No, it doesn't. If you make a fair system where everyone who does more gets more, you don't have a problem.
That would be nice, wouldn't it. In practice, weighing contribution of different people is almost never straightforward. Some jobs come fairly close, like a team of salespeople all selling the same product in the same territory - those are the jobs that tend to use comission. Yet even in that most simple of cases, there are problems. Companies trying to establish good reputations tend to shy away from straight comission pay because it makes salespeople do things that are bad for the company in the medium and long term. And 95% of us do jobs whose value is much harder to calculate, because no 2 people on the team are doing quite the same thing.
Of course, one "solution" is to choose a simple, arbitrary policy and assume whatever it does is fair by definition. Pure communists and capitalists both fall into this camp. They see no problem so long as the Party or the invisible hand is doing whatever it does. But of course few people hold on to such a simplistic view.
That remains to be seen. It's true that there would be no reason not to buy a Mac *if* it could run Windows as well, but so long as Windows on Mac hardware is a hack, buying a Mac might never be a viable option for everybody stuck on Windows.
I don't blame them for going down the tubes. The fact is, their niche doesn't exist anymore. When you look at all the top companies from 1990, hardly any are left. Those that survive are in niches which happen to still be profitable - and generally much less so (Oracle, Sun). Successfully re-inventing a sizeable business on the fly hardly ever happens. And no, I wouldn't count Apple as an example of that, unless they leave the PC business. HP has changed its business more, but they're not a standalone company anymore and also aren't doing terribly well.
There is no logical reason in my mind why apple would want palm. Apple has a very competent R&D team that could easily start from scratch and create a superior product.
The history is that Apple entered the market and failed, and Palm entered it and succeeded.
I could only imagine what the 1gb would be like, but I suppose if you wanted to give someone an iPod on the cheep or had a small music collection then it would be the best route to go.
I don't think 1 GB would be very different from 2 GB.
I'd say there are essentially only two sizes: "big" (store your whole collection) or "small" (you have to think about loading and unloading stuff).
That is the kind of thinking that produced the F-4 Phantom back in 1959, when weapons designers were sure that the era of dogfighting was over.
So? An idea premature for 1959 may now be correct. Saying there will "always" be a need for dogfighting is like saying there will always be a need for trebuchets.
The main reason we're still developing expensive new air superiority fighters like the F22 is simple: inertia.
Such a carelessness towards memory consumption would also suggest a similar lack of interest in writing code that is secure
That, I think, is wrong. Finite resources mean setting priorities. If you set hardware conservation above all else, it will come at a cost to conceptual integrity, security, usability, etc.
It's bizarre, isn't it, that we have such elaborate schemes for prioritizing processes' CPU access, yet nothing for IO access. IMHO Process priorities should apply to disk access. They should also apply to network send queues.
all too often there's a program that's spiking to 30% or more CPU intermittently, and so the program might flash at the top every now and then, but for the most part it's low on the list where you can't see it.
At each moment on a uniprocessor system, every program is either using 100% or 0% of the CPU. There's no in between. The idea of a program that never grabs 100% of the CPU on occasion is wrong.
Maybe you would be interested in running top and hitting 'T', which sorts programs by the total CPU time they've used since invocation. If you launch top with S (cumulative mode), those times will include the times of all child processes as well.
See, the thing is: these days, real-world programming skill is not about the language anymore, it's about the libraries.
You took the words right out of my mouth. This is true both for developers, and also for implementors of a language/platform. That's why there are 100 free JVMs and no complete and free CLASSPATH.
Is it guaranteed that each multi-thousand $$$ football is even used in the game once? Does the average game even have 120 plays, and do they really switch them every single play?
This sounds to me like VMWare is under a lot of pressure. I'd hate to see them go away, because we use VMWare Workstation for some pretty important stuff, and the license cost isn't that bad in the grand scheme of things anyways.
Except that the most common utility - electricity - has had tiered acess and variable rates for decades.
