My understanding is that no team of engineers have ever had a fully satisfactory understanding of a space launch system. Space launch designs have such tiny margins for error that some subtle phenomenon going on with the design could be serious enough to destroy the rocket.
The gunshot-sounding pop could easily have been the sound of the first thing to fail inside the rocket. That sound (and any other sounds from the launch pad) could have echoed off of the building. People are notoriously bad at recalling the sequence of events, so it's possible that a person could report hearing the echo before the sound itself.
The problem with the gunshot theory is that it has an incredibly low a priory probability. The ULA has a lot to lose by trying to sabotage their competitors in a blatantly illegal way. It's the last thing they would do. It could have been an employee acting on their own (perhaps someone with mental health problems), but what are the odds that the ULA has such an employee combined with the odds that that employee could have successfully smuggled a rifle all the way to the top of the building? Close to zero.
The gunshot theory only becomes probable if they find hard evidence, like bullet fragments, or tell-tale damage to the wreckage.
Yeah but the point is that the strut that failed in 2015 was directly attached to the helium tank that failed last month. There is literally a strong connection between the two. Perhaps there's a logical connection too.
I would suspect a flaw in the design process that lead to the design of that part of the rocket.
SpaceX says they've looked at all the obvious explanations, but they have probably only had time to look at the known obvious explanations. There could be 'unknown obvious' explanations, explanations that are not obvious because the engineers lack a full understanding of the rocket. Explanations are only obvious if you fully understand the thing that you're trying to explain.
The rifle theory is incredibly unlikely when you consider that new rocket designs (like the Falcon 9) tend to fail every now and then. So far it has always been because of some problem with the rocket or with the processes surrounding its use.
Like someone wrote on the SpaceX Reddit: SpaceX probably only wanted to go up on the roof to look for debris from the rocket in case any of it happened to land on the roof.
IIRC the previous malfunction was caused by a faulty strut attached to one of the helium tanks. This would be the second time that that part of the rocket malfunctioned.
It sounds like the Falcon 9 helium tanks need a lot more analysis and testing.
I would not put my money on the rifle theory, not unless they have found bullet fragments or suspicious-looking damage.
Well the answer to your question is that Type.NULL is something that you define yourself or read about in the docs/code if it is coming from a dependency and use instead of NULL.
There are times when having your software lie to you is better than having it do nothing. That's when you use Type.NULL.
There are also times when having the software crash is better than having it lie. That's what NULL is for.
Of course, if you and everyone on your team always writes correct code then it doesn't matter how you encode non-existent values. Your code will neither crash nor lie. The problem is that writing correct code is sometimes hard and often takes a lot of time.
And what do you do when something doesn't have a value? A new employee has been hired but you don't know his birthday? Plug in 1900-01-01? And check for that? Nulls serve a purpose.
Pretty much, yeah.
Using Type.NULL instead of NULL lets your code fail partially or quantitatively instead of failing completely. Perhaps the value Date.NULL will suffice for the birth date for most use cases. Perhaps it's okay if your system thinks you're employing a bunch of 116-year olds.
The only fatal flaw with this scheme is that it fails as soon as you introduce inheritance, because Person.NULL != Employee.NULL, even though logically speaking the absence of a Person implies the absence of an Employee.
The flat head nails that your hammer interacts with have an interface that is both sufficiently good and sufficiently difficult to reinvent that even the most creative of people and large of firms have been unable to replace it with something incompatible. (I'm sure Apple will eventually move into the carpentry business and invent a new type of nail that requires a special $59.95 Apple hammer and $4.95 a piece Apple nails. But that hasn't happened yet.)
So in order to make software with longevity you simply need to make sure that your application's components speak with one another and the surrounding world through API:s that are really good and really difficult to reinvent. Like SOAP over HTTP.
There are flaws in every sufficiently complex piece of technology, and no, you can't just magically find them all by looking Real Hard.
You can find a lot of the ones that are most likely to cause a complete failure.
My understanding is that they could have found last year's problem by doing a sufficient amount of parts testing to verify the numbers on the data sheets from their suppliers. It would have cost a lot of money to do that, but so does losing a payload and having to suspend launches for months.
Yeah, but the presence of two flaws (the faulty strut last year and now this unknown flaw) indicate that there could be many more lurking in the design of the Falcon 9. SpaceX is going to want to go hunting for flaws everywhere in the design before they fly the F9 again.
