There's nothing special about Retina displays. They're not a new technology or anything like that, it's just a marketing label applied to displays with a certain pixel density. Nobody was competing on PPI before that because the market didn't care. It was only after Apple applied their marketing whammy that people started asking for it.
Oh please how many people have we got here on slashdot that cry over 1920x1200 vs 1920x1080? The market didn't care given the ridiculous premiums, sure IBM offered a 4K monitor back in 2001 for $18k - I wonder why people didn't buy it? By your standards Henry Ford did nothing special, the car was well invented before he got involved. Oh and OS support, also had nothing to do with it. It was absolutely all Apple's marketing machine.
As far as I understand it, the primary reason to send telescopes into space is because the atmosphere is opaque to certain wavelengths. There's also distortions caused by the atmosphere for other wavelengths, but we've found number crunching techniques that are cheaper than sending them out into space because on the ground we can build ridiculously sized telescopes like the 2800 ton E-ELT.
No, my Samsung S-IPS monitor was already 'Retina' well before Retina was ever a buzzword. That was a few years before the iPad. 6.3M physical pixels in a 32" screen. 206 PPI. Too bad it was stuck at 1920x1080 and not 5760x3240, though I bet a firmware modification could give it that performance.
Uh, so a firmware upgrade would turn your 6.3M physical pixels into a 5760x3240 = 18.7MP display? And LCD panels have only one native resolution, so if it was 1920x1080 then I'm guessing your 6,3 million "pixels" is 1920x1080xRGB? I'm sorry, but that's not the way we count them...
Now now, "high-end" is too strong of a word for what apple makes. They send expensive boutique computers with a generous markup, but the hardware is far from high-end.
A few years ago I would have agreed with you, but thinks like the iPad 3's screen or the Retina MBP were pretty unrivaled when they were released. In return, you get gouged pretty good on things like RAM or storage upgrades that are bog standard tech.
It depends on your business which metric is meaningful. E.g. for a global bank, quality is more important than time to market. Make sure that your business really gains something by playing release time against quality.
Actually, I was at a seminar earlier today and that's really not true. Of course the core systems for your bank balance and online bank must have very high quality, but analysts, sales, marketing and management only care about big numbers, trends and probabilities and there's very much information that just has to be mostly right and that's all the information they need going forwards, how much money people have in the bank right now or used to have is of course important to running the business but most the effort goes into creating new business or increasing the profitability of existing customers. That the bank works is like having a an engine that runs, you can't sell a car without it but it's not really a big selling point by itself.
IMHO, you can't. Having a deadline every week, programmers cut corners to make dates.
More than if they risk missing the spring/fall release? Because that's really the question, deadlines always makes you cut corners but does fewer and bigger work better than fewer and smaller? I'm not so sure, I've seen people work on compliance projects where it's this one huge, absolute deadline and from what I hear it can get really hairy on the final stretch.
I have a feeling this is very much a "you can't eat you cake and have it too" situation... Linux has managed to badger many companies into eventually providing open source drivers for their various sound cards and network cards and raid controllers and wifi cards and whatnot. I admit, I have used the nVidia drivers quite a bit I'm glad it's not two dozen binary blobs around an open source mini-kernel, which is almost certainly what would have happened if Linux had a well defined ABI from the start. Graphics cards have by far been the hardest since they're huge drivers on top of a quickly changing and complex hardware, far exceeding anything else but they're also now almost alone in needing binary drivers.
Intel's driver are open source as always while their performance is starting to not suck horribly, AMD's open source efforts are ongoing, even nVidia has said they'd release some tidbits of 2D specifications at least. If you're going for the quick win of give me binaries and give them to me now, why pick this specific point in Linux history? Linux on the desktop is not growing, but it's not dying either and if you include Android then it's absolutely not dying on mobile. With the latest Mesa release Intel is up to OpenGL 3.1 support (3.0 on most other cards) and that's 3.5 years behind the standard - and full Open GL ES 2.0 support. Great? No. That terrible? Also no. Linux has persisted and procrastinated against the allure of binary drivers for 20 years now, if they do it a decade more I think the last pieces will fall in place.
And for a number of reason I still don't think it'll be YotLD, all open source drivers or not.
And any distro that would sign binary code coming from NVIDIA or whoever else would be mad.
