Fragmentation typically isn't an issue anymore because Windows will defrag its own drives daily by default. Fragmentation is also irrelevant if you've got an SSD, which I'd hope most people have for their primary drive nowadays. Your registry only grows indefinitely if you're constantly adding new software AND it doesn't clean up after itself properly, something that's a bit less common than it used to now that most programs use the standard Windows Installer libraries.
Microsoft Windows runtime requirements actually haven't increased significantly since Microsoft Vista, which was a *big* jump in hardware requirements (and they significantly understated the minimum requirements as well). MS has actually done a good job at keeping the runtime requirements fairly constant the last few releases, because we haven't seen the constant increase in CPU hardware speed like we used to. If you've got a machine that ran Vista reasonably well, it will probably run Windows 10 as well.
The second issue looks more reasonable to me, as I've seen this happen with terrible AV products and underpowered machines at work before as well. Frankly, the scans should NOT be taking place during the day, as long as people's machines are left on and connected. The whole drive encryption is likely not the issue, as that has much lower overhead than most people expect - it's the security software constantly thrashing the hard drive. If you get two competing programs trying to access the hard drive, access tends to slow down exponentially, because the drive head now has to constantly swap between two different points on the drive to service multiple requests. IT is correct in that SSDs would make a huge difference here, as they don't have this issue.
Those hundreds of MB of security patches are just the same binaries that already exist on your system, simply modified and re-compiled to patch security vulnerabilities. They don't appreciably change the runtime requirement of the OS. The *disk space* requirements ARE affected, but only because Windows retains the older copies of the systems files via it's "snapshot" system so you can roll back if needed. Keep in mind that this has NO effect on the runtime requirements of the OS.
Heh, you don't have to tell me how painful short falls can be, as I've also had my own rather nasty injury at a very close proximity to the ground (albeit down a couple of steps).
The point is that that anything *above* a few meters - which is where you're likely to be flying the vast majority of the time - should be much safer, since the parachute should deploy at those higher altitudes. We'll have to see what the safety limit actually is in practice, but again, this is like seatbelts and airbags for cars. They're not going to save everyone, but it's a hell of a lot better than not having them.
They're opposing the building of a modern institution of science and learning for the sake of "sacred land". I can't think of a better way to describe this than "ignorance". Or, they could admit this is a political issue of the separatist movement, and not really a cultural one.
The Hawaiians had their nation stolen from them about five generations ago. US citizens overthrew the Hawaiian constitutional monarchy, then the US annexed the territory. I feel bad about, and even the US government has apologized at this point. Yes, I have some sympathy with those who feel disaffected because of this, but after five generations... sorry, we're not going to hand the state back to the native Hawaiians.
Just watched the video, and Martin reps are claiming the parachute system will start functioning in just a few meters. As Ellis mentioned below, this obviously must mean they're ejecting and inflating the parachute via some sort of mortar, and as such, it probably starts working almost instantly.
I wouldn't be surprised if the parachute was automatically triggered, as it's pretty simple to detect freefall acceleration or a catastrophic engine failure, although this is just my sceculation. So, in this case, human reaction time is also not a factor, which would be significant at low flight levels.
Also, that thing can fly at 1km in the air and ascend faster than a typical helicopter. It's actually more of a personal VTOL aircraft than a jetpack IMO.
Sorry, I need to correct myself. I typed "turbofan", which is not what the summary or article said. A turbofan is a type of jet, of course, but I believe these are simply using normal ducted fans. So I guess it's not actually a jet, strictly speaking.
I think it's a lot more accurate to call this a single-person VTOL aircraft than a "jetpack", but that doesn't sound nearly as sexy I guess.
It's a little sad to see people fighting so hard for the cause of ignorance... not that it's so rare, I guess. It's almost more depressing if you consider that some of them are probably sincere, instead of simply wanting either payola or publicity.
Design by committee often fails, in my experience, if you have a group of equals on the committee and no one is empowered to make the final decision, or is not willing to make the final decision when it needs to be made.
