Right! Society should figure out how to take from each according to his ability, and give to reach according to his need. That kind of economic system would fix what ails Venezuela, to name just one country.
The company does pay for a personal security service for him, but the company's annual reports don't show that as very much ($1-2M, I think?). So even if you include benefits like that, it won't be much of a fine in Amazon's case.
So Spoke just need to get 165 people to open or close their cafe doors? Hint: That's not how static friction works.
Basically no pivot will work well for a 330-ton swinging door, and you'd need something like railcar wheels to make it roll. How well can you push a few fully loaded railway freight cars?
I don't always build environmentally friendly campuses, but when I do, the restaurant doors are 92 feet talk and weigh 330 tons. Because energy efficiency when opening them.
I am no Hillary Clinton fan, but for me, it suggested "current job", "cuba" and "cut out". What have you been searching for that made it suggest something else?
Who determines what convictions are related to the job someone is applying for? The police? They're not going to understand the details of the job or the applicant's history and current situation. They're not going to take responsibility for answering that yes/no question wrong. The authority to answer that question needs to stay with the responsibility for answering it incorrectly.
Yes, that is what "ban the box" is about. Ask these people what policies they push.
Ban-the-box laws typically make it illegal for employers to inquire about an applicant's criminal history before an interview or job offer (the details about which one varies by state; the activist group I linked to above wants to only allow it after a conditional job offer is made). They limit the kind or age of convictions that can be considered, and the categories of employers or where convictions can be considered at all. States like Hawaii (the first state to ban the box for both public and private employers) also prohibit consideration of arrests and other parts of a criminal record beyond convictions.
"Ban the box" absolutely is about putting strict limits on what employers can do with criminal record checks.
Activists in the US want to strictly limit or ban credit and criminal record checks by landlords and employers (e.g. "ban the box"), on the basis that these things "disproportionately" affect minorities -- especially blacks.
If you like your credit-check-screened apartments, you probably can't keep your credit-check-screened apartments.
The so-called "right to be forgotten" disagrees with you. Once a criminal serves their time in prison, or more likely are simply released with a caution or ASBO, they are Reformed. They will be good for their entire life, so landlords and the like have no right to know whether their would-be tenant blew up his last two apartments and left meth-lab chemicals saturating the walls to neighboring units.
Don't worry -- as evidenced in comments earlier about Ted Cruz's DNS stewardship bill, European elites would *never* do something like limit Europeans' online rights over something like criticizing religious zealots (Germany), or use their security apparatus to snoop virtually all Internet activity (UK), or outlaw the use of encryption (France), or require a three-strikes policy where someone can allege you pirated things three times to ban you from the Internet (France again). Only the American government does things like that. European governments are enlightened!</sarc>
The concept of premature optimization only assumes that software will eventually achieve representative functionality. Which it does long before anyone decides that it is "done".
Deciding what level of functionality is representative enough to start optimizing is hard. That's why the rules of optimization are:
Optimization isn't premature if it's totally awesome optimalization, am I right? I'm pretty sure some old Unix guy say "First, make it work, then make it work awesome, then make it work right."
I gave the reasons that I think a monitor with a discrete graphics card would be worse (for most of Apple's target market) than one with a well-chosen, integrated GPU. That was my consideration of the entire product, and I don't appreciate your straw-man reduction of my comment.
You could maybe install the video card adjacent to a controller board, but even so, you will need additional space beyond the card's height. I do think a 2" deeper monitor would be a problem. (Most current 27" monitors seem to be between 3" and 3.5" deep, so a 5" or 6" deep panel would look pretty hefty next to them. You might not see that as a problem, but Apple would, and apparently thinks their customers will.) If you kept the hinge for the stand near the back of the panel, you're probably moving the center of gravity for the monitor-minus-stand forward by that 2". Alternatively, you could put the GPU card at the top of the monitor, making it top-heavy instead of front-heavy, but the balance gets substantially worse either way.
The only way you'll get PC-like robustness is by screwing the GPU's rear panel into something solidly mounted. Is that something going to be attached to the controller board or to the monitor chassis? In either case, I suspect tolerance stack-up will be an issue, and the monitor will need that much more space for someone to reach in with a screwdriver.
