From LEO, high surface-area/mass objects can deorbit within a couple orbits or so
A couple of orbits or so along a predictable trajectory is a lot easier to shoot down than a low-altitude cruise missile. It might make sense to put something like a massive laser in space, but getting it into the right orbit for a strike and providing it with enough power to punch through the atmosphere and do more than give people on the ground a mild sunburn would be nontrivial.
And a big benefit of on-orbit munitions is that they may have a good chance of surviving a first strike in a nuclear war
Not really. Both the USA and China have tested ground-to-space missiles for shooting down satellites and laser systems that can disable or destroy satellites from the ground. Creating an orbital weapons platform that can survive missile and laser strikes from the ground would be a massive engineering challenge. In any modern nuclear first strike scenario, these things would be the first to be launched, because you want to destroy the enemy's ability to track your launches.
How easy is it to lose the power adaptor you leave at home plugged into the same socket for its entire life? How easy is it to lose something about as big as a thumb nail? Which of these would you rather use for emergency recovery?
It's different in one very important way: you are much less likely to lose the power adaptor than the security token. You'll use the power adaptor every day or two, while you'll only use the security token when you get locked out of your device. It's like the original iPod dock: my iPod was the only mobile device I owned that never had the battery go flat. The dock plugged into my HiFi so I could listen to music at home and so I had a reason for always dropping it in the charger. Every other mobile device got plugged in when I noticed it needed charging (Apple, cleverly, no longer includes the dock, so loses this advantage).
No one (sensible) cares about whether you can solve the brain teaser. That's not what they're for. They care how you try to solve it. If they ask a question and you immediately give the right answer then all that they learn is that you've seen it before. It's the intermediate steps that are important.
Is using code snippets from the internet really an issue?
Depends, what's the license on the snippets? If they're only a couple of lines long then they won't be covered by copyright, but on the other hand if you're copying things that short from the 'net then you probably shouldn't be writing code for a living. If you're copying long bits without checking that they are under a valid license then you are potentially going to expose a future employer to a huge amount of liability, so shouldn't be hired.
He said over 3 months. You, therefore, fail to get the job due to either lack of attention to detail or basic reasoning abilities. Unfortunately, this test was not specific enough to determine which of these you actually failed.
There are lots of reasons why people in Nokia management would think it's a good idea. They're staggeringly incompetent and cashing out now by selling the company is probably a lot easier than learning how to do the whole 'management' thing.
But why would Microsoft want to buy it? They sell Windows Phone 7 to other companies, and they are planning on making Windows 8 much more attractive to other companies in the tablet market, especially to existing PC makers. Hardware makers see MS as a safe bet because (unlike Google) they don't have a horse in the hardware race. They are willing to gouge all of their OEM customers equally and without prejudice. If they bought Nokia then anyone considering licensing WP7 or Windows 8 for tablets / smartphones would be immediately aware that they were competing with one of their key component suppliers, which puts them at an immediate disadvantage. This would be a very strong incentive to pick a different smartphone OS...
I wouldn't be surprised if Nokia was aiming to be bought by MS. I would be surprised if MS was interested in buying. If MS bought Nokia, then they would be directly competing with their own customers, which is something that historically Microsoft has avoided (well, right up until the point they decide they want 100% of a market one of their customers controls).
Since Vista (I think - 2K was the last Windows version I used properly), normal users haven't had write permission to Program Files, so applications that try to store user data there simply won't work. Program Files is analogous to/usr/[local/]{bin,lib,share} on a *NIX system. The registry has local machine (shared) and current user (private) settings. Anything that would be in/etc is in HKLM and anything that would be in ~/.something is in HKCU.
I've often wondered though. . . yeah, you can make money that way, but for every person who spends hundreds or thousands of dollars on DLC, I bet you have hundreds or thousands of players who never buy anything?
That kind of logic is what's causing problems for the movie and music industries. It doesn't matter how many people are not paying for your product, it matters how many people are. Your job, as a capitalist business, is to maximise the product of the number of people who are paying by the amount that they are paying. The movie industry has spent a huge amount on marketing campaigns to try to get people to stop pirating their products, forgetting that their goal should have been to make people start paying for them.
but on the whole Open Source is devastating the software development and sell model
True, but that doesn't really matter to most programmers because this model has never been responsible for employing more than 10% of software developers. The vast majority are employed working on in-house software, or software paid for by the customer directly. Open source fits perfectly into this model - indeed, most of it always was Free Software, as the customer received all of the rights to the code.
