Sample size of one does not give any odds for intelligent life elsewhere, nor any timescale prediction. We know hardly anything about life on Earth for the first 3,5 billion years, we have very few data points, so we can't even estimate how long it "should" have taken to evolve here.
I've been long enough in IT security to no longer believe that simply because something is on someone's computer or because some data originated from someone's computer that it actually has something to do with that someone. Computers are far harder to secure, far easier to break into and such a break in far harder to notice than your average home would be, especially considering the recent discoveries how "your" system is not necessarily "yours" at all.
But about the same can be said about cash found in a mattress. It' quite possible it was put there in a factory already as a part of mob money transfer scheme (and somebody got their kneecaps removed, because it was lost), or perhaps a fugitive broke in and hid the booty there (and was then caught or killed and never came to get it, lucky you), or perhaps it simply belongs to the Significant Other, who is secretly running an entirely different illegal business, completely unrelated to the first case.
All of these are unlikely, but I think they are still at least as plausible, as having an unencrypted bitcoin walled with that much money in a "zombie" host.
All programs are jigsaw puzzles, by your definition.
Except.... Jigsaw puzzles have fixed number of pieces, and can be put together in only one way, and are static and unchanging once put together. In fact jigsaw puzzle is about the worst puzzle analogy for computer programs I can think of.
Also, not all jigsaw puzzles are equally easy. In fact, without using highly sophisticated image recognition and heuristics combined with parallel processing, their time complexity is O(n*n).
If that kind of "evidence" holds a drop in water in court, good night legal system.
If they seized that amount in dollars from inside his mattress and transferred the cash to FBI custody, how much proving would they have to do? What would happen to the money?
After answering that, how is seizing bitcoin wallet from inside his computer and transferring the coins to FBI custody any different?
I'm no fan of the patent wars, but if Samsung played by the rules...
Yeah, but the rules for Standards Essential Patents are a bit different, because anti-competitive behaviour laws enter the picture much more easily. And then things quickly get very complicated, I think the general intent of EU here is: play fair with SEP, agree on a fair licensing price, work it out, or else...
So, if NSA did not have full access to CIA employee and contractor data, does it mean they're not that good after all? Because normally, why would anybody bother telling NSA anything? Wouldn't that be redundant, just unnecessary bureaucracy?
Maybe your code is different, but have you noticed how many C functions return signed value, where -1 means error? Often same functions take unsigned size in, and then return actual size written/read/whatever. Except the maximum returned size has one bit less for positive values, than the maximum size given as argument. Do you see the problem?
As a result, lot of times an unsigned is used in C, even in standard library, a signed value should be used instead.
Wake me up when java supports unsigned integers. Until then it's not a real language.
While unsigned numbers are great for a few things, mixing them with signed numbers is a real pain. Just consider all the C functions, which take in unsigned but return signed, and casting galore that follows. Of course you can just disable relevant warnings entirely and blindly hope implicit casts anywhere will never overflow, but that is kind of sloppy, and just asking for someone to find a way to use it for an exploit. Which incidentally is what most C code does.
I didn't mean you'd use Qt on everything, I meant you're pretty well prepared for everything if you know all that. It was tongue-in-cheek too, because it's such a blanket statement, same could be said about.NET or JVM worlds.
Of course the statement isn't completely true either, all these don't help much with VHDL, for example (or maybe writing GPU shader code helps with that, not sure).
GUI programming is a new thing and still it's rapidly transforming. Change that to event-driven application architecture, which almost all GUI apps have.
Relational databases, well... NoSQL would be enough for this list.
Model-view stuff seems to be missing from the list, and all kinds of patterns in general. Also, totally missing from TFS is client-server models, be it the backend-frontend model of web apps, or traditional TCP/IP protocols.
That being said, if you master every aspect of Qt5 including QML and using network and databases, and development for different mobile devices, while also embracing the functional aspect of JavaScript (for QML and HTML), you should be pretty well set for everything on every platform;-)
Is it wholly un-American to think that this whole article, the subject overall and this discussion are absurd, when talking about a civilized democratic country? I mean, discussion about bad stuff about police, sure, but this reeks of things having gone to the next level, something you'd expect in a 3rd world totalitarian "democratic republic".
"Have you ever looked at how much CPU a Linux OS takes, when certain cron jobs kick in? Just horrible!"