I think you also underestimate the cooling difficulties. It is much easier to manage cooling with a known GPU on an existing controller board, because you can select heat sinks or pipes or fans that can be permanently installed and spec'ed for a single heat dissipation number, and you can make assumptions about heat dissipation through the board that PCIe card designers usually can't make. Yes, the cooling problems for a discrete GPU can be solved, but they require more work, more space, and more components than an integrated GPU would.
There may be open space inside a typical LCD enclosure, but not enough to easily add a single-slot PCIe edge card plus fans to keep it cool. I'd bet a lot of it is there for heat management. Also, I said the connector for an in-monitor card would have to either be on a riser *or* perpendicular to the display. A riser would allow it to be parallel to the main control board. Maybe you missed the parentheses and the "or" in my earlier comment that indicated that the perpendicular arrangement was not the most obvious or preferable one?
As for compatibility, I was thinking strictly of physical, power supply and heat dissipation. Software and electrical compatibility are not the big issues. Regardless, you're not likely to have space for a double-height PCIe card in an LCD that somebody would want to buy, much less want fans to cool it right behind your monitor. As mentioned elsewhere, Thunderbolt is PCIe 3.0 x4 at most, so there is little point in having a hugely powerful GPU without bandwidth to keep it busy, but a lot of users won't (or won't want to) consider that. Just like you don't.
Let me guess: You've never been (seriously, professionally) involved in developing and delivering a combined hardware/software system. That is probably why you think my concerns -- which are merely basic issues that occurred to me in about 30 seconds -- are unduly negative. There are good reasons that products involving out-of-box GPUs have not taken the world by storm. In particular, the same difficulties that make it hard to scale pixel pipes to huge bandwidths make it at least as hard to scale bidirectional bit pipes to similar bandwidths, and cable latency affects bidirectional pipes twice.
How would it be cheaper? Except for modularity, it would be a worse product. The video card would need to sit on a riser (or be perpendicular to the display). It would be bulkier, physically less robust, harder to cool, harder to manage compatibility, and probably more expensive due to the extra components and cooling.
There is probably a market for a 5K monitor with user-replaceable video card, but I think a fixed GPU makes more sense for almost all of Apple's target market.
Beyond what crunchygranola wrote, gun suicides make up slightly more than half of suicides. Maybe the increase in suicides last year was by poison or other means, but if someone tells me that gun deaths and suicides moved in the same direction (and doesn't provide more info that contradicts this), my inclination is to think most of both changes are due to gun suicides.
I doubt it. That kind of violence is a tragedy we should try to prevent, but constitutes a tiny fraction of all deaths. Gun homicides are only roughly a third of gun deaths, and firearm deaths are about 1.5% of all deaths. So the 11% increase in murders (in the ten largest US cities, according to your DailySignal link) represents some fraction of 0.35 deaths per 100,000 people -- no more than 5% of the increase in death rate, and perhaps negated by reductions in other causes of death. Again, we shouldn't ignore those deaths, but they're not a driving factor in the increase here.
If you look at the components of the increase, it does not look much like an obesity epidemic. There are increases in suicide, Alzheimer's, gun deaths (probably because of suicides), and opioid overdoses. Most of the increase was among whites, especially white women, but whites have a slightly lower obesity rate than most other racial categories in the US.
It is easy, but probably wrong, to blame this on people's bad eating habits.
The problem with your approach is that it doesn't fix the problems that this change is intended to address.
With KillUserProcesses=no, several GNOME packages keep processes running after the user logs out, when those processes should have terminated at the same time the user logged out. With KillUserProcesses=yes, systemd cleans them up instead. When people file bugs about the GNOME packages being buggy, maintainers will probably close them because KillUserProcesses=yes is the intended (and only supported) use case.
The people who expect standard behavior are hit twice: Once by having to specifically ask for that bog-standard, well-documented, entirely reasonable, behavior -- and again because some software expects their bugs to be worked around by the non-standard behavior in question.
Slashdot Mirror
Right! Society should figure out how to take from each according to his ability, and give to reach according to his need. That kind of economic system would fix what ails Venezuela, to name just one country.
Um, Bezos draws $1 in salary each year.