Imagine how much market confidence SpaceX would lose if there were to be a third explosion within the next couple of years.
SpaceX states that it blew up in preparation for the static fire test.
Looks like an inherent problem with the rocket or the procedures leading up to the static test, which I assume are basically the same as those leading up to launch.
I'm afraid that this will set SpaceX back by months while they figure out what's wrong with their systems and procedures. Well, unexpected setbacks are to be expected in the space industry.
We're almost past the point of whether or not it works and moving onto why it works.
Yeah. Don't get your hopes up.
Based on many, many prior instances of machines with measurable effects just above experimental noise, that point of 'almost but not quite' being able to demonstrate that it works is where the EmDrive will remain.
The combination of a tiny effect size (or signal to noise ratio) a lack of a prior theoretical predictions is a huge warning sign. New exciting science usually comes either with a strong unexpected effect size or with a prior theoretical prediction.
Yes, but it may turn out that a large percentage of companies have needs that overlap. I think it's safe to say that wrapping databases in UI is not going to keep being a $100k+ skill for much longer.
Nobody cares if it'll last for decades. Most apps are not expected to deliver useful value at all by the people who order them. It's really only after the app has proven to be useful and profitable that the suits begin to think about longevity and scalability.
Now, repairing poorly made applications... That's going to keep being lucrative for decades to come, or at least until the government gets sick of poorly made applications and decides to impose 'building codes' for software.
Google is focusing a lot now on the "next 5 billion", which I guess is basically people in households that make $2000-20000 a year.
You can easily see how having a smartphone with multiple user accounts coupled with a cheap wireless keyboard and a cheap monitor could make a big difference for a family in that income range.
With solid state storage you can cram in effectively arbitrarily large amounts of storage in a tiny amount of space, so that will not be a problem with a thinner laptop. But it doesn't matter. Battery capacity will be lower, which means it won't be a professional laptop, which means it won't really need terrabytes of storage.
It sounds like the next MacBook Pro will basically be a better and slightly larger MacBook Air. That could be a smart short-term move by Apple, but it will kill them in the long run when their professional users jump ship and buy computers from Dell.
I'm glad Dell is still serious about building pro segment laptops. Samsung is probably going to want a bigger piece of the action too if Apple moves out of the pro space.
Older people have more patience. They are as unhappy to put up with "shit" as the next guy but they have patience that comes with experience.
It seems to me that one's impatience ought to grow with age. The time that something takes in proportion to your remaining life expectancy grows larger the older you get.
I don't think older people are more patient. I think the patience that old people appear to have is really lower and more realistic expectations.
50% change of loosing if 2008 was an average year...
It's probably closer to a 51% chance of winning and 49% chance of losing.
A much safer and more scientifically relevant bet would be that the average temperature of the entire 2010's will be at least a half degree Celsius higher than the average temperature of the 1910's.
Blender is a nice name. Inkscape is a really nice name.
But "GIMP"? WTF?
It is an ill-chosen name for english-speaking users.
Good thing it's free software. It should be fairly simple to do a fork of every major GIMP release that only changes the name and some of the graphics and leaves everything else as-is. It's not like it would be hard to come up with a name and some graphics resources...
I guess the name isn't such a big problem after all.
I think AR apps and games where you keep your screen on for extended periods of time are going to put an end to the thinner phone trend. At least for a while until they figure out a way to get radically more screen time per unit of energy.
The long term answer to AR usage patters will probably be screens that will switch to a low power reflective mode for use in sunlight and other high ambient light situations.
Yeah, Nintendo has seen that their audience has gone over to mobile gaming. There is probably no point in trying to fight that.
By the way Nintendo and Niantic may finally have found a way to prompt Apple and Samsung to put bigger batteries in their phones. Games like Pokemon GO where you keep your screen on will suck the battery dry in 2-3 hours on most phones.
Weren't these exact specs going through the console gaming news cycle some 2-3 months ago? I believe this is the OFA: http://www.giantbomb.com/artic...
But yes, this makes a lot of sense for Sony. It is not going to be technologically feasible to release a gaming system with at least 10x more GPU power than the original PS4 within the foreseeable future. Certainly not at a reasonable price point. It is feasible to do a 2x upgrade next year because the PS4 was somewhat underpowered to begin with. Then there could easily be another 2x upgrade sometime in the early 2020's.
