The main problem is: with a gun, you have a good chance of defeating the mugger[s]. When unarmed, about none. With the police being as unreliable as they are, there's hardly any deterrent to mugging.
From my own benchmarks, clang behaves like gcc at one optimization level lower, both for compilation speed and speed of the resulting binary.
It's support for C++ has been rife with bugs, but things have improved a lot, so clang gets roughly close to gcc (other than highest, rarely used levels of optimization), except for two areas: warnings (clang produces tons of bogus ones), and portability (clang supports only i386, amd64, 32 bit arm, big-endian mips, s390, while gcc does pretty much everything).
I can build binaries for any other system from any other system. Except for OS X, where you need someone with that exact piece of overpriced junk, and it's hard to find such a person who's willing to run nightlies for you. If you'd want to prove the last part incorrect, it would be nice if someone could run Dungeon Crawl builds at least weekly.
For any other modern system, you either have nicely packaged cross toolchain, or could build it yourself with not much hassle but most folks don't bother as it's easier to run a virtual machine at no cost.
I'm afraid you put political correctness above data. Here's why races being equal is impossible:
There's a vast number of genetical variation between people: some are stronger, some weaker, some have bigger jaws, some blond hair, some... and so on. It's quite obvious at least some mental attributes have genetic variation.
Now let's have two vectors of numbers, one labelled "intelligence", second "skin pigmentation". You can use a more complex graph, of course, but a linear one is enough. Initialize all values to the same number. Now, in every iteration, randomly select an entry, then randomly either put it halfway towards one of its neighbours, or add or subtract a random value. Repeat for umpteen zillion of iterations.
Now check if these vectors show correlation. Unless you are extremely lucky and they happen to zero out by chance, there will be significant correlation, even though there's obviously no causation (at no point values of "intelligence" and "pigmentation" ever depended on one another).
Unlike this simulation, in real life it's possible we have at least common causes: survival in temperate regions that suffer from winters requires more preparation than savannahs and jungles, yet as I've shown above, even in an complete absence of causation correlation is extremely likely.
This doesn't mean a given person from one group is necessarily stronger/stupider/bigger-jawed than one from the other, thanks to individual variation. Yet at least the averages between two groups will show correlation with many unrelated attributes.
A "positive" discrimination is still discrimination.
Any kind of affirmative action is harmful not only to potential employees involved, but also to the employer as well -- and I'm not speaking about what is "moral", but actual utility.
Try a simulation: assign every simulated person a random ability score, according to some distribution. In most cases, you want the normal distribution, but you may try others as well. Do the simulation twice: once for both groups being equal, once for one of them being on the average better than the other. The recruitment process consists of making candidates pass an exam and adding a bias to one of the groups, then picking X top scoring candidates.
It's obvious that with equal groups any discrimination will decrease the total utility: you'd replace some candidates from the group you discriminate against with worse-qualified ones from the group you give preferential treatment to. It's somewhat less obvious this happens also when there is an actual difference between abilities of both groups -- and if you run such simulations, you'll see this happens for other models of racism/affirmative action as well: flat-out banning the worse group, giving them a penalty, giving them a bonus, enforcing some quotas. You get the most utility for being strictly gender/race/Zodiac sign/colour of underwear/etc blind.
It's an unpopular thing to say around here, but there are actual differences in intelligence between races: Ashkenazi jews > asians > whites > hispanics > blacks, and men > women. Does this mean that the highest measured IQ should belong to an Israeli male? Uhm no, it's a white woman without noticeable jewish ancestry. Comparing the averages for two populations doesn't predict outliers -- and when looking for good engineers, you care about a small portion at the top rather than the bulk of the population. With the normal distribution, though, you can expect a group with lower average to have a smaller representation among the top -- this is natural, and not your fault.
There are some non-genetic causes that can be rectified, like giving "boy toys" to girls, disrupting ghettos/"white trash", etc, but it's far too late to attempt such things by high school or university. At that time, loss of IQ due to development causes is undistinguishable for that due to genetics.
Obviously, this is not a reason not to fire jerks who create a hostile workplace for those women/etc who proved their superiority by passing an unbiased entrance exam. That woman is in some regards better than you, as she rose higher above the average than you did, so do respect her. After both of you passed the same barrier, there is no reason whatsoever to assume one of you is more able, according to statistics. If your employer did the right thing and avoided biases, you don't have a right to introduce biases of your own.