The company does pay for a personal security service for him, but the company's annual reports don't show that as very much ($1-2M, I think?). So even if you include benefits like that, it won't be much of a fine in Amazon's case.
Good grief, this phone keyboard is awful. That post should start "So Apple..."
So Spoke just need to get 165 people to open or close their cafe doors? Hint: That's not how static friction works.
Basically no pivot will work well for a 330-ton swinging door, and you'd need something like railcar wheels to make it roll. How well can you push a few fully loaded railway freight cars?
You did notice that he wrote "Islamism", not "Islam", right?
I don't always build environmentally friendly campuses, but when I do, the restaurant doors are 92 feet talk and weigh 330 tons. Because energy efficiency when opening them.
I am no Hillary Clinton fan, but for me, it suggested "current job", "cuba" and "cut out". What have you been searching for that made it suggest something else?
Who determines what convictions are related to the job someone is applying for? The police? They're not going to understand the details of the job or the applicant's history and current situation. They're not going to take responsibility for answering that yes/no question wrong. The authority to answer that question needs to stay with the responsibility for answering it incorrectly.
Yes, that is what "ban the box" is about. Ask these people what policies they push.
Ban-the-box laws typically make it illegal for employers to inquire about an applicant's criminal history before an interview or job offer (the details about which one varies by state; the activist group I linked to above wants to only allow it after a conditional job offer is made). They limit the kind or age of convictions that can be considered, and the categories of employers or where convictions can be considered at all. States like Hawaii (the first state to ban the box for both public and private employers) also prohibit consideration of arrests and other parts of a criminal record beyond convictions.
"Ban the box" absolutely is about putting strict limits on what employers can do with criminal record checks.
Activists in the US want to strictly limit or ban credit and criminal record checks by landlords and employers (e.g. "ban the box"), on the basis that these things "disproportionately" affect minorities -- especially blacks.
If you like your credit-check-screened apartments, you probably can't keep your credit-check-screened apartments.
The so-called "right to be forgotten" disagrees with you. Once a criminal serves their time in prison, or more likely are simply released with a caution or ASBO, they are Reformed. They will be good for their entire life, so landlords and the like have no right to know whether their would-be tenant blew up his last two apartments and left meth-lab chemicals saturating the walls to neighboring units.
Welp, there's one more country that got the emails of our BlackBerry-toting former Secretary of State. I just wish the US had a copy of them all.
Don't worry -- as evidenced in comments earlier about Ted Cruz's DNS stewardship bill, European elites would *never* do something like limit Europeans' online rights over something like criticizing religious zealots (Germany), or use their security apparatus to snoop virtually all Internet activity (UK), or outlaw the use of encryption (France), or require a three-strikes policy where someone can allege you pirated things three times to ban you from the Internet (France again). Only the American government does things like that. European governments are enlightened!</sarc>
The concept of premature optimization only assumes that software will eventually achieve representative functionality. Which it does long before anyone decides that it is "done".
Deciding what level of functionality is representative enough to start optimizing is hard. That's why the rules of optimization are:
1. Don't.
2. Don't Yet (for experts only).
Optimization isn't premature if it's totally awesome optimalization, am I right? I'm pretty sure some old Unix guy say "First, make it work, then make it work awesome, then make it work right."
I gave the reasons that I think a monitor with a discrete graphics card would be worse (for most of Apple's target market) than one with a well-chosen, integrated GPU. That was my consideration of the entire product, and I don't appreciate your straw-man reduction of my comment.
You could maybe install the video card adjacent to a controller board, but even so, you will need additional space beyond the card's height. I do think a 2" deeper monitor would be a problem. (Most current 27" monitors seem to be between 3" and 3.5" deep, so a 5" or 6" deep panel would look pretty hefty next to them. You might not see that as a problem, but Apple would, and apparently thinks their customers will.) If you kept the hinge for the stand near the back of the panel, you're probably moving the center of gravity for the monitor-minus-stand forward by that 2". Alternatively, you could put the GPU card at the top of the monitor, making it top-heavy instead of front-heavy, but the balance gets substantially worse either way.
The only way you'll get PC-like robustness is by screwing the GPU's rear panel into something solidly mounted. Is that something going to be attached to the controller board or to the monitor chassis? In either case, I suspect tolerance stack-up will be an issue, and the monitor will need that much more space for someone to reach in with a screwdriver.