Really? Does your electric company make you pay extra for "TV electricity" and limit you to 4 slices of toast per day? Because that's what they're proposing.
Even if there were some point to the analogy, the bad old days aren't necessarily the best model for the future. If the phone companes had their way, we'd still be leasing home telephones for $7/mo (instead of buying them outright for the same price) and would still be banned from using "unapproved devices" like modems. (And of course some direly foolish consumers would still be trumpeting the telco's "rights" to do whatever they want with their (monopoly) network).
Me too (and it's not just my pythonesque sensibilities.) Did they really need to put a camera onboard too?
My big question (not addressed in the article) is the power supply. What's the fun if each birdphone only lasts a couple days?
I guess solar would be the first choice, but how about a little wind turbine or an induction coil spanning the bird's wings?
Is there any conflict between the government and the GPL?
I doubt it. Why would this application require modifying the kernel at all?
If so, that will be neat, becuase the government will use its sovereign powers to trump anything in the GPL.
I doubt that too. IME the US government is quite scrupulous about abiding by licenses. Anyways, they're using LynxWorks. I'm sure LynxWorks has a pretty good handle on the issues by now.
Apparently it looks exactly like an innocent tourist video
But not when you play it to a really tense and dramatic piece of background music with portions in slow-mo, as when I saw the clips on TV. Whew, scary stuff.
Of course, one "solution" is to choose a simple, arbitrary policy and assume whatever it does is fair by definition. Pure communists and capitalists both fall into this camp. They see no problem so long as the Party or the invisible hand is doing whatever it does. But of course few people hold on to such a simplistic view.
What's good for the goose is good for the gander.
I don't blame them for going down the tubes. The fact is, their niche doesn't exist anymore. When you look at all the top companies from 1990, hardly any are left. Those that survive are in niches which happen to still be profitable - and generally much less so (Oracle, Sun). Successfully re-inventing a sizeable business on the fly hardly ever happens. And no, I wouldn't count Apple as an example of that, unless they leave the PC business. HP has changed its business more, but they're not a standalone company anymore and also aren't doing terribly well.
Maybe if Palm were a little more fast-moving itself, they wouldn't be choosing between a massive rewrite and utter obsolescence right now.
If it can really beat F-15s, it could only be using standoff weapons (missiles). Surely you're not saying it can dogfight 6 F-15s at once?
I'd say there are essentially only two sizes: "big" (store your whole collection) or "small" (you have to think about loading and unloading stuff).
The main reason we're still developing expensive new air superiority fighters like the F22 is simple: inertia.
That, I think, is wrong. Finite resources mean setting priorities. If you set hardware conservation above all else, it will come at a cost to conceptual integrity, security, usability, etc.
It's bizarre, isn't it, that we have such elaborate schemes for prioritizing processes' CPU access, yet nothing for IO access. IMHO Process priorities should apply to disk access. They should also apply to network send queues.
Maybe you would be interested in running top and hitting 'T', which sorts programs by the total CPU time they've used since invocation. If you launch top with S (cumulative mode), those times will include the times of all child processes as well.
The problem is, there's just no good way to handle low memory conditions.
Those who stressed out and did well, or those who kicked back and blamed failure on others?
This sounds to me like VMWare is under a lot of pressure. I'd hate to see them go away, because we use VMWare Workstation for some pretty important stuff, and the license cost isn't that bad in the grand scheme of things anyways.
Even if there were some point to the analogy, the bad old days aren't necessarily the best model for the future. If the phone companes had their way, we'd still be leasing home telephones for $7/mo (instead of buying them outright for the same price) and would still be banned from using "unapproved devices" like modems. (And of course some direly foolish consumers would still be trumpeting the telco's "rights" to do whatever they want with their (monopoly) network).
My big question (not addressed in the article) is the power supply. What's the fun if each birdphone only lasts a couple days? I guess solar would be the first choice, but how about a little wind turbine or an induction coil spanning the bird's wings?