So, I can't imagine that most Google employees are eating TWO meals a day there - maybe a meal and a snack. So, the benefit is probably $2500/yr or so.
And that is gross income. Even if their marginal rate is 35% they'd only pay an extra $800 in actual taxes on that.
If you gave me a choice of paying for typical cafeteria fare at a typical fortune 500 at typical rates for cafeteria food (ie mediocre food at premium prices), or paying an extra $800 in taxes so that I could have gourmet food at lunch every day (just grab whatever you want and stop by for ice cream in the afternoon if you have a craving), I think I'd take the gourmet food. I pay way more than $800/yr on lunches already most likely, and I don't eat like they do at Google.
A cheese steak, drink, and mushy fries at work costs me $7.50. Gourmet food for $3.50/meal in taxes - sign me up! Oh, and if your marginal rate is lower then it is even cheaper.
There is no comparison between an arc flash and a magazine explosion. The former kills a few people. The latter would wipe out half a town if it happened at port, and shatter every window for miles.
Just watch videos of WW2 battleships firing their main guns. The shock waves are incredible, and those are just single rounds being fired (and only the force of the propellant - not the shell).
No question that ship-mounted weapons are nothing to play around with no matter what technology they employ, but an electrical weapon only contains the potential energy for a few shots most likely (the rest being generated on demand), and a magazine contains the potential energy for every shot the ship could fire without resupply.
The beam just travels up into space and beyond. Unless there was a satellite right behind the target and the beam is REALLY well-focused your missed shot will likely travel forever (assuming the universe isn't closed), maybe heating up the odd bit of intergalactic gas.
Many of the weapons on naval ships can't fire forward. Usually a ship needs to keep a threat to either side for all the weapons to come to bear. For things like anti-missile defenses you want to have the missile approach perpendicular to your course anyway. A few reasons for this:
1. You're going to be firing flares and chaff, and you want the missile to go after those, and travelling at a right angle to the missile means that the bearing angle between you and the decoys is maximized (that is, after it passes the decoys without setting off the fuse you won't be still in the path of its sensors when it comes out on the other side).
2. Many missiles use radar for guidance, and if you're travelling perpendicular to the line of travel then your relative motion is zero compared to the water around you, which means that returned signals don't have a doppler shift, which means you don't stand out nearly as much. Granted, this makes a bigger difference for aircraft.
3. If for whatever reason another wave of missiles is coming in, you want to get away from the point where you were when you were spotted. As with #1 a right angle course means you're further from the center of the target bearing. With missiles bearing matters more than distance, since the missiles just travel in a line until they spot something and then they blow it up.
A portage mirror wouldn't really take THAT much space (hint, just about every Gentoo box out there already contains one). Mirroring the distfiles would take a ton of space, but you don't have to do that as you pointed out.
Running your own tree is a good way to manage updates on a large number of servers if you're running Gentoo. You can run a known-good tree, test new tree syncs in a test environment, and so on. You can also build binary packages so that production deployments go quickly.
Why run Gentoo on servers? Well, it makes sense if you need to do something unusual. Odd configurations are usually a lot easier to pull off in Gentoo than in a distro where automagic everything gets in the way. There is also hardened - Gentoo probably has one of the better hardened configuration options out there (it goes beyond just SELinux and gives you options from the application level through the kernel).
In fact, short of being a tech demo for something that might eventually be mature, it isn't entirely clear what this system can do that any of the better regarded WWII-era light cannon(retrofitted with modern targeting systems) couldn't...
I suspect that technology demonstrator is much of this device's purpose. Using weapons in the field helps you to gain experience with practical issues that you might not forsee in a lab setting.
However, this does have some use cases. The cost to shoot down a drone or damage an attack boat with one of these things is measured in tens of dollars most likely (just the cost of electricity - not counting the cost of installing/maintaining the weapon in general). If you wanted to take out a drone otherwise you're going to have to launch a SAM at it, and those cost hundreds of thousands of dollars a shot at least - unless the thing is REALLY close you aren't going to be shooting at it with any kind of manual or close-in weapon.. I'm not sure I buy the cost argument for fast attack boats - you could just fire at those with a 50 cal and they're not going to just stick around. Maybe if you want to take them out at near-horizon ranges the laser would be a better option (otherwise you're talking radar-guided cannon fire which probably costs hundreds of dollars a shell and fires fairly quickly). However, I'm not sure what kind of range you can get with one of these things with sea spray and all that attenuating the beam.
Sure, the up-front costs make the cost argument sound pretty silly - you'd have to fire a lot of million dollar SAMs to break even with the installation cost of these. That said, they're going to be spending the R&D dollars anyway - laser weapons on ships make a lot of sense (if perfected they make almost any kind of missile or surface attack impossible - short of things like ballistic rods and such).
