performance and expandability of a workstation, which has been patently false since the "workstation" segment was created many years ago.
The chipset limitation is that Apple is making excuses in order to justify making an already slim and light notebook into a slightly slimmer and lighter notebook at the cost of additional wanted functionality. The "battery life" excuse can only come from two places:
1. They would have to stack chips in a memory package because there aren't dense enough modules available yet in current large-scale manufacturing, causing (slightly) less volume for battery, and (slightly) more amperage to be used to refresh the DRAM.
2. Phil Schiller's ass.
And I'm typing this on a 2014 MacBook Pro, so no I'm not a 'hater.'
AWS I/O is garbage for large data moves unless you are using the ephemeral SSD volume which gets wiped when you stop the instance. And, the instant you move the data from that ephemeral disk to an EBS volume, you're back to shared-tenant storage linked through a weak-shit emulated gigabit interface (even with the "enhanced EBS storage" unless you fork over big cash for a huge instance where you don't need all the CPU and RAM, but they are nice enough to give you a 10GB interface).
Sincerely, someone who routinely is I/O bound on AWS EC2 while moving about databases to software development environments.
If AWS allowed you to a la carte the EBS bandwidth in addition to the provisioning of IOPS, it would be far better than it is today. But I think Amazon likes charging people that provision 20,000 IOPS SSDs, who will never see more than 5,000 IOPS of actual utilization because the EBS bandwidth is maxed out - it probably helps them oversubscribe the shit of our IOPS on their storage arrays without anyone actually noticing unless they look real hard.
Except for when you need to move 1TB of data into said environment. Goes pretty quick on a laptop via Thunderbolt / USB3. Pretty shitty pushing that over a congested cable modem or DSL via WiFi. Remember, we're talking about a laptop here, so you won't always be sitting on your company's metro ethernet connection to the Internet. And you're totally fucked if you happen to be on an airplane - somewhere that laptops absolutely never get used.
Oh, and if that data holds sensitive information, is your favorite cloud service a good place to be keeping that? How many violations of security policy / PCI audit failures would that entail? Sure, you might be able to scrub it first, but that just adds to the time it takes to get that data there, and you still have to have local assets to do the scrubbing.
It's also not a good sign when you are deciding what someone's acumen is based on the brand of tools they choose to use. In fact, it basically marks you as an idiot.
Here's a hint: it doesn't fucking matter as long as the job gets done, and done right.
Because it's pretty nice for a developer to be able to test code against actual data, without potentially fucking over production databases, or requiring a rebuild of a QA environment?
VM snapshotting is a thing, and has been for years.
Slashdot Mirror
Oh, so that way apps that automatically scale up to the size of your display, and expect "edge swipe" gestures don't work very good because the edge is a dead zone... but only on a few phones that have hardware designed for marketing rather than function.
Good call.
Wouldn't have been anywhere close to the revenue.
Which is more, $5/mo from all cable and satellite providers in the Southern California area, or $5/mo opt in from Dodgers fans in the Southern California area?
The almighty dollar runs the show when it comes to channel carriage, and especially professional sports properties.
So the EVP of Worldwide Marketing for a company is a shill for that company's products. I'm absolutely overwhelmed with surprise.
Why do we need to have a new episode of The Phil Schiller Emails every day?
Current website, versus link from 2014, and the person citing the link in 2014 thinks the person citing the current site is a retard.
Oh, how Slashdot has gone to hell.
This is the kind of question that was asked after the Apollo 1 fire. Yes, there are risks in spaceflight. That doesn't mean that there isn't way to mitigate risks, or that undue risk has to be taken.
I imagine the benefit to fueling with the people already in the capsule is that you'll have less liquid oxygen boil off before launch, if you can launch as soon as the thing is fueled. Is that worth risking the lives of people? No, especially as you can likely fix it with a procedural change.
While technically correct, they will still win that battle in anything but an actual legal judgement. They have lawyers and plenty of resources and time, while you have a broken car, a job you have to go to, and rent / mortgage to pay. And you probably don't have the retainer for a lawyer's time to fight the kind of asshat company that would have used the DMCA to sue you yesterday, but can't today.
It depends on your agreement with Microsoft. A large company I worked for had "n-1" version regression rights in their EA. So if they are buying systems stickered with Windows 8, they can legally install Windows 7 on it.
As with all things Microsoft, it's a matter of how much you pay them.
There won't be any challenges, they just won't sue you under the DMCA.
They'll still void any warranty you may have and either refuse to work on it, or just fuck you bigtime if anything goes wrong that's even remotely connected to the "hack".
