Here's a "eh" solidworks PCB. Give it a bit of cleanup, but it on a background like a model's palm/hand, put it right side up so you can read (how did that slip thru?), work a little on the textures, and its all good.
Here's a "eh" real photo of a PCB which may or may not link properly. Lets list the faults. Bad lighting leads to shadows, half of each resistor is illuminated and half in shadow. bright camera flash reflection in the center and off some solder joints. The focus plane is obviously near the camera... look at the solder joints holding the two PCBs together, you can see whats going on but they're not in sharp focus.
Now you have to realize the solidworks example is at least an attempt at a marketing mock up, and the genuine pic is merely trying to show a manufacturing fault to someone else, not create a work of art. Still, they're fairly typical examples of what you easily get with CGI vs pics.
CGI is replacing actual photos in the stock picture business, and in the catalog business. I haven't seen an office catalog in years that uses actual pictures. It's all semi-competent CGI.
I know for a fact, err, trustworthy heresay... that almost all electronic catalog "pictures" are 3-d CAD renderings, sometimes with a bit of photoshop. I'm talking about real EE component catalogs, not best buy consumer catalogs.
From trying to take pictures of things I've built, its an unholy PITA and depth of field and reflections and lighting are agonizing. You can look at the pic of a PCB, lets say a stereotypical switching power supply module, and try to figure out how I could get that depth of field and lighting without reflection issues and suddenly realize, this was done in Solidworks not a camera.
If you want to see how bad "real pictures" of electronic devices/components look, try trashy photos of that stuff on ebay. Some of those guys are obviously not even wiping the human grease off the cellphone camera lens first.
Video is a big problem for photographers and not in the way you specify.
A big part of being a photographer is getting just the right moment for the shot at the perfect angle. An instant in time.
Now you just run a high def video camera, walk around and wave the camera, and pick the best individual frame later.
Aside from weddings, this is also causing license chaos because people used to purchase and pay separate photo and video rights at sporting events... why pay for photo rights if you can just use a single video frame, and why miss the action if you can just use a video camera. So enforcement is all confused about that. The fairest way to charge for "rights" at sporting events seems to be by ounce of gear hauled into the venue. I know there have already been court cases over video camera guys selling single frames to newspapers, but I don't know how they've turned out.
Ah but those licenses were (probably not) paid by and for the previous regime... A reasonable argument could be made for the US to buy a country sized site license for all PCs in Iraq...
Historically, in the arms dealing arena, this was how "foreign assistance" worked. Congress can't hand a billion bucks to General Dynamics just for fun. But you can hand a billion bucks to Israel on the agreement that they hand that same billion bucks to General Dynamics in exchange for a couple fighter airplanes at list price.
That is probably how an Iraq re licensing deal would work. MS pays congress $1M to send $1B to Iraqi govt on the nod and handshake agreement that they spend that $1B on licenses for windows as part of "economic development" or some such B.S.
MS doesn't have to actually provide or "do" anything, they just need to convince people to send them money, which can be pretty cheap.
No, once a store is available, all you need is to have the devs behind adblock plus only distribute thru the store. I have exactly 13 addons. All I need is a couple to decide to switch from free mode to store mode and that makes FF much less useful to me.
The danger is making it necessary to switch addons makes it not look all that much harder to switch to Chrome.
For example, say somebody tries to monetize noscript or firebug... I'm sure there is a "almost as good" replacement that'll remain free on FF... but if I'm going to do the effort to substitute, I'll find a "almost as good" replacement on Chrome.
How does the interim solution get implemented if the machine won't boot?
In ye olden days, if you compiled a new linux kernel, deleted the old one, forgot to run lilo, it doesn't reboot, then the solution was to boot and run lilo, which was a task that separated the men from the boys, err it wasn't that difficult, maybe separated the 7 year olds from the 6 year olds or whatever. Anyway...
Also a note to the editors, that link would have been a million times funnier if it pointed to a ubuntu.com live cdrom/dvd/usb image, or maybe the android-x86 project, which is really quite usable.