Having seen negotiations between little fish and big fish from the perspective of IT in a big fish, I can vouch that this is good advice.
Little fish question big fish IT policies all the time. Those who want to do business with them will push back on IT. The little fish won't get that much say in the end, but nobody is going to be upset because they questioned a policy. Sometimes they even get their way, especially if a lot of people in the big fish already think the policy is dumb but they've been looking for somebody expendable (like a little fish) to be the lightning rod when it gets challenged.
What I find most odd about the denial of 'class' status in this case is that an illegal cartel arrangement to push down wages is exactly the sort of situation where it would be very difficult for any specific employee(unless they are allowed to take their case to discovery and dig up a bunch of juicy internal documents mentioning them by name) to prove any specific salary delta between the competitive and noncompetitive situations; but it should be relatively simple(by economic modeling standards) to arrive at an approximate figure for overall savings on wages by the cartel members.
They learned their lesson from the Vioxx lawsuit.
Suppose there is a 10% chance that any individual was cheated. If you just divide that evenly across the pool then everybody gets thousands of dollars. If you instead litigate everything one case at a time then there is a 90% chance that any particular plaintiff was not cheated, and therefore the juries rule against every single one of them. The result is that nobody bothers to pursue a claim, especially after a few people lose.
Class action lawsuits are probably the best remedy out there when everybody can agree that something fishy is going on, but they can't agree on exactly who did what when to who.
The idea behind this malware is kind of neat though. It's not stealing log in credentials, so it doesn't need to do browser interception and then have the hacker physically dealing with banks. It doesn't preform ddos attacks or send spam, so it doesn't use any network resources except for talking to the command and control server. If it's written correctly, it should run at low priority with a small memory footprint. It might be using 100% CPU, but on a desktop machine, the user would probably never even know its there.
Indeed, it doesn't even need to have an exploit. If you implemented a miner in Javascript you could just stick it in an advertisement and have it crunch away in a sandbox. Granted, you couldn't keep it running when the tab is closed and it would be slow in Javascript, but it would work just fine.
Even if mining on non-specialized hardware is inefficient it doesn't cost the operator anything, and it greatly reduces their risk of being caught, assuming they don't use the stolen bitcoins in any traceable transactions (the bitcoins are always traceable, but to be caught you have to use them in some transaction that can link them up with your real-world identity).
Well, I'm sure the planning is ready - there have always been plans for NK, and no doubt they get dusted off from time to time (especially at times like these).
Logistics - sure, but the US military is basically ready to go 24x7 more-or-less. Getting all the tanks sealifted (beyond those pre-positioned - which are probably considerable) will take days, but the air war could start with fairly little notice (things like B2s are usually based out of the US anyway - they don't have to go anywhere).
Bigger issues are: 1. Cost - nobody wants to pay for yet another war. 2. Overcommitment - we're still in Iraq and Afganistan, and the US really doesn't have much left in the way of reserves (they are all deployed, aside from those who are basically resting up).
The only way the US is going to do anything is if NK starts shooting. Right now the plan seems to be to let China just starve them out until they start talking sense (China hasn't made an oil delivery to them in a month or two now).
That's one of the challenges of specialization in modern society.
Not everybody would prioritize an upper-middle income over proximity to family. Considering that it seems like 90% of the population never moves more than about 25 miles from where they were born, I'd say that those who do are an oddity, even if they probably do make 90% of the income in the US.
It is a great irony that all those revolutions in travel and communication are in some ways making it harder to stay in touch with people you care about...
This is an open source project, and the video is directed at people who work with the browser source code or who have an interested in webkit/blink. It isn't targeted at end users.
It is a rendering engine. How many end users have even heard of webkit?
Obviously the browser will work the same after as before. The only thing a user might notice is that it is faster (some day) or that it has some new feature that perhaps would not have been as easy to develop without the fork.
I for one appreciate that a company like Google takes the time to actually create videos like this. I can certainly say that my employer doesn't put half-hour videos on Youtube discussing the software engineering decisions it makes for its closed-source systems.
Once students get any information about the system, however, it's doomed - and in any case it's unlikely the system will give real, useful assistance in improving skills beyond what you'd get from a grammar checker.
Yup. I'm sure essay-writing programs optimized to various brands of essay-checking programs will be on the market soon.
Just feed it a "topic" and it will do the relevant google searches to find correlated words, and string them together into a bazillion sentences that satisfy the grading algorithm.
I am not sure there is a company where transition from engineer lead to financial lead produced any benefit to the products. And bad products push companies in death spiral.