(Yes, we are less than three and a half years away from the early 2020's. Sounds a little implausible, doesn't it?)
I think the PS4 series of consoles is likely to be Sony's final non-portable gaming system. One of Nvidia, Intel and AMD will probably be the company that pioneers the truly next generation of gaming consoles, which will require truly radical change in how GPU:s are made.
This has already been tried. The results have not exactly been encouraging.
The problem is that even the best experts in any field know very little. The theories that exist probably cover something like 0.1% of everything that we need to understand in order to make informed decisions that would optimise global happiness. And that's assuming we can agree that Utilitarianism makes sense. It's a complete non-starter. And it will have adverse effects when people stubbornly refuse to admit that it doesn't work.
Let's take one example. You are clearly going to need a science of history in order to predict what effect your policies will have in the long run. But historians have virtually nothing in terms of over-arching theory that can guide us.
The "solution" that people have come up with in the past is to simply make theories up out of thin air and make it obligatory by law to believe in them.
Not exactly. Physicists and laymen alike have been confused about the weirder aspects of quantum mechanics since long before the internet.
I remember silently refusing to believe that my middle school science teacher was telling the truth about the double slit experiment. It seemed obviously wrong. Some searching on Altavista changed (and blew) my mind.
This is admittedly interesting news, but what I'm really dying to know is what Kim Kardashian has to say about Pokemon Go.
I'd also like to know more about the top 10 weirdest things that have happened to people while playing Pokemon Go and how to easily make hundreds of dollars a week by playing Pokemon Go.
They're claiming that's not enough. What they're claiming implicitly is that your great-grandchildren have to be able to make money off of your work, otherwise there will be too little incentive for you to innovate.
No, actually what they're claiming is that they have enough lobbyist to enact whatever the hell they want for themselves.
And by innovation they mean clever ways to sell the exact same old recording that you already own, maybe with some slight changes to the mastering or what not.
Slashdot Mirror
My understanding is that no team of engineers have ever had a fully satisfactory understanding of a space launch system. Space launch designs have such tiny margins for error that some subtle phenomenon going on with the design could be serious enough to destroy the rocket.
The gunshot-sounding pop could easily have been the sound of the first thing to fail inside the rocket. That sound (and any other sounds from the launch pad) could have echoed off of the building. People are notoriously bad at recalling the sequence of events, so it's possible that a person could report hearing the echo before the sound itself.
The problem with the gunshot theory is that it has an incredibly low a priory probability. The ULA has a lot to lose by trying to sabotage their competitors in a blatantly illegal way. It's the last thing they would do. It could have been an employee acting on their own (perhaps someone with mental health problems), but what are the odds that the ULA has such an employee combined with the odds that that employee could have successfully smuggled a rifle all the way to the top of the building? Close to zero.
The gunshot theory only becomes probable if they find hard evidence, like bullet fragments, or tell-tale damage to the wreckage.
Yeah but the point is that the strut that failed in 2015 was directly attached to the helium tank that failed last month. There is literally a strong connection between the two. Perhaps there's a logical connection too.
I would suspect a flaw in the design process that lead to the design of that part of the rocket.
SpaceX says they've looked at all the obvious explanations, but they have probably only had time to look at the known obvious explanations. There could be 'unknown obvious' explanations, explanations that are not obvious because the engineers lack a full understanding of the rocket. Explanations are only obvious if you fully understand the thing that you're trying to explain.
The rifle theory is incredibly unlikely when you consider that new rocket designs (like the Falcon 9) tend to fail every now and then. So far it has always been because of some problem with the rocket or with the processes surrounding its use.
Like someone wrote on the SpaceX Reddit: SpaceX probably only wanted to go up on the roof to look for debris from the rocket in case any of it happened to land on the roof.
IIRC the previous malfunction was caused by a faulty strut attached to one of the helium tanks. This would be the second time that that part of the rocket malfunctioned.
It sounds like the Falcon 9 helium tanks need a lot more analysis and testing.
I would not put my money on the rifle theory, not unless they have found bullet fragments or suspicious-looking damage.
Well the answer to your question is that Type.NULL is something that you define yourself or read about in the docs/code if it is coming from a dependency and use instead of NULL.