Sometimes I wonder what reality Slashdot lives in, a distro signing a nVidia driver would be no different than say Microsoft signing a nVidia driver. Of course the puritan distributions won't but they don't ship proprietary code anyway, it's not like Ubuntu would "vouch" more for nVidia than Microsoft does. If the nVidia code is full of shitty security holes that's not really the distro's problem as long as it's "genuine" nVidia code.
No last thing I want is to get my thoughts spelled out on a screen when my secretary is bending over to reach for my pen/calculator/notebook/mobile, that I accidentally dropped. In front of her. For the 5th time. That hour.
If it's the fifth time that hour she's reading it and she's still bending over, it might be time to take a hint...
I would like to counter that point with a reference to the Debian Manifesto. It was published in 1994, and Debian, in addition to becoming one of the most popular distributions by itself, came to form the basis behind products like Ubuntu, Mint, Nokia Maemo, Knoppix and Mepis. So for them it seems to have worked.
Except the Debian Manifesto is pretty to the point about WHAT they want to build, WHY and HOW. The KDE manifesto is more a fluffy value statement that pretty much wraps being a community-driven open source project in pretty words. It reminds me of certain corporate value statements, like saying we have 7 values and they are honesty, boldness, trust, freedom, team spirit, modesty and fun. Great, but do you have any fogging idea where they're going with that? Or even what line of business they're in? Well okay buzzword bingo isn't that far off, it's a consulting company. The point is that if you cover over "KDE" I haven't got a clue whose manifesto this is, it could be practically any open source project. There's nothing here about who they are or where they're going.
Middle schoolers with a passion for programming will only get to exercise their passion in the tightly controlled environment of their school's computer lab, using the language their teacher demands they use. Programmers will use $10k computers with special licensing structures that most individuals cannot afford.
Yeah uh huh, because device manufacturers in the future won't need developers to write apps to sell their devices. All the SDKs will go away, companies will stop giving away free developer tools and all interfaces will only be available under NDAs *rollseyes*. Oh I'm sure there'll still be a war on jailbreaking and DRM with even more crypto signed shit you can't touch, but beyond that I think they'll all be chanting developers, developers, developers because they make apps, apps, apps that bring profit, profit, profit.
As long as desktop hardware is cheaper than comparable laptop/portable hardware, it will have a niche. You can hook up all the docking stations and external monitors in the world to your tablet, but a desktop rig will have more storage, more memory, more GHz and better longevity (if only due to superior air flow) at a lower cost.
Ah, but desktops don't have the advantage of shipping volume, laptops do by about a 2:1 advantage. What does that mean? That Intel sets their development goals to make Haswell a better mobile CPU, not particularly a desktop GPU. Since you mention RAM as an example, I find 2x8GB desktop memory for 734 NOK (includes VAT, so please don't compare to US prices) and 2x8GB laptop memory (SO-DIMMs) for 749 NOK. What's the price advantage? Zero. Sure you can get 32/64GB RAM in a desktop that you can't in a laptop but that's a niche workstation market. You can get up to a 1TB SSD in a 2.5" laptop form factor and it doesn't get cheaper using a 3.5" desktop form factor, you just add empty space.
Sure a few things are better, you can't turn a laptop into a 10TB storage server and for a fully decked out 100W processor + 2x300W graphics cards in CF/SLI there's no laptop replacement. But those uses aren't enough to sustain the market share and the less people use desktops, the less support and higher premiums it gets. To cut costs you're pretty soon getting a laptop core in a desktop shell, which again cuts the "true" desktop market of dedicated desktop parts even more. I wouldn't not be so sure that desktop parts will always remain cheaper.
That's not even getting into the ability to customize and replace hardware without a dozen proprietary bits.
Ah, let me bring you back to the 80s, when the motherboard really was the motherboard full of daughter cards. Your network and sound and disk controllers and so on all had their own card, so you could mix and match and replace parts. I could actually go even further back, when people fixed solders and replaced chips on the individual cards too. What happened? Integration. More and more got on the motherboard and if the motherboard broke, get a new one. All-in-one volume and convenience beat flexibility and repair ability. Laptops are the same, volume and convenience of an integrated laptop means the market doesn't care or at least not enough.
But I think a sparse Mars colony would be perfect for my socialphobia. I'd only have to get used to at most twenty people.