Yup, but the average stock analyst doesn't really understand how to make products, but they certainly understand MBA-speak.
That's the problem with most big US corporations. Sure, MBAs have a valuable role in a company, but they're aren't the ones that create wealth. The most they can really accomplish is improving efficiency. Well, companies don't become great by being efficient - they become great by delivering a product people want to buy. Sometimes efficiency is a part of that product (Walmart, Amazon, etc), but you first have to have some kind of product that people want to buy.
The other big problem is lazy executives. At work when they apply Six Sigma to a problem they tell people to just think of a half dozen CTQs (quality attributes to optimize). The problem is that they will then apply that methodology to some really big process that has many outputs and which many other processes depend on and there really are more than a dozen things that matter. But, it is hard to do statistical analysis when you have 47 variables so everything gets oversimplified. The result is that you get new efficient processes that simply neglect things that are important, and down the road you start having quality problems because when you're making something complicated you really need to have good control over every step along the way. But, executives don't want to think about complex processes with complex outputs - they want a powerpoint slide with 6 bullet points on it, and they want to hear that the six things that really matter can be delivered just as well at half the cost.
Maybe I'm just not part of the target audience, but does that paragraph from Forbes actually say anything?
It says about as much as any other article on big data.
Hey, I love Hadoop, but for the most part every time I hear some executive talk about big data it is because they are paying for a landfill and they just desperately want to believe that it contains gold. There are lots of valid business reasons for data retention, but that doesn't automatically mean that if you just run the right computer program that it will make you rich.
You bring up a very good point there. Since tuition has climbed, young males (primarily) would be much more interested in a seemingly guilt-free and positive benefit behavior of purchasing an expensive car without a college degree to 'look' more impressive to the females, rather than to actually have the money to support a lifestyle with them.
Based on my observation, at that age the last thing the females you're talking about are worried about is whether the males have much in the way of money, beyond whatever money is being immediately spent on them (and while that money is required to sustain a relationship, it does little to get one started).
From what I've seen, about the only thing that matters in impressing the sort of young girls you're talking about is charisma. Money is just something their parents use to pay their bills for them - of little consequence.
Now, once you get into the late 20s and 30s things change. After the first divorce, or experience with being a single parent, or just being tired of living with their parents for a decade or having to pay the bills attitudes change. At least, for those women who are not self-sufficient (not knocking them, but they're not really the target of the comment I responded to - they certainly won't be impressed by somebody who went into debt for a fancy car and no future).
Oh, and this is by no means intended as suggesting that the males that age are on average any better. It is more a matter of kids having no sense of long-term value than a male/female thing.
You often don't get 2 years of binary compatibility, even in a 2 years old supported distro. Unless your program just depends on glibc and bundles everything else, in that case the last incompatible version is about a decade old.
A big part of this is due to how dependencies are managed. On Windows there is a defined stable ABI and a set of core libraries provided by the OS, and then everything else is always bundled with the application. It isn't uncommon to find 14 versions of zlib floating around on a Windows system as a result. On Linux a distro provides all the libraries, and an application is really just supposed to provide itself. That is what leads to all the binary incompatibility across OS versions.
As you say, if you always bundle everything but glibc the compatibility lasts quite a bit longer. And, nothing is stopping distros from shipping a few versions of glibc as well, which would stretch the compatibility WAY back. Such an approach would of course waste RAM and disk space, and mean that software would often be using old/buggy/vulnerable dependencies, and that auditing for these sorts of problems would be more painful than it is today.
The model used for Linux works really well if the source code for everything you use is available to you. That is the dominant approach on Linux, and the only model that is really interesting to most of those developing for Linux. However, if I were running a company like Canonical that wanted to make it easier to install proprietary software on Linux I'd strongly consider ensuring binary compatibility over decade-long time scales, however that has to happen.
I love Android and Chrome and most Google products, but the last thing I want is Google throwing their weight around with the OS. The whole point of Android is that it is open. If somebody thinks they can make a better browser more power to them - these aren't the days of IE6 - I don't want an "Optimized for Chrome" experience.
If they come up with some good ideas then everybody wins. Competition is good for the consumer. I like Chrome for all the automatic syncing across all my platforms, but the last thing I want is for them to stagnate due to lack of competition.
Office workers? Great, Windows is a pretty good system for that usage since office workers have admins that can unf*ck their system when they pick up a virus off browser exploits.
The average slashdotter seems to have no concept of why Windows rules in the office still (though if they keep making stupid moves they could lose that market).