I remember nearly a decade ago I was at SD hosting a BOF on the impact of Open Source. We danced for hours around the issue and it was a very good BOF. At the end I summed it up and we all agreed, the software developer will not make the same amount of money as they had
Why would that be the case? Businesses still have problems that can only be solved by writing software. Whether that is written from scratch, or built using some existing components doesn't matter much to them, except in how much it costs. If it costs less due to open source then it means that they have more budget to spend on less vital things.
If the entire COTS market for software vanished overnight, it would have very little impact on most software developers.
It was better, but it was also more expensive. I bought an old MTX-512 (now there's a rare 8-bit machine) second hand a few years ago, and it came with a load of magazines from that era. The C-64 was significantly cheaper than even the BBC Model A. And you needed a BBC Master to have as much RAM as the C64. The BBC had a (much!) better dialect of BASIC, better graphics, better sound, and far better I/O, but did not have a better price. Unless you were a school in the '80s, then the government would pay something like 50% of the cost of a computer that met a strict set of requirements (including a programming environment with support for structured programming) that the BBC met and the C64 didn't.
Seriously, anyone who thinks Microsoft would turn around and immediately swap out OS/webserver because herp derp, needs to be fired
So, someone who pays attention to history should not have a job, according to you?
When Microsoft bought Hotmail, it was running on FreeBSD. This was slightly embarrassing for Microsoft and they made a big show of switching it over to Windows NT. It then died horribly, and they quietly switched it back. The second time they did the transition, it worked. Neither time was done for technical reasons - hotmail had already demonstrated that it scaled very well with the old infrastructure - it was done purely so Microsoft could point at Hotmail and say that they were using a 100% Microsoft back end.
Most of the time you see a server error, it means that the web server is working but something else (often the database, more often the fastcgi or whatever) has failed. It's when you don't see any error that the web server itself is broken.
Since Apache 1.x is now officially unsupported, OpenBSD has imported nginx into the base system and it will be the included web server in a near future release. I've been using Lighttpd for a few years because it was lighter-weight and easier to configure. Development seems to have stalled in the past couple of years though, and nginx looks like a promising alternative.
$100 says that your statement is 100% false in every example you can give me.
I'll take that bet. Two projects: gnuplot and pdflatex. I have the versions that I used for my PhD thesis in 2006 on my PowerPC Mac. I recompiled the same versions on my first MacBook Pro. This copy still works on my new MacBook Pro running 10.7 (I didn't upgrade, the new machine came with it, so no Rosetta - even if you do copy if from the old machine, you don't have the PowerPC libraries and it won't actually work, so that's more nonsense in your post).
ZFS actually does have a mode to skip the comparison step. I think it may even be the default. The rationale is that the probability of hash collisions between two blocks is much lower than the probability of data loss from a variety of other problems (including meteor strikes to the server), so there's not much point. If the hash is in RAM and the data is on disk, then skipping the seek-and-read can be quite significant.
Not for the two that he talked about. VM disk images contain filesystems, so they are perfect for deduplication (as I'm seeing on my NAS). If you modify 10MB in a VM disk image, then you will modify 10MB of blocks (possibly some more, depending on the filesystem), but you definitely won't be re-storing the entire thing. For mail spools, the common format for spools-in-a-single-file is mbox, where (precisely to avoid the problem you're talking about, where a small change generates a huge amount of I/O), new emails are appended to the file and mails that are deleted are not actually removed, they're just marked as not-present until a lot of space is wasted and only then is the entire file rewritten. Incremental backups of both gain a lot from dedup.
In my cases, the biggest files are VM images. In this case, it is only storing the differences, because changing a few files in the VM only modifies a few blocks of the image but causes Time Machine to create a new backup copy of the entire image. This means that it's backing up 10GB files quite regularly, but I'm only storing a few MBs for each one.
You got 40MB writing to memory cache possibly, not the ZFS store.
I did? That's interesting. I copied 500GB of data from an external FireWire disk attached to my laptop to the NAS via a GigE connection, yet the NAS only has 8GB of RAM. That's one hell of a compression algorithm they're using for the RAM cache...
Slashdot Mirror
is it a meteor or is it a bomb. The down side is it could lead to an accidental counter-strike if someone thought a meteor was a bomb re-entering
Today's injection of culture into Slashdot: Icarus Allsorts, by Roger McGough .