Except you'll be living practically on top of each other, if it's only the icebreaker phase then sure but I'd go completely bonkers if all I had to myself was a small alcove but otherwise there's always be people - the same people - and there's nowhere to go to get away.
I think we're going there eventually... I mean the CPU, GPU, RAM etc. keeps getting better - even smart phones have a gigabyte of RAM these days. They won't be doing anything heavy but for light workloads, the hardest is probably watching movies on YouTube but the latest generation of phones can already decode and stream 1080p to a TV like for example here. How much more power would the average non-gamer really need? I've been thinking about replacing one of my parents' boxes with a zbox nano which is pretty much as minimal as you get but in reality it's still overkill for their needs. As long as they got to use a big screen and a big keyboard a smart phone would be plenty.
The city determined that was fair market rate because they were just going to tear the house down anyway, and nobody else would offer anywhere near $35k for a house that was going to be torn down.
Of course they wouldn't, but that's a fucked up way to calculate a price. The price should at the very least be what it would have been, if the government wasn't trying to expropriate it. I'm not entirely opposed to public expropriation, but it should absolutely not be a way for the government to get property at below market value.
I think that highly depends on whether there's any more engine failures on the next flights. If it seems like an odd case and SpaceX can say that "and even such a thing were to happen again, we'd catch it" then all is well. If another one fails and it smells more like "our engines aren't exactly 100% reliable, but we're betting on statistics that two of them won't fail on the same flight" then that's not good.
* C _forces_ a programmer to _always_ be thinking about efficiency.
* C++ _allows_ programmers to be _sloppy_ and not even bother to _think_ about efficiency.
When you have idiots calling a virtual function inside an inner loop because they don't know how a virtual function is _implemented_ that is PRECISELY the type of programmer Linus says is a crap programmer because they have never learnt the 0th rule of programming: TINSTAAFL
There Is No Such Things As A Free Lunch
And if absolute performance was the only metric the Linux kernel should be written in hand-optimized assembler. I've worked on database tables that have hundreds of millions of rows - then you'd better be real smart about how you query and join them. But I've also written a ton of queries for a bunch of tables that don't have more than 100 rows and never will, then I don't care if it does it smart. I don't care if the tables are properly indexed. It's a report nobody cares if finishes in two seconds or five because they run it once a month, they just care that it's correct. They just care that it's understandable so someone else could modify the code.
A language that _always_ forces you to think about efficiency will often force you to focus on the wrong thing. I'll gladly admit that I in general don't write the most optimal code. For example I don't try to skimp one byte here and two bytes there on whether a int16 instead of an int(32) will do, for one it's not worth the time for the second the cost if this dies in production because somebody, somewhere managed to hit that limit it's not worth it. I often pass around larger structures than strictly necessary because it's easier to expand a structure of things that naturally belong together than to change function definitions all over the place because I figured I need value X over in function Y. I don't always care if I pass by value or const ref for a microscopic performance difference.
Of course the kernel *does* have a lot of performance critical code. But I care more that I plug in a device and it works than if it uses 3% more CPU than it should. Maybe it's a project that can justify an "efficient code or not at all" policy but it's in the 1% of open source projects that can. Most of the rest are still on steps 1 or 2 of "make it work, make it work well, make it work fast". Seriously, this computer doesn't have any kind of breathing problems unless it's doing heavy rendering or video encoding or something, I could accept it being a bit slow if everything else was top notch.
First off i have a very hard time believing backdoors are built in the large networks they sell. In complex systems like that its next to impossible to hide things in the long run. Anything suspicious would have been found in the audits.
Dormant backdoors are very hard to find, hit the firewall with a secret knock (timing/ports/payload) and it'll magically slip through or start relaying information or run a MITM attack or shut down or start a denial of service attack at a critical moment. You don't have to be so obvious as to send regular bits and bytes, you can use timing information, create intentional bit errors in the error correction or boost/lower the signal strength a fraction to create a covert subchannel, almost everything is possible if you control the hardware on both sides. Even if they're very low bandwidth remember you can have a rather extreme military value/bit, like say the GPS coordinates of all the US airplanes and ships. In a war situation, how much would you pay to know exactly where the stealth fighters/bombers are?