The answer is simple - win32. They've basically had the same API for 20 years now, give or take. You could probably take calc.exe from windows 3.1 and run it just fine on Windows 7. You could probably do the same thing with Wordperfect from that Era, minus their funky bundled printer drivers, or MS Office from the early 90s.
The other answer is 10 year support timelines. Windows XP still gets security patches. That came out 13 years ago. Support ends a decade after the OS STOPs selling. If you look at the consumer tablet/phone market the longest support comes from Apple, and they end support about 3 years after a product STARTs selling. If you look at any non-Nexus Android device you're lucky to get updates while the thing is still on store shelves.
Businesses love this. It means that they can buy a $100,000 industrial robot and the ancient software that operates it installs just fine on a PC that still gets security updates. They can spend $50M on some ERP solution that is married to IE 6 and they get a decade to work out the upgrade path. Traditionally there just aren't surprises with MS - if you buy their OS you can upgrade it on a very relaxed timeline.
Toss in active directory and domains and management tools and such and you have a complete package.
That's why the dumbest thing MS can do is abandon win32. Oh, sure, promote the latest and greatest with devs and get people to adopt some newer solution, but keep win32 support for an extra 10 years. Does this cost MS money? Absolutely! Does it make MS more money that it costs them? Almost certainly!
If you want to sell somebody something you need to understand what makes THEIR lives easier, not yours. People complain about all the legacy stuff in Windows, but the people who pay big money for it buy it mainly because those features are there. Sure, getting $30 for every PC ever sold is a nice thing for MS, but getting that umpteen-million a year professional service contract from every big company out there is worth a lot of money too.
I could see needing software to sort out 100 RFP responses for a fighter jet (where each response is 100,000 pages including gigabytes of engineering data/diagrams).
This is basically like getting 100 CVs for a job posting. You don't even need a secretary to handle that unless you post something like this every week.
The problem with that kind of data is that it is fairly restrospective, and college tuitions have risen VERY quickly.
If you're talking about a college education from the 80s followed by the job market of the 90s, there is little question that the education was worth it. However, students today don't get to pay 80s tuitions, and when they graduate they don't get to work in the 90s job market.
Due to the huge expansions in loans colleges charge WAY more today, and they accept many more people than they used to. I'm not convinced that anybody who can get accepted to college is better off taking $150k in loans to do so, no matter what their career plans or abilities are.
Having seen many kids make college plans today, I'm greatly concerned that college is just another four years of high school for them. I don't really see the value in that kind of education - why not just improve our high schools?
But in your hypothetical scenario you forgot the $2 billion they spent on advertising and marketing.
That's in addition to the R&D costs. It only gets spent on the drugs that pan out, obviously. The relative amount spent on marketing is actually lower than in many industries (Pharma has one of the highest relative R&D spends of any industry).
Hey, I'm all for getting rid of that, but it doesn't really change the issues with the patent model.
A decade ago I would have agreed that drug companies make too much money, but if you look at the last few years the whole industry is struggling. Much of that is because those who pay for drugs are setting higher standards, which means more drugs are just killed during development (no sense getting permission to sell a drug that nobody will buy). So, to some extent the market is taking care of the problem. Pills aren't any cheaper - there are just fewer of them out there.
And this doesn't mention all the potential "negative future revenue" drugs that might have been squashed or hidden away, but that's another topic.
People always talk about that, but that makes it sound like the cure of cancer is just sitting in some lab notebook and nobody wants to bother with it. Cures sell just fine - as long as they can be patented and whatever they cure is a common malady. Sure, they don't sell as much as a treatment, but no company has a treatment for every disease, and no company has a perpetual lock on the market. Sure, maybe company A might delay the cure if they make a lot of money on the treatment. However, when the patent expires on the treatment, why would company A not market the cure if they had no better treatment to follow-up with? And, what would stop company B from coming out with the cure - they're not making any money off the treatment?
What the private model doesn't work well for are diseases that aren't common (there is no money in it), or which aren't patentable.
In any case, I'm all for public end-to-end drug R&D that goes all the way to marketed products. I'd love to see that become predominant and for most of the private industry to go away. However, you can do that without touching drug patents. Let the private companies work on the Viagras that no government lab will want to touch, and taxpayers who don't want Viagra can just not buy it. While government R&D is figuring things out the private industry can continue to come out with patented drugs, and once the government gets up to speed they'll obviously have a cost advantage and consumers will tend to prefer their treatments unless a private one really provides a clear value.
Slashdot Mirror
So, I can't imagine that most Google employees are eating TWO meals a day there - maybe a meal and a snack. So, the benefit is probably $2500/yr or so.
And that is gross income. Even if their marginal rate is 35% they'd only pay an extra $800 in actual taxes on that.