In this sort of programming environment, I'd think it would be critical to have at least one person with the authority to make a final decision on any matters of disagreement or contention - likely the senior programmer when it comes to technical issues, or the feature/product owner when it comes to decisions related to functionality. Likewise, when part of a group dynamic, members need to learn the skill of knowing which battles are important enough to fight, and when it's best to just shrug and disagree. Otherwise, nothing would ever get done.
Keep in mind that human beings are pretty good at creating a shared mental context of their conversation, and programmers would also have a lot of shared knowledge, both of the problem at hand as well as more generalized knowledge that most programmers (should) know.
For instance, if I said "We should use a proxy object here to create a streamlined interface and minimize coupling between these various systems"... how much information did that convey? If you're a programmer, that probably said a quite a bit in a very few words, because you understand a deeper meaning behind all those words.
I guess you can consider shared knowledge to be a highly effective form of compression for personally-transmitted information (spoken communication is only part of it, remember).
I'm guessing it works for solving some sorts of problems fairly well. But there are some problems I've run into that require some silent contemplation over quite a bit of time to come up with a solution. I have a hard time envisioning that working with pair programming (which I've done only in very brief amounts) or with mob programming (never tried, probably like most). There are also some problems so technically difficult that I need maximum concentration to keep everything straight while implementing or debugging it. Having a group of programmers surrounding me seems distracting to that end.
I could see how it might help in some situations... you're essentially programming while having a constant design and review meeting, so I can see how the quality of code would improve. You're unlikely to simply accept a sub-par solution, because you've got a couple other programmers to readily suggest solutions you haven't thought of yet. The fact that it improves quality but not productivity should really come as no surprise, as you're essentially multiplying the brain-power focused on a single problem, but five programmers can't necessarily solve a problem five times faster.
An interesting concept. I don't think I'd want to *always* program that way (nor pair programming), but I could see it being helpful at times. At the moment, I either work on my own projects or as a remote contract programmer, so I'm largely in the position of *having* to solve everything myself, and it's often fairly difficult to not have immediate access to other programmers for advice or assistance.
Specifically designed low-altitude parachutes are effective at a few hundred feet. No, it won't save you at extreme low altitudes and velocities, but it's certainly better than nothing. The phrase "Perfect is the enemy of the good" comes to mind.
...It is the landing on the rock hard ground I'm concerned about.
From the article:
along with a low-altitude parachute for use should things go wrong.
I wouldn't use it without a parachute either. With an emergency parachute... um.... probably. After a few other brave souls try it first.
Also, 30 minutes is waaay better than the versions we've seen previously, which could only operate for a few minutes at a time. And... I guess we're still calling it a "jetpack" even though it's just using turbofans? I guess there's no other commonly-known term to describe it?
I'm guessing "personality test" simply means you're trying to screen people out who have horrible personalities that won't work well with others. If you've had the misfortune of working with people like that before, you might understand the reasoning behind that sort of test.
I can't personally conceive of how such a test might work, short of just getting a few members of the team together to chat casually with the interviewee for a while, and even then, most antisocial people probably know how to behave properly for a short time.
I left Foxit behind when they started pushing crapware installs, and more critically, when it had some problems rendering some fairly basic PDFs correctly. Back to Adobe Reader for me as well.
It's like that with MS Word docs as well. The damn things are so complicated that only the original code has a prayer of rendering it correctly, and even then not always.
I'm assuming "mobile home" is meant by "trailer". What's the harm? Unless you own the land as well, owning a trailer is risky. The land owner can, with a year's notice I believe, sell your land to some other developer and force you to move your trailer. Often, the expensive of disconnecting and moving the trailer is so high (especially if it's not all that structurally sound anymore), the owner is basically forced to simply dispose of it at their expense. This just happened in an area near me, and was a story in the local media.