And who ultimately loses? The people that don't subscribe to Comcast or DirecTV, because they get a double-helping from the bowl of shit. That is, until their provider buys / merges with some other content house in order to have a seat at the big boys' table too.
ABC/Disney has a shitload of money, and holds the ESPN properties. Now taking bets on who they buy and when, should this TWAT&T deal happen.
This merger is only about one thing: sticking it to Comcast the way that Comcast has been sticking it to everyone else for NBC/Universal content. With AT&T owning Turner Broadcasting / Warner Brothers Home Entertainment / HBO, they've got a way to negotiate lower rates from Comcast in trade for continued lower rates for TimeWarner crap.
Who loses? People that subscribe to companies that aren't AT&T or Comcast, as they're going to discover the length, width, and breadth of the shaft until their service gets their own content production company and buy their way into the new game.
The next penny I expect to drop: Disney/ABC buying either Dish Networks or Charter Communications in order to play this game.
I'd bet a box of donuts that the chip China came up with is a direct rip of a Xeon. It's not like China would say "oh, that's Intel's patented IP, we can't use it."
They are far more likely to say "Oh, the US wants to ban us from being able to buy this chip? We'll just reverse engineer it and make our own factory to crank them out." It's not like they'd have a hard time getting their hands on working Xeon chips, what with Lenovo being a Chinese company...
I get the draw of adopting an existing standard, but what if Apple comes up with something better? I'm not saying that they will, and I don't know what it would be if they did, but adopting standards for the sake of adopting standards could have a stagnating effect.
If they do it right, they could support Qi, and do their own thing. That way you can interoperate with all the Qi stuff out there (airports, 3rd party integrations) for a base functionality, and then do your "Magic Charger (tm)" that has some extra special Apple-y shit.
Unfortunately, that's the path that Apple hardly takes.
Yeah, because no other company starts working on the next version of a successful product after launching the current one. They just sit back and enjoy the breeze.
Are you cracked?
Did I say that I need almost 2TB of RAM? Did anyone say that? And, by the way, the only way you can get one of those X1 class instances right now is by doing a reserved instance, which means you're agreeing to pay $4/hr for the next three years whether you use it or not. That's over $100k. Sounds like a fucking bargain in comparison to just buying several laptops with more RAM in it for $300 more each, especially since for the same cost we could buy 20 developers the top spec'd MacBook Pro every single year and save money compared to that single X1 host, which is still backed by slow EBS storage; any MacBook Pro from the last 3 years has better I/O than you will see from EBS, and probably by an order of magnitude.
Good job knocking the hell out of a straw man though. Here's what my devs want:
4-core laptop, at least 32GB RAM. Why? So they can run an Ubuntu app server VM that mirrors the production stack config, a DB host VM or container, and still have ample room for the IDE, multiple web browsers for testing, bloated Outlook email, etc.
It's no secret that every single database engine runs better with more memory, and is massively I/O bound if you don't have the RAM to cache the whole thing. They can currently do the above with 16GB of RAM but the database runs like crap, because the data set they are working with is ~200GB, and you can't cache very good when you only have somewhere around 8GB to 10GB of RAM to give the whole VM, much less the DB engine.
I work in AWS every day, I know what the instances are capable of, and I know what the shortfalls are. Sometimes there is no substitute for having a self contained development environment that doesn't require network connectivity in order to function.
I work with software developers and "the cloud" every day - we have around 150 instances in AWS that are both production and development systems. Our developers still work off their local laptops whenever possible, and want to move their primary development environment locally using Vagrant and VMs because you never have connection issues, and you get to use the full bandwidth of the local NVMe SSD in your laptop rather than the slow EBS storage you get with EC2 or RDS. Not to mention that if you should not have an internet connection available (think: airplane), you can still work because your laptop isn't an expensive dumb terminal.
Some of the queries and reports they are running take 20+ minutes in "the cloud" and take far less time running locally. But you're saying that we should just move everything into AWS and the developers should just get over it? Because a laptop manufacturer can't figure out how to wedge a little more RAM into the computer?
Just on a cost perspective: $300 once for more RAM (probably less, but this is Apple we're talking about), or hourly charges forever on a cloud provider, and a developer twiddling their thumbs while waiting for a test to complete on a less efficient system. Let me think that one over real quick...