If you gave me a choice of paying for typical cafeteria fare at a typical fortune 500 at typical rates for cafeteria food (ie mediocre food at premium prices), or paying an extra $800 in taxes so that I could have gourmet food at lunch every day (just grab whatever you want and stop by for ice cream in the afternoon if you have a craving), I think I'd take the gourmet food. I pay way more than $800/yr on lunches already most likely, and I don't eat like they do at Google.
A cheese steak, drink, and mushy fries at work costs me $7.50. Gourmet food for $3.50/meal in taxes - sign me up! Oh, and if your marginal rate is lower then it is even cheaper.
There is no comparison between an arc flash and a magazine explosion. The former kills a few people. The latter would wipe out half a town if it happened at port, and shatter every window for miles.
Just watch videos of WW2 battleships firing their main guns. The shock waves are incredible, and those are just single rounds being fired (and only the force of the propellant - not the shell).
No question that ship-mounted weapons are nothing to play around with no matter what technology they employ, but an electrical weapon only contains the potential energy for a few shots most likely (the rest being generated on demand), and a magazine contains the potential energy for every shot the ship could fire without resupply.
The beam just travels up into space and beyond. Unless there was a satellite right behind the target and the beam is REALLY well-focused your missed shot will likely travel forever (assuming the universe isn't closed), maybe heating up the odd bit of intergalactic gas.
Many of the weapons on naval ships can't fire forward. Usually a ship needs to keep a threat to either side for all the weapons to come to bear. For things like anti-missile defenses you want to have the missile approach perpendicular to your course anyway. A few reasons for this:
1. You're going to be firing flares and chaff, and you want the missile to go after those, and travelling at a right angle to the missile means that the bearing angle between you and the decoys is maximized (that is, after it passes the decoys without setting off the fuse you won't be still in the path of its sensors when it comes out on the other side).
2. Many missiles use radar for guidance, and if you're travelling perpendicular to the line of travel then your relative motion is zero compared to the water around you, which means that returned signals don't have a doppler shift, which means you don't stand out nearly as much. Granted, this makes a bigger difference for aircraft.
3. If for whatever reason another wave of missiles is coming in, you want to get away from the point where you were when you were spotted. As with #1 a right angle course means you're further from the center of the target bearing. With missiles bearing matters more than distance, since the missiles just travel in a line until they spot something and then they blow it up.
Good point - wasn't really thinking of that but it would be hard to ID the bitcoins that used the botnet for aid.
A portage mirror wouldn't really take THAT much space (hint, just about every Gentoo box out there already contains one). Mirroring the distfiles would take a ton of space, but you don't have to do that as you pointed out.
Running your own tree is a good way to manage updates on a large number of servers if you're running Gentoo. You can run a known-good tree, test new tree syncs in a test environment, and so on. You can also build binary packages so that production deployments go quickly.
Why run Gentoo on servers? Well, it makes sense if you need to do something unusual. Odd configurations are usually a lot easier to pull off in Gentoo than in a distro where automagic everything gets in the way. There is also hardened - Gentoo probably has one of the better hardened configuration options out there (it goes beyond just SELinux and gives you options from the application level through the kernel).
In fact, short of being a tech demo for something that might eventually be mature, it isn't entirely clear what this system can do that any of the better regarded WWII-era light cannon(retrofitted with modern targeting systems) couldn't...
I suspect that technology demonstrator is much of this device's purpose. Using weapons in the field helps you to gain experience with practical issues that you might not forsee in a lab setting.
However, this does have some use cases. The cost to shoot down a drone or damage an attack boat with one of these things is measured in tens of dollars most likely (just the cost of electricity - not counting the cost of installing/maintaining the weapon in general). If you wanted to take out a drone otherwise you're going to have to launch a SAM at it, and those cost hundreds of thousands of dollars a shot at least - unless the thing is REALLY close you aren't going to be shooting at it with any kind of manual or close-in weapon.. I'm not sure I buy the cost argument for fast attack boats - you could just fire at those with a 50 cal and they're not going to just stick around. Maybe if you want to take them out at near-horizon ranges the laser would be a better option (otherwise you're talking radar-guided cannon fire which probably costs hundreds of dollars a shell and fires fairly quickly). However, I'm not sure what kind of range you can get with one of these things with sea spray and all that attenuating the beam.
Sure, the up-front costs make the cost argument sound pretty silly - you'd have to fire a lot of million dollar SAMs to break even with the installation cost of these. That said, they're going to be spending the R&D dollars anyway - laser weapons on ships make a lot of sense (if perfected they make almost any kind of missile or surface attack impossible - short of things like ballistic rods and such).