Actually, I need to correct myself. At least in the VC implementation, the shared_ptr object itself has both a pointer to the allocated object as well as the ref-counted block (which has a counter plus a pointer to the allocated object). So that means it's zero overhead when you dereference it - that is, no more overhead than a raw pointer.
If you use std::make_shared, there's a good chance the library allocates both the object and the extra block for the shared_ptr internal data in a single allocation. This not only improves allocation performance, but increases cache coherency as well. There's even ways to use custom block allocators to further improve small object allocation as well as data coherency.
Honestly, though, if you're using a pointer, you're probably already causing cache misses (unless you're using highly specialized allocators), so just live with it and use *safe* pointers. The extra indirection won't make much a difference. It doesn't really matter if you make one jump or two - you're pretty much guaranteed to jump out of the cache anyhow. If you really need cache-friendly data structures, you need to stuff all your data into a continuous chunk of memory (i.e. std::vector) and avoid pointer indirection completely.
The fact is, the majority of code is not performance-sensitive enough to be worrying about this. You're far better off focusing on broader algorithmic optimizations. Even in videogame programming, we don't typically worry about that sort of thing except for a very small percentage of code, typically fairly deep in the engine, such as graphics primitive processing, or in animation code... stuff like that. There's no reason shared pointers can't be used extensively throughout your codebase, and then just save the low-level optimizations for your hotspots.
So, someone decides they're ok with working longer hours for more money and you conclude that this is so bizarre that they must be a troll? Hell, there's a name for it: a workaholic. Or maybe he just likes the extra money that he brings in when choosing to work at jobs that demand longer hours. It's not all that unusual for people to enjoy making more money. Maybe he's young and single, or doesn't have kids, and so can choose to spend his free time earning money. Who knows?
No, it's not a choice I'd make myself, but I fail to see how this attitude is so unusual...
Honestly, that does sound more plausible than the hilarious lies we've heard. Do you happen to have a source for this? Or are you fluent enough in Korean to read the original transcripts?
BTW, the "unicorns" were actually kirin, a horned animal with parts of various creatures making up the body. This creature is said to accompany wise rulers (no coincidence there, eh?).
Shame on all mods who downmodded dgatwood's post redundant or overrated (his post currently stands at 0 Redundant). I disagreed completely with his conclusions, but he presented them clearly, politely, and intelligently. Moderation is not supposed to be a "I disagree with your opinion, and therefore I'm going to silence you" button. You continue to do that, and you'll simply be turning/. into an echo-chamber (well, more than it already is) where no one dares disagree on any substantial issue.
But, you ever notice how few 3-star reviews there are on Amazon for most products?
I'm not sure I completely agree, but I supposed it depends on the product. Let's take a peek at a technical book I was looking at:
It has 25 reviews, with an (algorithmically weighted) average of 3.8 stars. The break down is: 5 star 44% 4 star 24% 3 star 12% 2 star 12% 1 star 8%
That seems reasonable enough to me. My takeaway is that the book is probably solid, but has a few serious flaws. The people who gave three stars, interestingly enough, seemed to give the best critical feedback while also pointing out the positive aspects of the book.
The problem you mention is even worse when you look at user videogame reviews. For instance, on metacritic, you use user ratings of 1/10 (ZOMG worst game EVAR!!!!!) or 10/10 (WOWOW bestest game in history!!!!), and surprisingly little in-between.
Steam did away with this and simply gave users a choice of thumbs up or thumbs down, which I think helps to eliminate that ridiculousness of extremes, although it has some drawbacks as well. Many users can't be trusted to give a reasonably unbiased review of a product they feel passionately about. Others disagree wildly about what various degrees of a rating mean (I've seen one person say they never give five star ratings because no product is perfect, which elicited a *facepalm* from me). And what's the point of a 10 point scale if most games never score below a 6/10 (or likewise with a 100 point scale)?
Slashdot Mirror
Fragmentation typically isn't an issue anymore because Windows will defrag its own drives daily by default. Fragmentation is also irrelevant if you've got an SSD, which I'd hope most people have for their primary drive nowadays. Your registry only grows indefinitely if you're constantly adding new software AND it doesn't clean up after itself properly, something that's a bit less common than it used to now that most programs use the standard Windows Installer libraries.