I'm actually ok with this except for #3, because it would be a bit at odds with #4. Your ideas go towards preventing my biggest issue with the current political structure - large stacks of cash buying politicians wholesale. The issue I have is that the corporations and unions paying for and running their own ads is indistinguishable from giving craptons of cash to the campaign for ad buys. It would just change the accounts ledger for the campaign a bit - they could spend the individual donation money on non-media expenses that these organizations cannot fund, and let the 3rd party entities pick up the ad buys. Fundamentally, nothing changes - large bags of cash traded during the campaign for political favors in the future once they are in whatever office they are running for.
Yes, basic understanding of HTML 1.0 is now beyond the Slashdot editors.
Too bad you've got me pegged completely wrong. I'm not a "left winger" unless you are one of these guys who thinks that actually striking a compromise with "the other side of the aisle" in order to let the government actually function and serve the people is "selling out". And you know what? If you are that guy, that's cool for you - be your own guy; I don't have a problem with it the way that you seem to.
At the end of the day, the same problem that allows Soros to have as much influence as he does, is the same problem that allows the Koch brothers to have the undue influence that they do. Nobody's "speech" should have more weight than anyone else's just because they have a fatter stack of cash, and that's exactly what the system we have right now accomplishes.
Why are we afraid of letting ideas stand on their own merit, without having to buy undue influence? Remember when that was called "bribery" rather than "speech" ?
It's like they didn't even ask "professionals" what they really wanted. Thinner and only one type of albeit faster port? I highly doubt anybody was asking for that. It probably would have been "Faster, longer battery life, maybe add 1 or 2 USB-C ports, otherwise leave it alone".
The same could have been said about the Mac Pro in 2013. "Faster CPU and RAM, PCI-E 3.0, Throw some USB 3.0 and Thunderbolt on there, and give us a power supply that can fire up 4 GPUs."
Instead we got a little cylinder that left people unimpressed with specs at launch, and is just a sad joke 3 years later after being completely ignored. The thermal engineering of the thing is rather impressive, but it's only required because they seemed to think that size of a workstation is far more important than the performance and expandability of a workstation, which has been patently false since the "workstation" segment was created many years ago.
The chipset limitation is that Apple is making excuses in order to justify making an already slim and light notebook into a slightly slimmer and lighter notebook at the cost of additional wanted functionality. The "battery life" excuse can only come from two places:
1. They would have to stack chips in a memory package because there aren't dense enough modules available yet in current large-scale manufacturing, causing (slightly) less volume for battery, and (slightly) more amperage to be used to refresh the DRAM.
2. Phil Schiller's ass.
And I'm typing this on a 2014 MacBook Pro, so no I'm not a 'hater.'
extra (low bandwidth) space.
But you are right, this was yet another useful feature that has been mercilessly shitcanned.
AWS I/O is garbage for large data moves unless you are using the ephemeral SSD volume which gets wiped when you stop the instance. And, the instant you move the data from that ephemeral disk to an EBS volume, you're back to shared-tenant storage linked through a weak-shit emulated gigabit interface (even with the "enhanced EBS storage" unless you fork over big cash for a huge instance where you don't need all the CPU and RAM, but they are nice enough to give you a 10GB interface).
Sincerely,
someone who routinely is I/O bound on AWS EC2 while moving about databases to software development environments.
If AWS allowed you to a la carte the EBS bandwidth in addition to the provisioning of IOPS, it would be far better than it is today. But I think Amazon likes charging people that provision 20,000 IOPS SSDs, who will never see more than 5,000 IOPS of actual utilization because the EBS bandwidth is maxed out - it probably helps them oversubscribe the shit of our IOPS on their storage arrays without anyone actually noticing unless they look real hard.
Except for when you need to move 1TB of data into said environment. Goes pretty quick on a laptop via Thunderbolt / USB3. Pretty shitty pushing that over a congested cable modem or DSL via WiFi. Remember, we're talking about a laptop here, so you won't always be sitting on your company's metro ethernet connection to the Internet. And you're totally fucked if you happen to be on an airplane - somewhere that laptops absolutely never get used.
Oh, and if that data holds sensitive information, is your favorite cloud service a good place to be keeping that? How many violations of security policy / PCI audit failures would that entail? Sure, you might be able to scrub it first, but that just adds to the time it takes to get that data there, and you still have to have local assets to do the scrubbing.
It's also not a good sign when you are deciding what someone's acumen is based on the brand of tools they choose to use. In fact, it basically marks you as an idiot.
Here's a hint: it doesn't fucking matter as long as the job gets done, and done right.
Because it's pretty nice for a developer to be able to test code against actual data, without potentially fucking over production databases, or requiring a rebuild of a QA environment?
VM snapshotting is a thing, and has been for years.