There are times when having your software lie to you is better than having it do nothing. That's when you use Type.NULL.
There are also times when having the software crash is better than having it lie. That's what NULL is for.
Of course, if you and everyone on your team always writes correct code then it doesn't matter how you encode non-existent values. Your code will neither crash nor lie. The problem is that writing correct code is sometimes hard and often takes a lot of time.
And what do you do when something doesn't have a value? A new employee has been hired but you don't know his birthday? Plug in 1900-01-01? And check for that? Nulls serve a purpose.
Pretty much, yeah.
Using Type.NULL instead of NULL lets your code fail partially or quantitatively instead of failing completely. Perhaps the value Date.NULL will suffice for the birth date for most use cases. Perhaps it's okay if your system thinks you're employing a bunch of 116-year olds.
The only fatal flaw with this scheme is that it fails as soon as you introduce inheritance, because Person.NULL != Employee.NULL, even though logically speaking the absence of a Person implies the absence of an Employee.
The flat head nails that your hammer interacts with have an interface that is both sufficiently good and sufficiently difficult to reinvent that even the most creative of people and large of firms have been unable to replace it with something incompatible. (I'm sure Apple will eventually move into the carpentry business and invent a new type of nail that requires a special $59.95 Apple hammer and $4.95 a piece Apple nails. But that hasn't happened yet.)
So in order to make software with longevity you simply need to make sure that your application's components speak with one another and the surrounding world through API:s that are really good and really difficult to reinvent. Like SOAP over HTTP.
There are flaws in every sufficiently complex piece of technology, and no, you can't just magically find them all by looking Real Hard.
You can find a lot of the ones that are most likely to cause a complete failure.
My understanding is that they could have found last year's problem by doing a sufficient amount of parts testing to verify the numbers on the data sheets from their suppliers. It would have cost a lot of money to do that, but so does losing a payload and having to suspend launches for months.
Yeah, but the presence of two flaws (the faulty strut last year and now this unknown flaw) indicate that there could be many more lurking in the design of the Falcon 9. SpaceX is going to want to go hunting for flaws everywhere in the design before they fly the F9 again.
Imagine how much market confidence SpaceX would lose if there were to be a third explosion within the next couple of years.
SpaceX states that it blew up in preparation for the static fire test.
Looks like an inherent problem with the rocket or the procedures leading up to the static test, which I assume are basically the same as those leading up to launch.
I'm afraid that this will set SpaceX back by months while they figure out what's wrong with their systems and procedures. Well, unexpected setbacks are to be expected in the space industry.
We're almost past the point of whether or not it works and moving onto why it works.
Yeah. Don't get your hopes up.
Based on many, many prior instances of machines with measurable effects just above experimental noise, that point of 'almost but not quite' being able to demonstrate that it works is where the EmDrive will remain.
The combination of a tiny effect size (or signal to noise ratio) a lack of a prior theoretical predictions is a huge warning sign. New exciting science usually comes either with a strong unexpected effect size or with a prior theoretical prediction.
I missed that...
I do find it funny that I keep misreading EmDrive as EmDrivel.
Yes, but it may turn out that a large percentage of companies have needs that overlap. I think it's safe to say that wrapping databases in UI is not going to keep being a $100k+ skill for much longer.
Nobody cares if it'll last for decades. Most apps are not expected to deliver useful value at all by the people who order them. It's really only after the app has proven to be useful and profitable that the suits begin to think about longevity and scalability.
Now, repairing poorly made applications... That's going to keep being lucrative for decades to come, or at least until the government gets sick of poorly made applications and decides to impose 'building codes' for software.
Google is focusing a lot now on the "next 5 billion", which I guess is basically people in households that make $2000-20000 a year.
You can easily see how having a smartphone with multiple user accounts coupled with a cheap wireless keyboard and a cheap monitor could make a big difference for a family in that income range.
With solid state storage you can cram in effectively arbitrarily large amounts of storage in a tiny amount of space, so that will not be a problem with a thinner laptop. But it doesn't matter. Battery capacity will be lower, which means it won't be a professional laptop, which means it won't really need terrabytes of storage.
It sounds like the next MacBook Pro will basically be a better and slightly larger MacBook Air. That could be a smart short-term move by Apple, but it will kill them in the long run when their professional users jump ship and buy computers from Dell.