Certain cron jobs? Do you even know what cron is or how it works? Presumably you have a cron entry that runs "Eat-All-My-CPU-So-I-Can-Make-Absurd-Statements-On-Slashdot.sh".
"And 20 minutes, you say? The Windows box was probably downloading and processing updates. "
Since when does Windows automatically apply updates at power on without the new user telling it to do so? More likely it was still completing the boot process after doing the initial desktop paint "fake-out" that is done to make it appear that it boots much more quickly than it actually does.;-)
"My point is, even though Windows may suck in many exquisite ways, your post is way off the mark."
Where is it off the mark? While yours is clearly absurd, I see nowhere where the GPs was off the mark.
Indeed, there was probably something equal to cron job "Eat-All-My-CPU-So-I-Can-Make-Absurd-Statements-On-Slashdot.sh" on that PC, because now that I could check, normal CPU usage of idle Windows box seems to be 0..1%. The point is, that is not kernel, that's userspace... Read the post I replied to again, it talks about increasing kernel overhead. The overhead is exactly same for system services and other userspace programs.
Anyway, your guess about "completing boot process" may well be spot on, though it probably was "first boot process" then, because slowest I've seen Windows finishing startup for real, on a laptop with encrypted slow HD and a ton of corporate crapware, is around 10 minutes.
But GP was off the mark by considering the 10% figure (whatever the reason for it) including system services in the 50% microkernel overhead thought experiment. If system services suffer the overhead, then so would any userspace programs.
And 20 minutes, you say? The Windows box was probably downloading and processing updates.
No, just kernel + system services.
You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
In your assumed on-line scenario, it installs up-to-date, thank you.
Uh... The assumed scenario is, a new computer arrives with who-knows-how-old pre-installed Windows image... It's irrelevant wether it was installed up-to-date or not whenever the image was created, if it has not been updated since then.
Also, this was about changing kernel to be microkernel, with more overhead. Services are not kernel, so any number which includes them is irrelevant in the context presented in the post I replied to.
Either the computer was doing something related to being turned on for the first time, or there was some serious crapware installed. This Windows f'ing Vista running on this low-end laptop from 2009, with single core Celeron, and which has never been reinstalled, uses 0..1% CPU when I check Task Manager performance graph. If a new computer shows 10%, then it's not the kernel.
cron is userspace, so that's not an issue, he's talking about kernel-used CPU cycles, of stuff tighly integrated into the lower levels of the OS. crontab -e will let you edit cron tabs that you distro preconfigured easily, completely unrelated to Linux itself.
And 20 minutes, you say? The Windows box was probably downloading and processing updates. You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
Not unless I configure it to do so.
The mentioned Windows services are also userspace, and would not be hit by kernel getting more overhead, which was the whole point.
Also, it seems this Windows was pre-installed, and anybody shipping pre-installed Windows without turning automatic updates on should be shot for malicious negligence. If you install it yourself, IIRC you will be prompted how you want to handle updates (may depend on version).
Have you ever looked at how much CPU a Linux OS takes, when certain cron jobs kick in? Just horrible!
Oh, but they are in userspace and would not be affected by kernel becoming slower? Well, I'm fairly sure those Windows services were also in userspace.
And 20 minutes, you say? The Windows box was probably downloading and processing updates. You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
My point is, even though Windows may suck in many exquisite ways, your post is way off the mark.
A game is a place that you go and do things, not a story. Movie directors have a hard time with this. They want to lock the player into a track ride, like an amusement park.
Speak for yourself. I play games for the story, the rest matters only in context of the story and relative to it.
Which is why I'll never forgive SWTOR2, may the ones responsible be suffocated in bantha excrement.
While I will refrain from commenting on your post otherwise, I have to say that execution by a simple needle sounds like quite a nasty and slow way to kill someone. If that is not enough for you, I don't want to know how "medieval" you want to get.
Function of that sleep key should be configurable. Assuming you are running Windows, Vista or later, try hitting that Windows key you touch so rarely, then type "power", and select appropriate choice to go to correct control panel section.
Windows key (since Vista) is the greatest input innovation in Microsoft OSes since ctrl-alt-del. I start basically any new task by hitting it and then typing what I want to do.