From LEO, high surface-area/mass objects can deorbit within a couple orbits or so
A couple of orbits or so along a predictable trajectory is a lot easier to shoot down than a low-altitude cruise missile. It might make sense to put something like a massive laser in space, but getting it into the right orbit for a strike and providing it with enough power to punch through the atmosphere and do more than give people on the ground a mild sunburn would be nontrivial.
And a big benefit of on-orbit munitions is that they may have a good chance of surviving a first strike in a nuclear war
Not really. Both the USA and China have tested ground-to-space missiles for shooting down satellites and laser systems that can disable or destroy satellites from the ground. Creating an orbital weapons platform that can survive missile and laser strikes from the ground would be a massive engineering challenge. In any modern nuclear first strike scenario, these things would be the first to be launched, because you want to destroy the enemy's ability to track your launches.
How easy is it to lose the power adaptor you leave at home plugged into the same socket for its entire life? How easy is it to lose something about as big as a thumb nail? Which of these would you rather use for emergency recovery?
It's different in one very important way: you are much less likely to lose the power adaptor than the security token. You'll use the power adaptor every day or two, while you'll only use the security token when you get locked out of your device. It's like the original iPod dock: my iPod was the only mobile device I owned that never had the battery go flat. The dock plugged into my HiFi so I could listen to music at home and so I had a reason for always dropping it in the charger. Every other mobile device got plugged in when I noticed it needed charging (Apple, cleverly, no longer includes the dock, so loses this advantage).
There are, but only if you order them directly from China. They can't be sold in the US / EU due to the magsafe patent.
My CV lists the rifle shooting team I was in under 'achievements' but I wouldn't expect to be asked to shoot the interviewer...
No one (sensible) cares about whether you can solve the brain teaser. That's not what they're for. They care how you try to solve it. If they ask a question and you immediately give the right answer then all that they learn is that you've seen it before. It's the intermediate steps that are important.
Is using code snippets from the internet really an issue?
Depends, what's the license on the snippets? If they're only a couple of lines long then they won't be covered by copyright, but on the other hand if you're copying things that short from the 'net then you probably shouldn't be writing code for a living. If you're copying long bits without checking that they are under a valid license then you are potentially going to expose a future employer to a huge amount of liability, so shouldn't be hired.
He said over 3 months. You, therefore, fail to get the job due to either lack of attention to detail or basic reasoning abilities. Unfortunately, this test was not specific enough to determine which of these you actually failed.
There are lots of reasons why people in Nokia management would think it's a good idea. They're staggeringly incompetent and cashing out now by selling the company is probably a lot easier than learning how to do the whole 'management' thing.
But why would Microsoft want to buy it? They sell Windows Phone 7 to other companies, and they are planning on making Windows 8 much more attractive to other companies in the tablet market, especially to existing PC makers. Hardware makers see MS as a safe bet because (unlike Google) they don't have a horse in the hardware race. They are willing to gouge all of their OEM customers equally and without prejudice. If they bought Nokia then anyone considering licensing WP7 or Windows 8 for tablets / smartphones would be immediately aware that they were competing with one of their key component suppliers, which puts them at an immediate disadvantage. This would be a very strong incentive to pick a different smartphone OS...
I wouldn't be surprised if Nokia was aiming to be bought by MS. I would be surprised if MS was interested in buying. If MS bought Nokia, then they would be directly competing with their own customers, which is something that historically Microsoft has avoided (well, right up until the point they decide they want 100% of a market one of their customers controls).
Since Vista (I think - 2K was the last Windows version I used properly), normal users haven't had write permission to Program Files, so applications that try to store user data there simply won't work. Program Files is analogous to /usr/[local/]{bin,lib,share} on a *NIX system. The registry has local machine (shared) and current user (private) settings. Anything that would be in /etc is in HKLM and anything that would be in ~/.something is in HKCU.
I've often wondered though. . . yeah, you can make money that way, but for every person who spends hundreds or thousands of dollars on DLC, I bet you have hundreds or thousands of players who never buy anything?
That kind of logic is what's causing problems for the movie and music industries. It doesn't matter how many people are not paying for your product, it matters how many people are. Your job, as a capitalist business, is to maximise the product of the number of people who are paying by the amount that they are paying. The movie industry has spent a huge amount on marketing campaigns to try to get people to stop pirating their products, forgetting that their goal should have been to make people start paying for them.