I'm glad Dell is still serious about building pro segment laptops. Samsung is probably going to want a bigger piece of the action too if Apple moves out of the pro space.
Older people have more patience. They are as unhappy to put up with "shit" as the next guy but they have patience that comes with experience.
It seems to me that one's impatience ought to grow with age. The time that something takes in proportion to your remaining life expectancy grows larger the older you get.
I don't think older people are more patient. I think the patience that old people appear to have is really lower and more realistic expectations.
50% change of loosing if 2008 was an average year...
It's probably closer to a 51% chance of winning and 49% chance of losing.
A much safer and more scientifically relevant bet would be that the average temperature of the entire 2010's will be at least a half degree Celsius higher than the average temperature of the 1910's.
Blender is a nice name. Inkscape is a really nice name.
But "GIMP"? WTF?
It is an ill-chosen name for english-speaking users.
Good thing it's free software. It should be fairly simple to do a fork of every major GIMP release that only changes the name and some of the graphics and leaves everything else as-is. It's not like it would be hard to come up with a name and some graphics resources...
I guess the name isn't such a big problem after all.
I think AR apps and games where you keep your screen on for extended periods of time are going to put an end to the thinner phone trend. At least for a while until they figure out a way to get radically more screen time per unit of energy.
The long term answer to AR usage patters will probably be screens that will switch to a low power reflective mode for use in sunlight and other high ambient light situations.
Yeah, Nintendo has seen that their audience has gone over to mobile gaming. There is probably no point in trying to fight that.
By the way Nintendo and Niantic may finally have found a way to prompt Apple and Samsung to put bigger batteries in their phones. Games like Pokemon GO where you keep your screen on will suck the battery dry in 2-3 hours on most phones.
Intel is not remotely interested in gaming consoles, and would never partner with AMD anyway.
Intel is no doubt in search of a new area of focus. So that could change.
Intel will probably not want to license technology from AMD, but they might at some point buy AMD and chop it up for parts.
Weren't these exact specs going through the console gaming news cycle some 2-3 months ago? I believe this is the OFA: http://www.giantbomb.com/artic...
But yes, this makes a lot of sense for Sony. It is not going to be technologically feasible to release a gaming system with at least 10x more GPU power than the original PS4 within the foreseeable future. Certainly not at a reasonable price point. It is feasible to do a 2x upgrade next year because the PS4 was somewhat underpowered to begin with. Then there could easily be another 2x upgrade sometime in the early 2020's.
(Yes, we are less than three and a half years away from the early 2020's. Sounds a little implausible, doesn't it?)
I think the PS4 series of consoles is likely to be Sony's final non-portable gaming system. One of Nvidia, Intel and AMD will probably be the company that pioneers the truly next generation of gaming consoles, which will require truly radical change in how GPU:s are made.
This has already been tried. The results have not exactly been encouraging.
The problem is that even the best experts in any field know very little. The theories that exist probably cover something like 0.1% of everything that we need to understand in order to make informed decisions that would optimise global happiness. And that's assuming we can agree that Utilitarianism makes sense. It's a complete non-starter. And it will have adverse effects when people stubbornly refuse to admit that it doesn't work.
Let's take one example. You are clearly going to need a science of history in order to predict what effect your policies will have in the long run. But historians have virtually nothing in terms of over-arching theory that can guide us.
The "solution" that people have come up with in the past is to simply make theories up out of thin air and make it obligatory by law to believe in them.
Not exactly. Physicists and laymen alike have been confused about the weirder aspects of quantum mechanics since long before the internet.
I remember silently refusing to believe that my middle school science teacher was telling the truth about the double slit experiment. It seemed obviously wrong. Some searching on Altavista changed (and blew) my mind.
This is admittedly interesting news, but what I'm really dying to know is what Kim Kardashian has to say about Pokemon Go.
I'd also like to know more about the top 10 weirdest things that have happened to people while playing Pokemon Go and how to easily make hundreds of dollars a week by playing Pokemon Go.
They're claiming that's not enough. What they're claiming implicitly is that your great-grandchildren have to be able to make money off of your work, otherwise there will be too little incentive for you to innovate.
No, actually what they're claiming is that they have enough lobbyist to enact whatever the hell they want for themselves.
And by innovation they mean clever ways to sell the exact same old recording that you already own, maybe with some slight changes to the mastering or what not.