This is Slashdot. If you cant write your own browser extension to automatically put in your own HTML tags then this is not the site for you. Reddit is ----> thataway
Uh, sorry but that was utterly unimaginative rant, so much so that I'll call it a stupid rant. Some points:
Asteroid mining is pre-requisite for sensible manufacturing in space, and has actual commercial interest in it too. And space manufacturing is pre-requisite of doing anything really interesting in space, hauling all the stuff up from gravity wells just will not do in long term.
More efficient solar panels could mean a nanofilm with area measured in square kilometers, and ability to power it also with (solar powered) lasers once even that gets out of useful range of solar radiation. This is just an example btw, not prediction of real deployed tech.
I'm fairly sure a fission-powered thermoelectric generator could be made lighter than radioactive decay powered RTG, already at power range and RTG lifetime of Voyager probes. In fact having "cheap" plutonium for building RTGs with stagnated technology is holding us back, with their limited power output and lifetime. We need incentive to develop something better.
Mercury is the best place for solar neutrino detector, thanks to inverse square law and required shielding, and sending rovers there would be first step into building one.
You should really read up on balloons at Venus, in the upper atmosphere, it's quite an interesting concept and technical challenges are interesting too. Also, if we want to really understand planetary atmospheres (including our own), we really have to spend some time in the atmosphere of Venus too.
About radios, just think how much audio and video data, with what kind of tiny antennas, and with how little energy per uncompressed bit, we can transmit with our consumer mobile devices today, compared to just 10 years ago, and it doesn't look like things are slowing down, quite the opposite. That relative improvement will trickle down to space tech too, in due time.
Do we really need to send probes to the outer solar system for next half a century? Let's wait for better, lighter reactors and solar panels, and less power hungry electronics and more sensitive radios. In the mean time, let's mine some asteroids, send up better telescopes, practice human presence in space, send balloons to Venus and rovers to polar regions of Mercury...
Slashdot Mirror
Sample size of one does not give any odds for intelligent life elsewhere, nor any timescale prediction. We know hardly anything about life on Earth for the first 3,5 billion years, we have very few data points, so we can't even estimate how long it "should" have taken to evolve here.
I've been long enough in IT security to no longer believe that simply because something is on someone's computer or because some data originated from someone's computer that it actually has something to do with that someone. Computers are far harder to secure, far easier to break into and such a break in far harder to notice than your average home would be, especially considering the recent discoveries how "your" system is not necessarily "yours" at all.
But about the same can be said about cash found in a mattress. It' quite possible it was put there in a factory already as a part of mob money transfer scheme (and somebody got their kneecaps removed, because it was lost), or perhaps a fugitive broke in and hid the booty there (and was then caught or killed and never came to get it, lucky you), or perhaps it simply belongs to the Significant Other, who is secretly running an entirely different illegal business, completely unrelated to the first case.
All of these are unlikely, but I think they are still at least as plausible, as having an unencrypted bitcoin walled with that much money in a "zombie" host.
All programs are jigsaw puzzles, by your definition.
Except.... Jigsaw puzzles have fixed number of pieces, and can be put together in only one way, and are static and unchanging once put together. In fact jigsaw puzzle is about the worst puzzle analogy for computer programs I can think of.
Also, not all jigsaw puzzles are equally easy. In fact, without using highly sophisticated image recognition and heuristics combined with parallel processing, their time complexity is O(n*n).
If that kind of "evidence" holds a drop in water in court, good night legal system.
If they seized that amount in dollars from inside his mattress and transferred the cash to FBI custody, how much proving would they have to do? What would happen to the money?
After answering that, how is seizing bitcoin wallet from inside his computer and transferring the coins to FBI custody any different?
There (probably) are no laws which would allow that... I hope so at least.
I'm no fan of the patent wars, but if Samsung played by the rules...
Yeah, but the rules for Standards Essential Patents are a bit different, because anti-competitive behaviour laws enter the picture much more easily. And then things quickly get very complicated, I think the general intent of EU here is: play fair with SEP, agree on a fair licensing price, work it out, or else...
So, if NSA did not have full access to CIA employee and contractor data, does it mean they're not that good after all? Because normally, why would anybody bother telling NSA anything? Wouldn't that be redundant, just unnecessary bureaucracy?