Microsoft Windows runtime requirements actually haven't increased significantly since Microsoft Vista, which was a *big* jump in hardware requirements (and they significantly understated the minimum requirements as well). MS has actually done a good job at keeping the runtime requirements fairly constant the last few releases, because we haven't seen the constant increase in CPU hardware speed like we used to. If you've got a machine that ran Vista reasonably well, it will probably run Windows 10 as well.
The second issue looks more reasonable to me, as I've seen this happen with terrible AV products and underpowered machines at work before as well. Frankly, the scans should NOT be taking place during the day, as long as people's machines are left on and connected. The whole drive encryption is likely not the issue, as that has much lower overhead than most people expect - it's the security software constantly thrashing the hard drive. If you get two competing programs trying to access the hard drive, access tends to slow down exponentially, because the drive head now has to constantly swap between two different points on the drive to service multiple requests. IT is correct in that SSDs would make a huge difference here, as they don't have this issue.
Those hundreds of MB of security patches are just the same binaries that already exist on your system, simply modified and re-compiled to patch security vulnerabilities. They don't appreciably change the runtime requirement of the OS. The *disk space* requirements ARE affected, but only because Windows retains the older copies of the systems files via it's "snapshot" system so you can roll back if needed. Keep in mind that this has NO effect on the runtime requirements of the OS.
Heh, you don't have to tell me how painful short falls can be, as I've also had my own rather nasty injury at a very close proximity to the ground (albeit down a couple of steps).
The point is that that anything *above* a few meters - which is where you're likely to be flying the vast majority of the time - should be much safer, since the parachute should deploy at those higher altitudes. We'll have to see what the safety limit actually is in practice, but again, this is like seatbelts and airbags for cars. They're not going to save everyone, but it's a hell of a lot better than not having them.
They're opposing the building of a modern institution of science and learning for the sake of "sacred land". I can't think of a better way to describe this than "ignorance". Or, they could admit this is a political issue of the separatist movement, and not really a cultural one.
The Hawaiians had their nation stolen from them about five generations ago. US citizens overthrew the Hawaiian constitutional monarchy, then the US annexed the territory. I feel bad about, and even the US government has apologized at this point. Yes, I have some sympathy with those who feel disaffected because of this, but after five generations... sorry, we're not going to hand the state back to the native Hawaiians.
Just watched the video, and Martin reps are claiming the parachute system will start functioning in just a few meters. As Ellis mentioned below, this obviously must mean they're ejecting and inflating the parachute via some sort of mortar, and as such, it probably starts working almost instantly.
I wouldn't be surprised if the parachute was automatically triggered, as it's pretty simple to detect freefall acceleration or a catastrophic engine failure, although this is just my sceculation. So, in this case, human reaction time is also not a factor, which would be significant at low flight levels.
Also, that thing can fly at 1km in the air and ascend faster than a typical helicopter. It's actually more of a personal VTOL aircraft than a jetpack IMO.
Sorry, I need to correct myself. I typed "turbofan", which is not what the summary or article said. A turbofan is a type of jet, of course, but I believe these are simply using normal ducted fans. So I guess it's not actually a jet, strictly speaking.
I think it's a lot more accurate to call this a single-person VTOL aircraft than a "jetpack", but that doesn't sound nearly as sexy I guess.
It's a little sad to see people fighting so hard for the cause of ignorance... not that it's so rare, I guess. It's almost more depressing if you consider that some of them are probably sincere, instead of simply wanting either payola or publicity.
Design by committee often fails, in my experience, if you have a group of equals on the committee and no one is empowered to make the final decision, or is not willing to make the final decision when it needs to be made.