Oh sure, we had a few computing classes but since I was years ahead of the class it didn't teach me anything. But I did have the displeasure of using a Tiki 100, though they were already a decade out of date by then.
Just simulate in software, if you don't care about speed but want to learn to program parallel.
I was thinking virtualization, how hard would it be to virtualize more cores than you physically have in the same VM? Just make say 16 virtual cores point to the same physical core as 16 different processes and you'll have a "64 core" machine on a quad core. Of course you'll get less total performance but you'll very quickly see if your application actually scales before you get a real massively parallel box.
Trying to get them to move to a partial fixed bid structure, where I would realize all of the scoped hours for a project even if I only work half of them.
Be careful what you wish for, as long as you are costing them per hour they are all working to reduce the number of hours you will bill them. Once you switch to a fixed bid and change orders model, nobody cares how much time you use - it's your problem not theirs. You have to get anal about what's in the specification because you're not getting paid for extra bells and whistles and they will get anal about the specification because they don't care if it takes you two weeks to implement the nice-to-have feature someone spec'ed to two hours. And you will spend lots of time arguing about what's a bug and what's a change order and whether or not the functionality they think is missing or doesn't work like they thought is according to the spec or not. I know people that have literally spent weeks working for nothing due to one badly formulated sentence in the specification that they accepted. Don't assume that just because you can finish in half time on time&material contracts that you'll finish in half time on fixed bid projects. As many companies has discovered, it's not that easy.
We have salaried employment here in Norway too for leading and particularly independent positions, you just wouldn't qualify for one.
If they're either a) counting hours or b) tie you to a partially or wholly fixed work schedule or c) expect you to be on call when they want you to work, you're disqualified. Of course they can expect you to show up for meetings or such, but if you're explicitly or implicitly tied to office hours the employer can find themselves at the wrong end of a lawsuit for back pay. In the same vein if you can only work at the office you're disqualified, if they don't acknowledge work in places they don't control you're not independent. Third and probably the biggest is that you choose your work, if you're assigned specific work instead of areas of responsibility you're not independent either.
In the US, I have the impression that making you a salaried employee is almost unconditionally an advantage for the employers, a lot less employee rights and practically no extra restrictions. In Norway, it's a lot more that you can't both have your cake and eat it too. If you want to make your employees independent, you lose a lot of the control that employers normally like to have. Thus it becomes much more of a balancing act, is this really the kind of employee you'd trust to just do good work on their own? If so here's your paycheck, you're not getting overtime or domestic travel costs and you're off the corporate leash but we'll of course be following up on the results you deliver.
Yes indeed... a short-lived experiment at a democratic republic that (a) never managed to even come close to adhering to its own constitutional authorizations, veering further and further from them by the day, and (b) is currently only able to operate courtesy of loans from other countries, and (c) is now possessed of an entirely corrupt political and legal framework, starting at the bottom (legislation) and working right up into all the various human factors (law enforcement, judges, lawyers, politicians and the sundry minions of all the foregoing.)
You mean a government that set lofty ideals and don't quite live up to them? Shocker. If I had free choice of country to live in the US wouldn't be at the top of my list, but they wouldn't be all that bad either. Sure they're young compared to my country but that's because for a thousand years Norway has mostly been populated by Norwegians, most states and empires that have mixed completely different ethnicities, religions and cultures have ripped themselves apart in far less than two hundred and then some years. It's not yet 50 years since MLK had to march on Washington and deliver his "I have a dream" speech and now there's a black guy holding the presidency. Still waiting of course for a woman, gay or Muslim but it's not supposed to be a minority freak show either, just if they happen to be the one best qualified.
Besides, the US is a big country - it's not like you have to go live in the Bible Belt, I think California or New York would be just fine. Just as economically fucked though, but I think semi-secular Europe is having the same kind of debt crisis and we still haven't had a collapse like Argentina or Zimbabwe. Wanting to spend more than you have is very little related to religion or even corruption, it's a people and their politicians deciding to live off credit card debt. Sweet for a while, until it's not. And despite all the battering the constitution is taking, well at least that means there's a constitution still resisting the government. You'd be surprised in how many other countries your rights go down without a fight...