The free model seems great too
Nope, there's an article just a few down saying that this model is broken, and since it was on Slashdot it must be true!
but on the whole Open Source is devastating the software development and sell model
True, but that doesn't really matter to most programmers because this model has never been responsible for employing more than 10% of software developers. The vast majority are employed working on in-house software, or software paid for by the customer directly. Open source fits perfectly into this model - indeed, most of it always was Free Software, as the customer received all of the rights to the code.
I remember nearly a decade ago I was at SD hosting a BOF on the impact of Open Source. We danced for hours around the issue and it was a very good BOF. At the end I summed it up and we all agreed, the software developer will not make the same amount of money as they had
Why would that be the case? Businesses still have problems that can only be solved by writing software. Whether that is written from scratch, or built using some existing components doesn't matter much to them, except in how much it costs. If it costs less due to open source then it means that they have more budget to spend on less vital things.
If the entire COTS market for software vanished overnight, it would have very little impact on most software developers.
It was better, but it was also more expensive. I bought an old MTX-512 (now there's a rare 8-bit machine) second hand a few years ago, and it came with a load of magazines from that era. The C-64 was significantly cheaper than even the BBC Model A. And you needed a BBC Master to have as much RAM as the C64. The BBC had a (much!) better dialect of BASIC, better graphics, better sound, and far better I/O, but did not have a better price. Unless you were a school in the '80s, then the government would pay something like 50% of the cost of a computer that met a strict set of requirements (including a programming environment with support for structured programming) that the BBC met and the C64 didn't.
Seriously, anyone who thinks Microsoft would turn around and immediately swap out OS/webserver because herp derp, needs to be fired
So, someone who pays attention to history should not have a job, according to you?
When Microsoft bought Hotmail, it was running on FreeBSD. This was slightly embarrassing for Microsoft and they made a big show of switching it over to Windows NT. It then died horribly, and they quietly switched it back. The second time they did the transition, it worked. Neither time was done for technical reasons - hotmail had already demonstrated that it scaled very well with the old infrastructure - it was done purely so Microsoft could point at Hotmail and say that they were using a 100% Microsoft back end.
Most of the time you see a server error, it means that the web server is working but something else (often the database, more often the fastcgi or whatever) has failed. It's when you don't see any error that the web server itself is broken.
Since Apache 1.x is now officially unsupported, OpenBSD has imported nginx into the base system and it will be the included web server in a near future release. I've been using Lighttpd for a few years because it was lighter-weight and easier to configure. Development seems to have stalled in the past couple of years though, and nginx looks like a promising alternative.
Well, it died, but then it came back with sharper teeth and better dress sense. Just keep it out of the sunlight...
$100 says that your statement is 100% false in every example you can give me.
I'll take that bet. Two projects: gnuplot and pdflatex. I have the versions that I used for my PhD thesis in 2006 on my PowerPC Mac. I recompiled the same versions on my first MacBook Pro. This copy still works on my new MacBook Pro running 10.7 (I didn't upgrade, the new machine came with it, so no Rosetta - even if you do copy if from the old machine, you don't have the PowerPC libraries and it won't actually work, so that's more nonsense in your post).
ZFS actually does have a mode to skip the comparison step. I think it may even be the default. The rationale is that the probability of hash collisions between two blocks is much lower than the probability of data loss from a variety of other problems (including meteor strikes to the server), so there's not much point. If the hash is in RAM and the data is on disk, then skipping the seek-and-read can be quite significant.
Not for the two that he talked about. VM disk images contain filesystems, so they are perfect for deduplication (as I'm seeing on my NAS). If you modify 10MB in a VM disk image, then you will modify 10MB of blocks (possibly some more, depending on the filesystem), but you definitely won't be re-storing the entire thing. For mail spools, the common format for spools-in-a-single-file is mbox, where (precisely to avoid the problem you're talking about, where a small change generates a huge amount of I/O), new emails are appended to the file and mails that are deleted are not actually removed, they're just marked as not-present until a lot of space is wasted and only then is the entire file rewritten. Incremental backups of both gain a lot from dedup.
No, that's the opposite of deduplication...
In my cases, the biggest files are VM images. In this case, it is only storing the differences, because changing a few files in the VM only modifies a few blocks of the image but causes Time Machine to create a new backup copy of the entire image. This means that it's backing up 10GB files quite regularly, but I'm only storing a few MBs for each one.
You got 40MB writing to memory cache possibly, not the ZFS store.
I did? That's interesting. I copied 500GB of data from an external FireWire disk attached to my laptop to the NAS via a GigE connection, yet the NAS only has 8GB of RAM. That's one hell of a compression algorithm they're using for the RAM cache...