In this sort of programming environment, I'd think it would be critical to have at least one person with the authority to make a final decision on any matters of disagreement or contention - likely the senior programmer when it comes to technical issues, or the feature/product owner when it comes to decisions related to functionality. Likewise, when part of a group dynamic, members need to learn the skill of knowing which battles are important enough to fight, and when it's best to just shrug and disagree. Otherwise, nothing would ever get done.
Keep in mind that human beings are pretty good at creating a shared mental context of their conversation, and programmers would also have a lot of shared knowledge, both of the problem at hand as well as more generalized knowledge that most programmers (should) know.
For instance, if I said "We should use a proxy object here to create a streamlined interface and minimize coupling between these various systems"... how much information did that convey? If you're a programmer, that probably said a quite a bit in a very few words, because you understand a deeper meaning behind all those words.
I guess you can consider shared knowledge to be a highly effective form of compression for personally-transmitted information (spoken communication is only part of it, remember).
I'm guessing it works for solving some sorts of problems fairly well. But there are some problems I've run into that require some silent contemplation over quite a bit of time to come up with a solution. I have a hard time envisioning that working with pair programming (which I've done only in very brief amounts) or with mob programming (never tried, probably like most). There are also some problems so technically difficult that I need maximum concentration to keep everything straight while implementing or debugging it. Having a group of programmers surrounding me seems distracting to that end.
I could see how it might help in some situations... you're essentially programming while having a constant design and review meeting, so I can see how the quality of code would improve. You're unlikely to simply accept a sub-par solution, because you've got a couple other programmers to readily suggest solutions you haven't thought of yet. The fact that it improves quality but not productivity should really come as no surprise, as you're essentially multiplying the brain-power focused on a single problem, but five programmers can't necessarily solve a problem five times faster.
An interesting concept. I don't think I'd want to *always* program that way (nor pair programming), but I could see it being helpful at times. At the moment, I either work on my own projects or as a remote contract programmer, so I'm largely in the position of *having* to solve everything myself, and it's often fairly difficult to not have immediate access to other programmers for advice or assistance.
Specifically designed low-altitude parachutes are effective at a few hundred feet. No, it won't save you at extreme low altitudes and velocities, but it's certainly better than nothing. The phrase "Perfect is the enemy of the good" comes to mind.
...It is the landing on the rock hard ground I'm concerned about.
From the article:
along with a low-altitude parachute for use should things go wrong.
I wouldn't use it without a parachute either. With an emergency parachute... um.... probably. After a few other brave souls try it first.
Also, 30 minutes is waaay better than the versions we've seen previously, which could only operate for a few minutes at a time. And... I guess we're still calling it a "jetpack" even though it's just using turbofans? I guess there's no other commonly-known term to describe it?
And for bonus points, like Disney or Edison, have your old white guys being laid off train your brand new brown imported slaves.
Are companies hiring H1B's in droves or are they discriminating against brown-skinned people with accents?
I'm guessing "personality test" simply means you're trying to screen people out who have horrible personalities that won't work well with others. If you've had the misfortune of working with people like that before, you might understand the reasoning behind that sort of test.
I can't personally conceive of how such a test might work, short of just getting a few members of the team together to chat casually with the interviewee for a while, and even then, most antisocial people probably know how to behave properly for a short time.
The Year of the Lexus Hoverboard!
Yeah, of course, not a real, practical hoverboard, but a pretty cool gimmick nonetheless.
I left Foxit behind when they started pushing crapware installs, and more critically, when it had some problems rendering some fairly basic PDFs correctly. Back to Adobe Reader for me as well.
It's like that with MS Word docs as well. The damn things are so complicated that only the original code has a prayer of rendering it correctly, and even then not always.
I'm assuming "mobile home" is meant by "trailer". What's the harm? Unless you own the land as well, owning a trailer is risky. The land owner can, with a year's notice I believe, sell your land to some other developer and force you to move your trailer. Often, the expensive of disconnecting and moving the trailer is so high (especially if it's not all that structurally sound anymore), the owner is basically forced to simply dispose of it at their expense. This just happened in an area near me, and was a story in the local media.