Slashdot Mirror
There's nothing special about Retina displays. They're not a new technology or anything like that, it's just a marketing label applied to displays with a certain pixel density. Nobody was competing on PPI before that because the market didn't care. It was only after Apple applied their marketing whammy that people started asking for it.
Oh please how many people have we got here on slashdot that cry over 1920x1200 vs 1920x1080? The market didn't care given the ridiculous premiums, sure IBM offered a 4K monitor back in 2001 for $18k - I wonder why people didn't buy it? By your standards Henry Ford did nothing special, the car was well invented before he got involved. Oh and OS support, also had nothing to do with it. It was absolutely all Apple's marketing machine.
As far as I understand it, the primary reason to send telescopes into space is because the atmosphere is opaque to certain wavelengths. There's also distortions caused by the atmosphere for other wavelengths, but we've found number crunching techniques that are cheaper than sending them out into space because on the ground we can build ridiculously sized telescopes like the 2800 ton E-ELT.
No, my Samsung S-IPS monitor was already 'Retina' well before Retina was ever a buzzword. That was a few years before the iPad. 6.3M physical pixels in a 32" screen. 206 PPI. Too bad it was stuck at 1920x1080 and not 5760x3240, though I bet a firmware modification could give it that performance.
Uh, so a firmware upgrade would turn your 6.3M physical pixels into a 5760x3240 = 18.7MP display? And LCD panels have only one native resolution, so if it was 1920x1080 then I'm guessing your 6,3 million "pixels" is 1920x1080xRGB? I'm sorry, but that's not the way we count them...
Now now, "high-end" is too strong of a word for what apple makes. They send expensive boutique computers with a generous markup, but the hardware is far from high-end.
A few years ago I would have agreed with you, but thinks like the iPad 3's screen or the Retina MBP were pretty unrivaled when they were released. In return, you get gouged pretty good on things like RAM or storage upgrades that are bog standard tech.
It depends on your business which metric is meaningful. E.g. for a global bank, quality is more important than time to market. Make sure that your business really gains something by playing release time against quality.
Actually, I was at a seminar earlier today and that's really not true. Of course the core systems for your bank balance and online bank must have very high quality, but analysts, sales, marketing and management only care about big numbers, trends and probabilities and there's very much information that just has to be mostly right and that's all the information they need going forwards, how much money people have in the bank right now or used to have is of course important to running the business but most the effort goes into creating new business or increasing the profitability of existing customers. That the bank works is like having a an engine that runs, you can't sell a car without it but it's not really a big selling point by itself.
IMHO, you can't. Having a deadline every week, programmers cut corners to make dates.
More than if they risk missing the spring/fall release? Because that's really the question, deadlines always makes you cut corners but does fewer and bigger work better than fewer and smaller? I'm not so sure, I've seen people work on compliance projects where it's this one huge, absolute deadline and from what I hear it can get really hairy on the final stretch.
I have a feeling this is very much a "you can't eat you cake and have it too" situation... Linux has managed to badger many companies into eventually providing open source drivers for their various sound cards and network cards and raid controllers and wifi cards and whatnot. I admit, I have used the nVidia drivers quite a bit I'm glad it's not two dozen binary blobs around an open source mini-kernel, which is almost certainly what would have happened if Linux had a well defined ABI from the start. Graphics cards have by far been the hardest since they're huge drivers on top of a quickly changing and complex hardware, far exceeding anything else but they're also now almost alone in needing binary drivers.
Intel's driver are open source as always while their performance is starting to not suck horribly, AMD's open source efforts are ongoing, even nVidia has said they'd release some tidbits of 2D specifications at least. If you're going for the quick win of give me binaries and give them to me now, why pick this specific point in Linux history? Linux on the desktop is not growing, but it's not dying either and if you include Android then it's absolutely not dying on mobile. With the latest Mesa release Intel is up to OpenGL 3.1 support (3.0 on most other cards) and that's 3.5 years behind the standard - and full Open GL ES 2.0 support. Great? No. That terrible? Also no. Linux has persisted and procrastinated against the allure of binary drivers for 20 years now, if they do it a decade more I think the last pieces will fall in place.
And for a number of reason I still don't think it'll be YotLD, all open source drivers or not.
And any distro that would sign binary code coming from NVIDIA or whoever else would be mad.