TL;DR: any form of discrimination by anything else than pure ability is not only evil, but also harmful.
Taking it home, while strictly speaking not ok, is something that shouldn't raise much noise after all those years. On the other hand, he intends to publicly sell it, which may make Sega rightfully angry.
Why would you ever install the same program for two architectures at once, barring an interpreter that heavily uses binary modules? That's not supported by multiarch, and it's explicitely a non-goal.
armhf+armel multiarch seems quite pointless to me as well, although it is supported. I for one have both i386 and armhf enabled on my amd64 box and i386 on my armhf one (you need external patches for qemu so wine can work, though).
IIRC, the default DE for Debian 7 will be XFCE, not GNOME.
Sadly, that change has been reverted, and wheezy installs the steaming pile of crap by default. To get working GNOME, you need to pull MATE from an external repository.
Actually, that's untrue -- most developers don't even know the version number, operating exclusively on the name. "7.0" is just a random number someone in the release times increments in an arbitrary manner and slaps on. This pops up on debian-devel roughly once a year, but a flamewar never lasts long.
Besides, 2000 < XP < Vista < 7 < 8 is that obvious and sortable too, isn't it?
I was quoting Ciaphas Cain books, which often make fun transplants of M2 customs into M41. And for Sinuhe, I can't seem to find anything similar by a text search (Day of the False King seems to be too far, and unrelevant), and it's been >20 years since I've read it (a great book indeed, by the way). Care to point out what joke you understood?
This is not a case of Paracelsian "the dose makes the poison", these amounts are already harmful. Yet for the seeds, it's a hint that something bad happened to already grown plants in the area, and that if they germinate right now, they will have an opening they can grow unopposed into. This is a gamble, yeah -- the harmful agent might be still there in several days when it can possibly hurt the sapling, but considering how small a fraction of seeds get to produce a viable plant normally instead of being blocked by others, it's like Emperor's Day came early.
There are many places in America where gays can marry, and more states are considering it. We are moving in the right direction.
The same group that pushes for gays to marry also presses the hardest to outlaw polygamy, and 1-on-1 marriage between biological adults. The latter even carries massive prison sentences, and at least brands you for life.
Would you say that "light" prison rape without permanent harm isn't bad? No matter what criminals from MAFIAA say, DRM kills our basic rights, and deprives us of culture. And unlike lives, culture can last virtually forever. Killing a person is a worse act than burning a book, but burn enough books and I'll deem you worse than a murderer. Especially if you ensure no other copy of that book survives, which is a common side effect of DRM.
And this is why extremism, even with the best intentions is also bad.
Why are you so adamant about forbidding slavery? You don't want people to be able to secure cheap work force without the risk of employees quitting? Come on, learn to compromise!
And HDCP is solved by any camera. You can increase the camera's resolution (expensive) or use multiple cheap ones, combining their output with some math, and thus obtain any quality you want up to being pixel-perfect with the source.
High quality audio recording is even easier: just buy any non-shitty piece of gear. Unless you screw it up somehow, it will record better than human ear can possibly discern, out of the box.
So you say I should take elaborate steps just to work around damage done by greylisting? So what about just not installing the damn thing in the first place?
A text message can't include any relevant data, a voice mail takes ages to listen to. Email suffers no such flaws, is reliable and instanteous (the last part, in absence of greylisting).
A phone doesn't queue, can't handle you being in the loo, and so on. And for dealing with live people, it's usually "the program left a crash dump here and there, just mail it to me" -- ie, the mail is in addition to the phone call. And at that point, the live human already knows you are aware, so you're adding an hour to your response time for no gain (greylisting's 90% rejected spam doesn't add up with most other techniques).
... and all mails you get will be delayed by an hour or more, pretty unacceptable when you get an urgent complaint that something is down. And even in not work-related matters, making people wait for no reason is rude.
There are many spam fighting techniques without such flaws. And other than gmail, server admins are generally smart enough to handle failures properly (ie, with instant notification that something went wrong).
Slashdot Mirror
It doesn't need even those 5 years. Elop's rule started only half that time ago.
The main problem is: with a gun, you have a good chance of defeating the mugger[s]. When unarmed, about none. With the police being as unreliable as they are, there's hardly any deterrent to mugging.
Newsflash: knives can kill, a broken bottle can kill, a brick to the head can kill, and if the muggers feel like having fun, even bare hands can kill.