I think you also underestimate the cooling difficulties. It is much easier to manage cooling with a known GPU on an existing controller board, because you can select heat sinks or pipes or fans that can be permanently installed and spec'ed for a single heat dissipation number, and you can make assumptions about heat dissipation through the board that PCIe card designers usually can't make. Yes, the cooling problems for a discrete GPU can be solved, but they require more work, more space, and more components than an integrated GPU would.
There may be open space inside a typical LCD enclosure, but not enough to easily add a single-slot PCIe edge card plus fans to keep it cool. I'd bet a lot of it is there for heat management. Also, I said the connector for an in-monitor card would have to either be on a riser *or* perpendicular to the display. A riser would allow it to be parallel to the main control board. Maybe you missed the parentheses and the "or" in my earlier comment that indicated that the perpendicular arrangement was not the most obvious or preferable one?
As for compatibility, I was thinking strictly of physical, power supply and heat dissipation. Software and electrical compatibility are not the big issues. Regardless, you're not likely to have space for a double-height PCIe card in an LCD that somebody would want to buy, much less want fans to cool it right behind your monitor. As mentioned elsewhere, Thunderbolt is PCIe 3.0 x4 at most, so there is little point in having a hugely powerful GPU without bandwidth to keep it busy, but a lot of users won't (or won't want to) consider that. Just like you don't.
Let me guess: You've never been (seriously, professionally) involved in developing and delivering a combined hardware/software system. That is probably why you think my concerns -- which are merely basic issues that occurred to me in about 30 seconds -- are unduly negative. There are good reasons that products involving out-of-box GPUs have not taken the world by storm. In particular, the same difficulties that make it hard to scale pixel pipes to huge bandwidths make it at least as hard to scale bidirectional bit pipes to similar bandwidths, and cable latency affects bidirectional pipes twice.
How would it be cheaper? Except for modularity, it would be a worse product. The video card would need to sit on a riser (or be perpendicular to the display). It would be bulkier, physically less robust, harder to cool, harder to manage compatibility, and probably more expensive due to the extra components and cooling.
There is probably a market for a 5K monitor with user-replaceable video card, but I think a fixed GPU makes more sense for almost all of Apple's target market.
Beyond what crunchygranola wrote, gun suicides make up slightly more than half of suicides. Maybe the increase in suicides last year was by poison or other means, but if someone tells me that gun deaths and suicides moved in the same direction (and doesn't provide more info that contradicts this), my inclination is to think most of both changes are due to gun suicides.
I doubt it. That kind of violence is a tragedy we should try to prevent, but constitutes a tiny fraction of all deaths. Gun homicides are only roughly a third of gun deaths, and firearm deaths are about 1.5% of all deaths. So the 11% increase in murders (in the ten largest US cities, according to your DailySignal link) represents some fraction of 0.35 deaths per 100,000 people -- no more than 5% of the increase in death rate, and perhaps negated by reductions in other causes of death. Again, we shouldn't ignore those deaths, but they're not a driving factor in the increase here.
Take two Twitter and call me in the morning. Classic doctor's advice.
If you look at the components of the increase, it does not look much like an obesity epidemic. There are increases in suicide, Alzheimer's, gun deaths (probably because of suicides), and opioid overdoses. Most of the increase was among whites, especially white women, but whites have a slightly lower obesity rate than most other racial categories in the US.
It is easy, but probably wrong, to blame this on people's bad eating habits.
systemd could be modified to do that, but it would still break tmux, dvtm, and millions of site-specific programs.
The problem is that systemd's new default is to also kill things that were started with nohup.
The problem with your approach is that it doesn't fix the problems that this change is intended to address.
With KillUserProcesses=no, several GNOME packages keep processes running after the user logs out, when those processes should have terminated at the same time the user logged out. With KillUserProcesses=yes, systemd cleans them up instead. When people file bugs about the GNOME packages being buggy, maintainers will probably close them because KillUserProcesses=yes is the intended (and only supported) use case.
The people who expect standard behavior are hit twice: Once by having to specifically ask for that bog-standard, well-documented, entirely reasonable, behavior -- and again because some software expects their bugs to be worked around by the non-standard behavior in question.