Having seen negotiations between little fish and big fish from the perspective of IT in a big fish, I can vouch that this is good advice.
Little fish question big fish IT policies all the time. Those who want to do business with them will push back on IT. The little fish won't get that much say in the end, but nobody is going to be upset because they questioned a policy. Sometimes they even get their way, especially if a lot of people in the big fish already think the policy is dumb but they've been looking for somebody expendable (like a little fish) to be the lightning rod when it gets challenged.
What I find most odd about the denial of 'class' status in this case is that an illegal cartel arrangement to push down wages is exactly the sort of situation where it would be very difficult for any specific employee(unless they are allowed to take their case to discovery and dig up a bunch of juicy internal documents mentioning them by name) to prove any specific salary delta between the competitive and noncompetitive situations; but it should be relatively simple(by economic modeling standards) to arrive at an approximate figure for overall savings on wages by the cartel members.
They learned their lesson from the Vioxx lawsuit.
Suppose there is a 10% chance that any individual was cheated. If you just divide that evenly across the pool then everybody gets thousands of dollars. If you instead litigate everything one case at a time then there is a 90% chance that any particular plaintiff was not cheated, and therefore the juries rule against every single one of them. The result is that nobody bothers to pursue a claim, especially after a few people lose.
Class action lawsuits are probably the best remedy out there when everybody can agree that something fishy is going on, but they can't agree on exactly who did what when to who.
The idea behind this malware is kind of neat though. It's not stealing log in credentials, so it doesn't need to do browser interception and then have the hacker physically dealing with banks. It doesn't preform ddos attacks or send spam, so it doesn't use any network resources except for talking to the command and control server. If it's written correctly, it should run at low priority with a small memory footprint. It might be using 100% CPU, but on a desktop machine, the user would probably never even know its there.
Indeed, it doesn't even need to have an exploit. If you implemented a miner in Javascript you could just stick it in an advertisement and have it crunch away in a sandbox. Granted, you couldn't keep it running when the tab is closed and it would be slow in Javascript, but it would work just fine.
Even if mining on non-specialized hardware is inefficient it doesn't cost the operator anything, and it greatly reduces their risk of being caught, assuming they don't use the stolen bitcoins in any traceable transactions (the bitcoins are always traceable, but to be caught you have to use them in some transaction that can link them up with your real-world identity).
Well, I'm sure the planning is ready - there have always been plans for NK, and no doubt they get dusted off from time to time (especially at times like these).
Logistics - sure, but the US military is basically ready to go 24x7 more-or-less. Getting all the tanks sealifted (beyond those pre-positioned - which are probably considerable) will take days, but the air war could start with fairly little notice (things like B2s are usually based out of the US anyway - they don't have to go anywhere).
Bigger issues are:
1. Cost - nobody wants to pay for yet another war.
2. Overcommitment - we're still in Iraq and Afganistan, and the US really doesn't have much left in the way of reserves (they are all deployed, aside from those who are basically resting up).
The only way the US is going to do anything is if NK starts shooting. Right now the plan seems to be to let China just starve them out until they start talking sense (China hasn't made an oil delivery to them in a month or two now).
That's one of the challenges of specialization in modern society.
Not everybody would prioritize an upper-middle income over proximity to family. Considering that it seems like 90% of the population never moves more than about 25 miles from where they were born, I'd say that those who do are an oddity, even if they probably do make 90% of the income in the US.
It is a great irony that all those revolutions in travel and communication are in some ways making it harder to stay in touch with people you care about...
This is an open source project, and the video is directed at people who work with the browser source code or who have an interested in webkit/blink. It isn't targeted at end users.
It is a rendering engine. How many end users have even heard of webkit?
Obviously the browser will work the same after as before. The only thing a user might notice is that it is faster (some day) or that it has some new feature that perhaps would not have been as easy to develop without the fork.
I for one appreciate that a company like Google takes the time to actually create videos like this. I can certainly say that my employer doesn't put half-hour videos on Youtube discussing the software engineering decisions it makes for its closed-source systems.
Once students get any information about the system, however, it's doomed - and in any case it's unlikely the system will give real, useful assistance in improving skills beyond what you'd get from a grammar checker.
Yup. I'm sure essay-writing programs optimized to various brands of essay-checking programs will be on the market soon.
Just feed it a "topic" and it will do the relevant google searches to find correlated words, and string them together into a bazillion sentences that satisfy the grading algorithm.
I am not sure there is a company where transition from engineer lead to financial lead produced any benefit to the products. And bad products push companies in death spiral.
Yup, but the average stock analyst doesn't really understand how to make products, but they certainly understand MBA-speak.