Maybe your code is different, but have you noticed how many C functions return signed value, where -1 means error? Often same functions take unsigned size in, and then return actual size written/read/whatever. Except the maximum returned size has one bit less for positive values, than the maximum size given as argument. Do you see the problem?
As a result, lot of times an unsigned is used in C, even in standard library, a signed value should be used instead.
Wake me up when java supports unsigned integers. Until then it's not a real language.
While unsigned numbers are great for a few things, mixing them with signed numbers is a real pain. Just consider all the C functions, which take in unsigned but return signed, and casting galore that follows. Of course you can just disable relevant warnings entirely and blindly hope implicit casts anywhere will never overflow, but that is kind of sloppy, and just asking for someone to find a way to use it for an exploit. Which incidentally is what most C code does.
I didn't mean you'd use Qt on everything, I meant you're pretty well prepared for everything if you know all that. It was tongue-in-cheek too, because it's such a blanket statement, same could be said about .NET or JVM worlds.
Of course the statement isn't completely true either, all these don't help much with VHDL, for example (or maybe writing GPU shader code helps with that, not sure).
GUI programming is a new thing and still it's rapidly transforming. Change that to event-driven application architecture, which almost all GUI apps have.
Relational databases, well... NoSQL would be enough for this list.
Model-view stuff seems to be missing from the list, and all kinds of patterns in general. Also, totally missing from TFS is client-server models, be it the backend-frontend model of web apps, or traditional TCP/IP protocols.
That being said, if you master every aspect of Qt5 including QML and using network and databases, and development for different mobile devices, while also embracing the functional aspect of JavaScript (for QML and HTML), you should be pretty well set for everything on every platform ;-)
Is it wholly un-American to think that this whole article, the subject overall and this discussion are absurd, when talking about a civilized democratic country? I mean, discussion about bad stuff about police, sure, but this reeks of things having gone to the next level, something you'd expect in a 3rd world totalitarian "democratic republic".
Certain cron jobs? Do you even know what cron is or how it works? Presumably you have a cron entry that runs "Eat-All-My-CPU-So-I-Can-Make-Absurd-Statements-On-Slashdot.sh".
Since when does Windows automatically apply updates at power on without the new user telling it to do so? More likely it was still completing the boot process after doing the initial desktop paint "fake-out" that is done to make it appear that it boots much more quickly than it actually does. ;-)
Where is it off the mark? While yours is clearly absurd, I see nowhere where the GPs was off the mark.
Indeed, there was probably something equal to cron job "Eat-All-My-CPU-So-I-Can-Make-Absurd-Statements-On-Slashdot.sh" on that PC, because now that I could check, normal CPU usage of idle Windows box seems to be 0..1%. The point is, that is not kernel, that's userspace... Read the post I replied to again, it talks about increasing kernel overhead. The overhead is exactly same for system services and other userspace programs.
Anyway, your guess about "completing boot process" may well be spot on, though it probably was "first boot process" then, because slowest I've seen Windows finishing startup for real, on a laptop with encrypted slow HD and a ton of corporate crapware, is around 10 minutes.
But GP was off the mark by considering the 10% figure (whatever the reason for it) including system services in the 50% microkernel overhead thought experiment. If system services suffer the overhead, then so would any userspace programs.
And 20 minutes, you say? The Windows box was probably downloading and processing updates.
No, just kernel + system services.
You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
In your assumed on-line scenario, it installs up-to-date, thank you.
Uh... The assumed scenario is, a new computer arrives with who-knows-how-old pre-installed Windows image... It's irrelevant wether it was installed up-to-date or not whenever the image was created, if it has not been updated since then.
Also, this was about changing kernel to be microkernel, with more overhead. Services are not kernel, so any number which includes them is irrelevant in the context presented in the post I replied to.
Either the computer was doing something related to being turned on for the first time, or there was some serious crapware installed. This Windows f'ing Vista running on this low-end laptop from 2009, with single core Celeron, and which has never been reinstalled, uses 0..1% CPU when I check Task Manager performance graph. If a new computer shows 10%, then it's not the kernel.
cron is userspace, so that's not an issue, he's talking about kernel-used CPU cycles, of stuff tighly integrated into the lower levels of the OS.
crontab -e will let you edit cron tabs that you distro preconfigured easily, completely unrelated to Linux itself.
And 20 minutes, you say? The Windows box was probably downloading and processing updates. You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
Not unless I configure it to do so.