Sometimes I wonder what reality Slashdot lives in, a distro signing a nVidia driver would be no different than say Microsoft signing a nVidia driver. Of course the puritan distributions won't but they don't ship proprietary code anyway, it's not like Ubuntu would "vouch" more for nVidia than Microsoft does. If the nVidia code is full of shitty security holes that's not really the distro's problem as long as it's "genuine" nVidia code.
It's a bit like finding the get away car for the bank job in your house and all the neighbors agree you use it to drive to work.
With the slight analogy breakdown that it could be used for a bank job while you drive it to work and most people wouldn't notice.
No last thing I want is to get my thoughts spelled out on a screen when my secretary is bending over to reach for my pen/calculator/notebook/mobile, that I accidentally dropped. In front of her. For the 5th time. That hour.
If it's the fifth time that hour she's reading it and she's still bending over, it might be time to take a hint...
I would like to counter that point with a reference to the Debian Manifesto. It was published in 1994, and Debian, in addition to becoming one of the most popular distributions by itself, came to form the basis behind products like Ubuntu, Mint, Nokia Maemo, Knoppix and Mepis. So for them it seems to have worked.
Except the Debian Manifesto is pretty to the point about WHAT they want to build, WHY and HOW. The KDE manifesto is more a fluffy value statement that pretty much wraps being a community-driven open source project in pretty words. It reminds me of certain corporate value statements, like saying we have 7 values and they are honesty, boldness, trust, freedom, team spirit, modesty and fun. Great, but do you have any fogging idea where they're going with that? Or even what line of business they're in? Well okay buzzword bingo isn't that far off, it's a consulting company. The point is that if you cover over "KDE" I haven't got a clue whose manifesto this is, it could be practically any open source project. There's nothing here about who they are or where they're going.
Middle schoolers with a passion for programming will only get to exercise their passion in the tightly controlled environment of their school's computer lab, using the language their teacher demands they use. Programmers will use $10k computers with special licensing structures that most individuals cannot afford.
Yeah uh huh, because device manufacturers in the future won't need developers to write apps to sell their devices. All the SDKs will go away, companies will stop giving away free developer tools and all interfaces will only be available under NDAs *rollseyes*. Oh I'm sure there'll still be a war on jailbreaking and DRM with even more crypto signed shit you can't touch, but beyond that I think they'll all be chanting developers, developers, developers because they make apps, apps, apps that bring profit, profit, profit.
As long as desktop hardware is cheaper than comparable laptop/portable hardware, it will have a niche. You can hook up all the docking stations and external monitors in the world to your tablet, but a desktop rig will have more storage, more memory, more GHz and better longevity (if only due to superior air flow) at a lower cost.
Ah, but desktops don't have the advantage of shipping volume, laptops do by about a 2:1 advantage. What does that mean? That Intel sets their development goals to make Haswell a better mobile CPU, not particularly a desktop GPU. Since you mention RAM as an example, I find 2x8GB desktop memory for 734 NOK (includes VAT, so please don't compare to US prices) and 2x8GB laptop memory (SO-DIMMs) for 749 NOK. What's the price advantage? Zero. Sure you can get 32/64GB RAM in a desktop that you can't in a laptop but that's a niche workstation market. You can get up to a 1TB SSD in a 2.5" laptop form factor and it doesn't get cheaper using a 3.5" desktop form factor, you just add empty space.
Sure a few things are better, you can't turn a laptop into a 10TB storage server and for a fully decked out 100W processor + 2x300W graphics cards in CF/SLI there's no laptop replacement. But those uses aren't enough to sustain the market share and the less people use desktops, the less support and higher premiums it gets. To cut costs you're pretty soon getting a laptop core in a desktop shell, which again cuts the "true" desktop market of dedicated desktop parts even more. I wouldn't not be so sure that desktop parts will always remain cheaper.
That's not even getting into the ability to customize and replace hardware without a dozen proprietary bits.
Ah, let me bring you back to the 80s, when the motherboard really was the motherboard full of daughter cards. Your network and sound and disk controllers and so on all had their own card, so you could mix and match and replace parts. I could actually go even further back, when people fixed solders and replaced chips on the individual cards too. What happened? Integration. More and more got on the motherboard and if the motherboard broke, get a new one. All-in-one volume and convenience beat flexibility and repair ability. Laptops are the same, volume and convenience of an integrated laptop means the market doesn't care or at least not enough.