That's the problem with most big US corporations. Sure, MBAs have a valuable role in a company, but they're aren't the ones that create wealth. The most they can really accomplish is improving efficiency. Well, companies don't become great by being efficient - they become great by delivering a product people want to buy. Sometimes efficiency is a part of that product (Walmart, Amazon, etc), but you first have to have some kind of product that people want to buy.
The other big problem is lazy executives. At work when they apply Six Sigma to a problem they tell people to just think of a half dozen CTQs (quality attributes to optimize). The problem is that they will then apply that methodology to some really big process that has many outputs and which many other processes depend on and there really are more than a dozen things that matter. But, it is hard to do statistical analysis when you have 47 variables so everything gets oversimplified. The result is that you get new efficient processes that simply neglect things that are important, and down the road you start having quality problems because when you're making something complicated you really need to have good control over every step along the way. But, executives don't want to think about complex processes with complex outputs - they want a powerpoint slide with 6 bullet points on it, and they want to hear that the six things that really matter can be delivered just as well at half the cost.
Maybe I'm just not part of the target audience, but does that paragraph from Forbes actually say anything?
It says about as much as any other article on big data.
Hey, I love Hadoop, but for the most part every time I hear some executive talk about big data it is because they are paying for a landfill and they just desperately want to believe that it contains gold. There are lots of valid business reasons for data retention, but that doesn't automatically mean that if you just run the right computer program that it will make you rich.
You bring up a very good point there. Since tuition has climbed, young males (primarily) would be much more interested in a seemingly guilt-free and positive benefit behavior of purchasing an expensive car without a college degree to 'look' more impressive to the females, rather than to actually have the money to support a lifestyle with them.
Based on my observation, at that age the last thing the females you're talking about are worried about is whether the males have much in the way of money, beyond whatever money is being immediately spent on them (and while that money is required to sustain a relationship, it does little to get one started).
From what I've seen, about the only thing that matters in impressing the sort of young girls you're talking about is charisma. Money is just something their parents use to pay their bills for them - of little consequence.
Now, once you get into the late 20s and 30s things change. After the first divorce, or experience with being a single parent, or just being tired of living with their parents for a decade or having to pay the bills attitudes change. At least, for those women who are not self-sufficient (not knocking them, but they're not really the target of the comment I responded to - they certainly won't be impressed by somebody who went into debt for a fancy car and no future).
Oh, and this is by no means intended as suggesting that the males that age are on average any better. It is more a matter of kids having no sense of long-term value than a male/female thing.
You often don't get 2 years of binary compatibility, even in a 2 years old supported distro. Unless your program just depends on glibc and bundles everything else, in that case the last incompatible version is about a decade old.
A big part of this is due to how dependencies are managed. On Windows there is a defined stable ABI and a set of core libraries provided by the OS, and then everything else is always bundled with the application. It isn't uncommon to find 14 versions of zlib floating around on a Windows system as a result. On Linux a distro provides all the libraries, and an application is really just supposed to provide itself. That is what leads to all the binary incompatibility across OS versions.
As you say, if you always bundle everything but glibc the compatibility lasts quite a bit longer. And, nothing is stopping distros from shipping a few versions of glibc as well, which would stretch the compatibility WAY back. Such an approach would of course waste RAM and disk space, and mean that software would often be using old/buggy/vulnerable dependencies, and that auditing for these sorts of problems would be more painful than it is today.
The model used for Linux works really well if the source code for everything you use is available to you. That is the dominant approach on Linux, and the only model that is really interesting to most of those developing for Linux. However, if I were running a company like Canonical that wanted to make it easier to install proprietary software on Linux I'd strongly consider ensuring binary compatibility over decade-long time scales, however that has to happen.
Yup, and the advancement of web/citrix/cloud-based solutions is making the workstation OS less relevant anyway.
However, their stability was the whole advantage of Windows.
Linux isn't exactly the bedrock of stability either. No distro provides a decade of support, though RHEL does a fairly decent job.
I love Android and Chrome and most Google products, but the last thing I want is Google throwing their weight around with the OS. The whole point of Android is that it is open. If somebody thinks they can make a better browser more power to them - these aren't the days of IE6 - I don't want an "Optimized for Chrome" experience.
If they come up with some good ideas then everybody wins. Competition is good for the consumer. I like Chrome for all the automatic syncing across all my platforms, but the last thing I want is for them to stagnate due to lack of competition.
Office workers? Great, Windows is a pretty good system for that usage since office workers have admins that can unf*ck their system when they pick up a virus off browser exploits.
The average slashdotter seems to have no concept of why Windows rules in the office still (though if they keep making stupid moves they could lose that market).