Actually, I need to correct myself. At least in the VC implementation, the shared_ptr object itself has both a pointer to the allocated object as well as the ref-counted block (which has a counter plus a pointer to the allocated object). So that means it's zero overhead when you dereference it - that is, no more overhead than a raw pointer.
If you use std::make_shared, there's a good chance the library allocates both the object and the extra block for the shared_ptr internal data in a single allocation. This not only improves allocation performance, but increases cache coherency as well. There's even ways to use custom block allocators to further improve small object allocation as well as data coherency.
Honestly, though, if you're using a pointer, you're probably already causing cache misses (unless you're using highly specialized allocators), so just live with it and use *safe* pointers. The extra indirection won't make much a difference. It doesn't really matter if you make one jump or two - you're pretty much guaranteed to jump out of the cache anyhow. If you really need cache-friendly data structures, you need to stuff all your data into a continuous chunk of memory (i.e. std::vector) and avoid pointer indirection completely.
The fact is, the majority of code is not performance-sensitive enough to be worrying about this. You're far better off focusing on broader algorithmic optimizations. Even in videogame programming, we don't typically worry about that sort of thing except for a very small percentage of code, typically fairly deep in the engine, such as graphics primitive processing, or in animation code... stuff like that. There's no reason shared pointers can't be used extensively throughout your codebase, and then just save the low-level optimizations for your hotspots.
So, someone decides they're ok with working longer hours for more money and you conclude that this is so bizarre that they must be a troll? Hell, there's a name for it: a workaholic. Or maybe he just likes the extra money that he brings in when choosing to work at jobs that demand longer hours. It's not all that unusual for people to enjoy making more money. Maybe he's young and single, or doesn't have kids, and so can choose to spend his free time earning money. Who knows?
No, it's not a choice I'd make myself, but I fail to see how this attitude is so unusual...
Not with smart pointers. Those will automatically call delete for you.
Honestly, that does sound more plausible than the hilarious lies we've heard. Do you happen to have a source for this? Or are you fluent enough in Korean to read the original transcripts?
BTW, the "unicorns" were actually kirin, a horned animal with parts of various creatures making up the body. This creature is said to accompany wise rulers (no coincidence there, eh?).
At the risk of being a bit off-topic...
Shame on all mods who downmodded dgatwood's post redundant or overrated (his post currently stands at 0 Redundant). I disagreed completely with his conclusions, but he presented them clearly, politely, and intelligently. Moderation is not supposed to be a "I disagree with your opinion, and therefore I'm going to silence you" button. You continue to do that, and you'll simply be turning /. into an echo-chamber (well, more than it already is) where no one dares disagree on any substantial issue.
But, you ever notice how few 3-star reviews there are on Amazon for most products?
I'm not sure I completely agree, but I supposed it depends on the product. Let's take a peek at a technical book I was looking at:
It has 25 reviews, with an (algorithmically weighted) average of 3.8 stars. The break down is:
5 star 44%
4 star 24%
3 star 12%
2 star 12%
1 star 8%
That seems reasonable enough to me. My takeaway is that the book is probably solid, but has a few serious flaws. The people who gave three stars, interestingly enough, seemed to give the best critical feedback while also pointing out the positive aspects of the book.
The problem you mention is even worse when you look at user videogame reviews. For instance, on metacritic, you use user ratings of 1/10 (ZOMG worst game EVAR!!!!!) or 10/10 (WOWOW bestest game in history!!!!), and surprisingly little in-between.
Steam did away with this and simply gave users a choice of thumbs up or thumbs down, which I think helps to eliminate that ridiculousness of extremes, although it has some drawbacks as well. Many users can't be trusted to give a reasonably unbiased review of a product they feel passionately about. Others disagree wildly about what various degrees of a rating mean (I've seen one person say they never give five star ratings because no product is perfect, which elicited a *facepalm* from me). And what's the point of a 10 point scale if most games never score below a 6/10 (or likewise with a 100 point scale)?