These machines at least can give terrorists cancer, witching sticks are not even good at beating someone. You don't want terrorists to have cancer?
From my own benchmarks, clang behaves like gcc at one optimization level lower, both for compilation speed and speed of the resulting binary.
It's support for C++ has been rife with bugs, but things have improved a lot, so clang gets roughly close to gcc (other than highest, rarely used levels of optimization), except for two areas: warnings (clang produces tons of bogus ones), and portability (clang supports only i386, amd64, 32 bit arm, big-endian mips, s390, while gcc does pretty much everything).
How exactly is that cross-compilation?
I can build binaries for any other system from any other system. Except for OS X, where you need someone with that exact piece of overpriced junk, and it's hard to find such a person who's willing to run nightlies for you. If you'd want to prove the last part incorrect, it would be nice if someone could run Dungeon Crawl builds at least weekly.
For any other modern system, you either have nicely packaged cross toolchain, or could build it yourself with not much hassle but most folks don't bother as it's easier to run a virtual machine at no cost.
I'm afraid you put political correctness above data. Here's why races being equal is impossible:
There's a vast number of genetical variation between people: some are stronger, some weaker, some have bigger jaws, some blond hair, some... and so on. It's quite obvious at least some mental attributes have genetic variation.
Now let's have two vectors of numbers, one labelled "intelligence", second "skin pigmentation". You can use a more complex graph, of course, but a linear one is enough. Initialize all values to the same number. Now, in every iteration, randomly select an entry, then randomly either put it halfway towards one of its neighbours, or add or subtract a random value. Repeat for umpteen zillion of iterations.
Now check if these vectors show correlation. Unless you are extremely lucky and they happen to zero out by chance, there will be significant correlation, even though there's obviously no causation (at no point values of "intelligence" and "pigmentation" ever depended on one another).
Unlike this simulation, in real life it's possible we have at least common causes: survival in temperate regions that suffer from winters requires more preparation than savannahs and jungles, yet as I've shown above, even in an complete absence of causation correlation is extremely likely.
This doesn't mean a given person from one group is necessarily stronger/stupider/bigger-jawed than one from the other, thanks to individual variation. Yet at least the averages between two groups will show correlation with many unrelated attributes.
A "positive" discrimination is still discrimination.
Any kind of affirmative action is harmful not only to potential employees involved, but also to the employer as well -- and I'm not speaking about what is "moral", but actual utility.
Try a simulation: assign every simulated person a random ability score, according to some distribution. In most cases, you want the normal distribution, but you may try others as well. Do the simulation twice: once for both groups being equal, once for one of them being on the average better than the other. The recruitment process consists of making candidates pass an exam and adding a bias to one of the groups, then picking X top scoring candidates.
It's obvious that with equal groups any discrimination will decrease the total utility: you'd replace some candidates from the group you discriminate against with worse-qualified ones from the group you give preferential treatment to. It's somewhat less obvious this happens also when there is an actual difference between abilities of both groups -- and if you run such simulations, you'll see this happens for other models of racism/affirmative action as well: flat-out banning the worse group, giving them a penalty, giving them a bonus, enforcing some quotas. You get the most utility for being strictly gender/race/Zodiac sign/colour of underwear/etc blind.
It's an unpopular thing to say around here, but there are actual differences in intelligence between races: Ashkenazi jews > asians > whites > hispanics > blacks, and men > women. Does this mean that the highest measured IQ should belong to an Israeli male? Uhm no, it's a white woman without noticeable jewish ancestry. Comparing the averages for two populations doesn't predict outliers -- and when looking for good engineers, you care about a small portion at the top rather than the bulk of the population. With the normal distribution, though, you can expect a group with lower average to have a smaller representation among the top -- this is natural, and not your fault.
There are some non-genetic causes that can be rectified, like giving "boy toys" to girls, disrupting ghettos/"white trash", etc, but it's far too late to attempt such things by high school or university. At that time, loss of IQ due to development causes is undistinguishable for that due to genetics.
Obviously, this is not a reason not to fire jerks who create a hostile workplace for those women/etc who proved their superiority by passing an unbiased entrance exam. That woman is in some regards better than you, as she rose higher above the average than you did, so do respect her. After both of you passed the same barrier, there is no reason whatsoever to assume one of you is more able, according to statistics. If your employer did the right thing and avoided biases, you don't have a right to introduce biases of your own.