The mentioned Windows services are also userspace, and would not be hit by kernel getting more overhead, which was the whole point.
Also, it seems this Windows was pre-installed, and anybody shipping pre-installed Windows without turning automatic updates on should be shot for malicious negligence. If you install it yourself, IIRC you will be prompted how you want to handle updates (may depend on version).
Who would want free BSD software relicensed under a shittier unfree license?
Users of a fork or otherwise copied code, who would like to see the source.
Developers, who see a bugfix or improvement of their code in a fork, and would like to merge that back to their codebase.
Have you ever looked at how much CPU a Linux OS takes, when certain cron jobs kick in? Just horrible!
Oh, but they are in userspace and would not be affected by kernel becoming slower? Well, I'm fairly sure those Windows services were also in userspace.
And 20 minutes, you say? The Windows box was probably downloading and processing updates. You know, when booting up a Linux which has never been updated, it would also be prudent to have it updated during the first 20 minutes.
My point is, even though Windows may suck in many exquisite ways, your post is way off the mark.
A game is a place that you go and do things, not a story. Movie directors have a hard time with this. They want to lock the player into a track ride, like an amusement park.
Speak for yourself. I play games for the story, the rest matters only in context of the story and relative to it.
Which is why I'll never forgive SWTOR2, may the ones responsible be suffocated in bantha excrement.
While I will refrain from commenting on your post otherwise, I have to say that execution by a simple needle sounds like quite a nasty and slow way to kill someone. If that is not enough for you, I don't want to know how "medieval" you want to get.
Function of that sleep key should be configurable. Assuming you are running Windows, Vista or later, try hitting that Windows key you touch so rarely, then type "power", and select appropriate choice to go to correct control panel section.
Windows key (since Vista) is the greatest input innovation in Microsoft OSes since ctrl-alt-del. I start basically any new task by hitting it and then typing what I want to do.
I'm not familiar with this "Dash" thing. Can't it just be taken out when you install the new Ubuntu?
Easiest is to install Kubuntu, Xubuntu or Lubuntu instead of the "main" distro.
This is Slashdot. If you cant write your own browser extension to automatically put in your own HTML tags then this is not the site for you. Reddit is ----> thataway
There, fixed that for you.
Except, sadly, I think 99% of /. readers can't.
Uh, sorry but that was utterly unimaginative rant, so much so that I'll call it a stupid rant. Some points:
Asteroid mining is pre-requisite for sensible manufacturing in space, and has actual commercial interest in it too. And space manufacturing is pre-requisite of doing anything really interesting in space, hauling all the stuff up from gravity wells just will not do in long term.
More efficient solar panels could mean a nanofilm with area measured in square kilometers, and ability to power it also with (solar powered) lasers once even that gets out of useful range of solar radiation. This is just an example btw, not prediction of real deployed tech.
I'm fairly sure a fission-powered thermoelectric generator could be made lighter than radioactive decay powered RTG, already at power range and RTG lifetime of Voyager probes. In fact having "cheap" plutonium for building RTGs with stagnated technology is holding us back, with their limited power output and lifetime. We need incentive to develop something better.
Mercury is the best place for solar neutrino detector, thanks to inverse square law and required shielding, and sending rovers there would be first step into building one.
You should really read up on balloons at Venus, in the upper atmosphere, it's quite an interesting concept and technical challenges are interesting too. Also, if we want to really understand planetary atmospheres (including our own), we really have to spend some time in the atmosphere of Venus too.
About radios, just think how much audio and video data, with what kind of tiny antennas, and with how little energy per uncompressed bit, we can transmit with our consumer mobile devices today, compared to just 10 years ago, and it doesn't look like things are slowing down, quite the opposite. That relative improvement will trickle down to space tech too, in due time.
Do we really need to send probes to the outer solar system for next half a century? Let's wait for better, lighter reactors and solar panels, and less power hungry electronics and more sensitive radios. In the mean time, let's mine some asteroids, send up better telescopes, practice human presence in space, send balloons to Venus and rovers to polar regions of Mercury...
Hacking in and of itself is cheating. So if you can cheat at cheating, you're doing it right -- you're smarter than the beast you're facing.
...unless you get caught, of course. Then you're just incompetent (you were caught) and stupid (you did not know or care about your incompetence).