But I think a sparse Mars colony would be perfect for my socialphobia. I'd only have to get used to at most twenty people.
Except you'll be living practically on top of each other, if it's only the icebreaker phase then sure but I'd go completely bonkers if all I had to myself was a small alcove but otherwise there's always be people - the same people - and there's nowhere to go to get away.
I think we're going there eventually... I mean the CPU, GPU, RAM etc. keeps getting better - even smart phones have a gigabyte of RAM these days. They won't be doing anything heavy but for light workloads, the hardest is probably watching movies on YouTube but the latest generation of phones can already decode and stream 1080p to a TV like for example here. How much more power would the average non-gamer really need? I've been thinking about replacing one of my parents' boxes with a zbox nano which is pretty much as minimal as you get but in reality it's still overkill for their needs. As long as they got to use a big screen and a big keyboard a smart phone would be plenty.
The city determined that was fair market rate because they were just going to tear the house down anyway, and nobody else would offer anywhere near $35k for a house that was going to be torn down.
Of course they wouldn't, but that's a fucked up way to calculate a price. The price should at the very least be what it would have been, if the government wasn't trying to expropriate it. I'm not entirely opposed to public expropriation, but it should absolutely not be a way for the government to get property at below market value.
I think that highly depends on whether there's any more engine failures on the next flights. If it seems like an odd case and SpaceX can say that "and even such a thing were to happen again, we'd catch it" then all is well. If another one fails and it smells more like "our engines aren't exactly 100% reliable, but we're betting on statistics that two of them won't fail on the same flight" then that's not good.
* C _forces_ a programmer to _always_ be thinking about efficiency.
* C++ _allows_ programmers to be _sloppy_ and not even bother to _think_ about efficiency.
When you have idiots calling a virtual function inside an inner loop because they don't know how a virtual function is _implemented_ that is PRECISELY the type of programmer Linus says is a crap programmer because they have never learnt the 0th rule of programming: TINSTAAFL
There Is No Such Things As A Free Lunch
And if absolute performance was the only metric the Linux kernel should be written in hand-optimized assembler. I've worked on database tables that have hundreds of millions of rows - then you'd better be real smart about how you query and join them. But I've also written a ton of queries for a bunch of tables that don't have more than 100 rows and never will, then I don't care if it does it smart. I don't care if the tables are properly indexed. It's a report nobody cares if finishes in two seconds or five because they run it once a month, they just care that it's correct. They just care that it's understandable so someone else could modify the code.
A language that _always_ forces you to think about efficiency will often force you to focus on the wrong thing. I'll gladly admit that I in general don't write the most optimal code. For example I don't try to skimp one byte here and two bytes there on whether a int16 instead of an int(32) will do, for one it's not worth the time for the second the cost if this dies in production because somebody, somewhere managed to hit that limit it's not worth it. I often pass around larger structures than strictly necessary because it's easier to expand a structure of things that naturally belong together than to change function definitions all over the place because I figured I need value X over in function Y. I don't always care if I pass by value or const ref for a microscopic performance difference.
Of course the kernel *does* have a lot of performance critical code. But I care more that I plug in a device and it works than if it uses 3% more CPU than it should. Maybe it's a project that can justify an "efficient code or not at all" policy but it's in the 1% of open source projects that can. Most of the rest are still on steps 1 or 2 of "make it work, make it work well, make it work fast". Seriously, this computer doesn't have any kind of breathing problems unless it's doing heavy rendering or video encoding or something, I could accept it being a bit slow if everything else was top notch.
You want Linus to do a book project?
First off i have a very hard time believing backdoors are built in the large networks they sell. In complex systems like that its next to impossible to hide things in the long run. Anything suspicious would have been found in the audits.
Dormant backdoors are very hard to find, hit the firewall with a secret knock (timing/ports/payload) and it'll magically slip through or start relaying information or run a MITM attack or shut down or start a denial of service attack at a critical moment. You don't have to be so obvious as to send regular bits and bytes, you can use timing information, create intentional bit errors in the error correction or boost/lower the signal strength a fraction to create a covert subchannel, almost everything is possible if you control the hardware on both sides. Even if they're very low bandwidth remember you can have a rather extreme military value/bit, like say the GPS coordinates of all the US airplanes and ships. In a war situation, how much would you pay to know exactly where the stealth fighters/bombers are?