The answer is simple - win32. They've basically had the same API for 20 years now, give or take. You could probably take calc.exe from windows 3.1 and run it just fine on Windows 7. You could probably do the same thing with Wordperfect from that Era, minus their funky bundled printer drivers, or MS Office from the early 90s.
The other answer is 10 year support timelines. Windows XP still gets security patches. That came out 13 years ago. Support ends a decade after the OS STOPs selling. If you look at the consumer tablet/phone market the longest support comes from Apple, and they end support about 3 years after a product STARTs selling. If you look at any non-Nexus Android device you're lucky to get updates while the thing is still on store shelves.
Businesses love this. It means that they can buy a $100,000 industrial robot and the ancient software that operates it installs just fine on a PC that still gets security updates. They can spend $50M on some ERP solution that is married to IE 6 and they get a decade to work out the upgrade path. Traditionally there just aren't surprises with MS - if you buy their OS you can upgrade it on a very relaxed timeline.
Toss in active directory and domains and management tools and such and you have a complete package.
That's why the dumbest thing MS can do is abandon win32. Oh, sure, promote the latest and greatest with devs and get people to adopt some newer solution, but keep win32 support for an extra 10 years. Does this cost MS money? Absolutely! Does it make MS more money that it costs them? Almost certainly!
If you want to sell somebody something you need to understand what makes THEIR lives easier, not yours. People complain about all the legacy stuff in Windows, but the people who pay big money for it buy it mainly because those features are there. Sure, getting $30 for every PC ever sold is a nice thing for MS, but getting that umpteen-million a year professional service contract from every big company out there is worth a lot of money too.
Google doesn't pay people to promote the Nexus devices on non-Nexus review pages.
I'm sure many of their fans do, but that isn't the same thing.
The parent post was complaining about MS actually paying people to post reviews like the this.
Yeah, this is nuts.
I could see needing software to sort out 100 RFP responses for a fighter jet (where each response is 100,000 pages including gigabytes of engineering data/diagrams).
This is basically like getting 100 CVs for a job posting. You don't even need a secretary to handle that unless you post something like this every week.
The problem with that kind of data is that it is fairly restrospective, and college tuitions have risen VERY quickly.
If you're talking about a college education from the 80s followed by the job market of the 90s, there is little question that the education was worth it. However, students today don't get to pay 80s tuitions, and when they graduate they don't get to work in the 90s job market.
Due to the huge expansions in loans colleges charge WAY more today, and they accept many more people than they used to. I'm not convinced that anybody who can get accepted to college is better off taking $150k in loans to do so, no matter what their career plans or abilities are.
Having seen many kids make college plans today, I'm greatly concerned that college is just another four years of high school for them. I don't really see the value in that kind of education - why not just improve our high schools?
But in your hypothetical scenario you forgot the $2 billion they spent on advertising and marketing.
That's in addition to the R&D costs. It only gets spent on the drugs that pan out, obviously. The relative amount spent on marketing is actually lower than in many industries (Pharma has one of the highest relative R&D spends of any industry).
Hey, I'm all for getting rid of that, but it doesn't really change the issues with the patent model.
A decade ago I would have agreed that drug companies make too much money, but if you look at the last few years the whole industry is struggling. Much of that is because those who pay for drugs are setting higher standards, which means more drugs are just killed during development (no sense getting permission to sell a drug that nobody will buy). So, to some extent the market is taking care of the problem. Pills aren't any cheaper - there are just fewer of them out there.
And this doesn't mention all the potential "negative future revenue" drugs that might have been squashed or hidden away, but that's another topic.
People always talk about that, but that makes it sound like the cure of cancer is just sitting in some lab notebook and nobody wants to bother with it. Cures sell just fine - as long as they can be patented and whatever they cure is a common malady. Sure, they don't sell as much as a treatment, but no company has a treatment for every disease, and no company has a perpetual lock on the market. Sure, maybe company A might delay the cure if they make a lot of money on the treatment. However, when the patent expires on the treatment, why would company A not market the cure if they had no better treatment to follow-up with? And, what would stop company B from coming out with the cure - they're not making any money off the treatment?
What the private model doesn't work well for are diseases that aren't common (there is no money in it), or which aren't patentable.
In any case, I'm all for public end-to-end drug R&D that goes all the way to marketed products. I'd love to see that become predominant and for most of the private industry to go away. However, you can do that without touching drug patents. Let the private companies work on the Viagras that no government lab will want to touch, and taxpayers who don't want Viagra can just not buy it. While government R&D is figuring things out the private industry can continue to come out with patented drugs, and once the government gets up to speed they'll obviously have a cost advantage and consumers will tend to prefer their treatments unless a private one really provides a clear value.