TL;DR: any form of discrimination by anything else than pure ability is not only evil, but also harmful.
Taking it home, while strictly speaking not ok, is something that shouldn't raise much noise after all those years. On the other hand, he intends to publicly sell it, which may make Sega rightfully angry.
Until you find your own piece of software closed so you can't use it, and a majority of users switch to that enhanced version by FooCorp.
Or, try to cross-compile anything to OS X.
Why would you ever install the same program for two architectures at once, barring an interpreter that heavily uses binary modules? That's not supported by multiarch, and it's explicitely a non-goal.
armhf+armel multiarch seems quite pointless to me as well, although it is supported. I for one have both i386 and armhf enabled on my amd64 box and i386 on my armhf one (you need external patches for qemu so wine can work, though).
IIRC, the default DE for Debian 7 will be XFCE, not GNOME.
Sadly, that change has been reverted, and wheezy installs the steaming pile of crap by default. To get working GNOME, you need to pull MATE from an external repository.
Actually, that's untrue -- most developers don't even know the version number, operating exclusively on the name. "7.0" is just a random number someone in the release times increments in an arbitrary manner and slaps on. This pops up on debian-devel roughly once a year, but a flamewar never lasts long.
Besides, 2000 < XP < Vista < 7 < 8 is that obvious and sortable too, isn't it?
I was quoting Ciaphas Cain books, which often make fun transplants of M2 customs into M41. And for Sinuhe, I can't seem to find anything similar by a text search (Day of the False King seems to be too far, and unrelevant), and it's been >20 years since I've read it (a great book indeed, by the way). Care to point out what joke you understood?
This is not a case of Paracelsian "the dose makes the poison", these amounts are already harmful. Yet for the seeds, it's a hint that something bad happened to already grown plants in the area, and that if they germinate right now, they will have an opening they can grow unopposed into. This is a gamble, yeah -- the harmful agent might be still there in several days when it can possibly hurt the sapling, but considering how small a fraction of seeds get to produce a viable plant normally instead of being blocked by others, it's like Emperor's Day came early.
Democrats were the [...] anti-racism party
Ah, the 1860s...
Unless you are gay and want to marry.
There are many places in America where gays can marry, and more states are considering it. We are moving in the right direction.
The same group that pushes for gays to marry also presses the hardest to outlaw polygamy, and 1-on-1 marriage between biological adults. The latter even carries massive prison sentences, and at least brands you for life.
Lightweight, non-intrusive DRM isn't bad.
Would you say that "light" prison rape without permanent harm isn't bad? No matter what criminals from MAFIAA say, DRM kills our basic rights, and deprives us of culture. And unlike lives, culture can last virtually forever. Killing a person is a worse act than burning a book, but burn enough books and I'll deem you worse than a murderer. Especially if you ensure no other copy of that book survives, which is a common side effect of DRM.
And this is why extremism, even with the best intentions is also bad.
Why are you so adamant about forbidding slavery? You don't want people to be able to secure cheap work force without the risk of employees quitting? Come on, learn to compromise!
And HDCP is solved by any camera. You can increase the camera's resolution (expensive) or use multiple cheap ones, combining their output with some math, and thus obtain any quality you want up to being pixel-perfect with the source.
High quality audio recording is even easier: just buy any non-shitty piece of gear. Unless you screw it up somehow, it will record better than human ear can possibly discern, out of the box.
So you say I should take elaborate steps just to work around damage done by greylisting? So what about just not installing the damn thing in the first place?
A text message can't include any relevant data, a voice mail takes ages to listen to. Email suffers no such flaws, is reliable and instanteous (the last part, in absence of greylisting).
A phone doesn't queue, can't handle you being in the loo, and so on. And for dealing with live people, it's usually "the program left a crash dump here and there, just mail it to me" -- ie, the mail is in addition to the phone call. And at that point, the live human already knows you are aware, so you're adding an hour to your response time for no gain (greylisting's 90% rejected spam doesn't add up with most other techniques).
So you get breakage notifications more often than once per few weeks, for every source you have? Impressive.
... and all mails you get will be delayed by an hour or more, pretty unacceptable when you get an urgent complaint that something is down. And even in not work-related matters, making people wait for no reason is rude.
There are many spam fighting techniques without such flaws. And other than gmail, server admins are generally smart enough to handle failures properly (ie, with instant notification that something went wrong).