Oh sure, we had a few computing classes but since I was years ahead of the class it didn't teach me anything. But I did have the displeasure of using a Tiki 100, though they were already a decade out of date by then.
Just simulate in software, if you don't care about speed but want to learn to program parallel.
I was thinking virtualization, how hard would it be to virtualize more cores than you physically have in the same VM? Just make say 16 virtual cores point to the same physical core as 16 different processes and you'll have a "64 core" machine on a quad core. Of course you'll get less total performance but you'll very quickly see if your application actually scales before you get a real massively parallel box.
Trying to get them to move to a partial fixed bid structure, where I would realize all of the scoped hours for a project even if I only work half of them.
Be careful what you wish for, as long as you are costing them per hour they are all working to reduce the number of hours you will bill them. Once you switch to a fixed bid and change orders model, nobody cares how much time you use - it's your problem not theirs. You have to get anal about what's in the specification because you're not getting paid for extra bells and whistles and they will get anal about the specification because they don't care if it takes you two weeks to implement the nice-to-have feature someone spec'ed to two hours. And you will spend lots of time arguing about what's a bug and what's a change order and whether or not the functionality they think is missing or doesn't work like they thought is according to the spec or not. I know people that have literally spent weeks working for nothing due to one badly formulated sentence in the specification that they accepted. Don't assume that just because you can finish in half time on time&material contracts that you'll finish in half time on fixed bid projects. As many companies has discovered, it's not that easy.
We have salaried employment here in Norway too for leading and particularly independent positions, you just wouldn't qualify for one.
If they're either a) counting hours or b) tie you to a partially or wholly fixed work schedule or c) expect you to be on call when they want you to work, you're disqualified. Of course they can expect you to show up for meetings or such, but if you're explicitly or implicitly tied to office hours the employer can find themselves at the wrong end of a lawsuit for back pay. In the same vein if you can only work at the office you're disqualified, if they don't acknowledge work in places they don't control you're not independent. Third and probably the biggest is that you choose your work, if you're assigned specific work instead of areas of responsibility you're not independent either.
In the US, I have the impression that making you a salaried employee is almost unconditionally an advantage for the employers, a lot less employee rights and practically no extra restrictions. In Norway, it's a lot more that you can't both have your cake and eat it too. If you want to make your employees independent, you lose a lot of the control that employers normally like to have. Thus it becomes much more of a balancing act, is this really the kind of employee you'd trust to just do good work on their own? If so here's your paycheck, you're not getting overtime or domestic travel costs and you're off the corporate leash but we'll of course be following up on the results you deliver.
Yes indeed... a short-lived experiment at a democratic republic that (a) never managed to even come close to adhering to its own constitutional authorizations, veering further and further from them by the day, and (b) is currently only able to operate courtesy of loans from other countries, and (c) is now possessed of an entirely corrupt political and legal framework, starting at the bottom (legislation) and working right up into all the various human factors (law enforcement, judges, lawyers, politicians and the sundry minions of all the foregoing.)
You mean a government that set lofty ideals and don't quite live up to them? Shocker. If I had free choice of country to live in the US wouldn't be at the top of my list, but they wouldn't be all that bad either. Sure they're young compared to my country but that's because for a thousand years Norway has mostly been populated by Norwegians, most states and empires that have mixed completely different ethnicities, religions and cultures have ripped themselves apart in far less than two hundred and then some years. It's not yet 50 years since MLK had to march on Washington and deliver his "I have a dream" speech and now there's a black guy holding the presidency. Still waiting of course for a woman, gay or Muslim but it's not supposed to be a minority freak show either, just if they happen to be the one best qualified.
Besides, the US is a big country - it's not like you have to go live in the Bible Belt, I think California or New York would be just fine. Just as economically fucked though, but I think semi-secular Europe is having the same kind of debt crisis and we still haven't had a collapse like Argentina or Zimbabwe. Wanting to spend more than you have is very little related to religion or even corruption, it's a people and their politicians deciding to live off credit card debt. Sweet for a while, until it's not. And despite all the battering the constitution is taking, well at least that means there's a constitution still resisting the government. You'd be surprised in how many other countries your rights go down without a fight...