"Most people" don't need much more than a working web browser. Even at work, "they" were recently f-ing around with a firewall and managed to block ssh, so I installed ajaxterm to work around that, and I mostly use the webmail page anyway, so all I really need at work is a working browser. I would not want to use ajaxterm on a regular basis 10 hours a day, but its survivable in a pinch until the firewall guys get their stuff together. For example I really enjoy how it floods the logs unless you do extra work....
Ballmer may not be special, maybe even lousy, but worst EVAR!!!? I would pick some of the CEOs around the world that lead us into this global recession - who not only did so but (distinguishing them from their counterparts in government) personally took home millions of tens of millions of dollars for doing so and are living lavishly to this day.
As quislings, I think the authors definition of "worst" might not quite match yours.
But even given that many of the banks would have imploded if not for the bail out, GM would be gone if not for the bail out and plenty of marginally successful companies have gone through quite a bit of economic turmoil that MS has avoided
That is the failure mode. The way "leadership" is defined, MS should have dumped a multiple of their net worth into the Zune, then paid millions in campaign contributions to politicians to get billions in bailout funds. Heads we win, tails you lose.
The landscape is quite a bit different for big companies operating under a corporate owned government, than it is for, say, a cupcake store. In the world of cupcake stores, the MS strategy IS more intelligent than the GM strategy, but this article was talking about the big companies that own the govt and order it around, not a scrapbooking supplies store or other small operators like that.
MS could have paid millions to politicians to force the.mil to buy MS licenses for every Iraqi owned PC in Iraq, that would probably have a pretty good profit. Or they could have purchased politicans to declare linux distros as hacking tools and have border control sieze any laptop with linux installed, or sieze any linux install media. So many kleptocratic ways to turn billions into trillions, and instead they... failed.
In a way, its bad news, if they get rid of Ballmer and but someone competent in his place, then the public will suffer greatly.
Its interesting watching the journalist / news release re publishers solely focus on competition from download services and totally ignore mail order.
Download took away all the "I want it on release day" and "I want it now". Mail order took away all the "I want it cheap" and "I've got metered/slow internet access" and "I want something unusual" That leaves retail stores with... gift sales? Apparently the book industry has given up on retail except for gift sales, and that works for them, but video games are too fast moving of a target.
Printing a jpeg as a PCB layout? Could happen I suppose.
I wonder if DPI is like "megapixels" or "sears air compressor horsepower" where the engineering definition no longer has any relationship to the marketing definition.
For example, marketing could sell four toner colors at 100 dpi as being "400 dpi" after all its four toners each at 100 dpi, right? Something like that would explain why the OP's printer can't successfully output at better than 100 dpi, despite marketing claiming 600 dpi. Hell I could see marketing claiming 50 dpi at 4 colors and duplex printing (two sided) as being "400 dpi".
Latency can be a huge issue. Sure, we got wifi. No we don't trust sensitive data over the air, so you gotta use the wifi, out to the internet, into the VPN concentrator... the corporate VPN concentrator... on the other side of the country... So to VNC into a local server you get millisecond latency connections over wired, or hundred ms (sometimes more) latency over the wifi. Yuck.
I specified that based on visual acuity limits. There's a lot of optical theory explaining why over 300 dpi is mostly useless for toner on paper. Unless your eyeball lens diameter is 10 times bigger than the average human or your retina cell layout is different than all known humans, it is not optically possible to resolve 3000 dpi or whatever on paper under normal conditions and lighting. Depending on how close you can hold the paper before you can't focus on it anymore, and tangentially depending on how bright the light it (little pinhole camera iris) humans top out around 300 dpi.
Now, projected thru transparencies onto a overhead, higher res works, if you have old fashioned overhead projectors and sit close to the screen. Also there are ugly aliasing and anti-aliasing effects that can be avoided by higher res with real vector scaling. And high res allows better/smoother color mixing, in that bluring together 2**8 pixels of 2**16 color is the same as one 2**24 pixel, more or less. There are also relative brightness/consistency effects where making a "line" that varies from 8 to 9 pixels wide looks a lot less consistent than a line that is 85 or 86 pixels wide at 10 times the res, look at the percentage variation of one pixel. If the lighting is really bad, there are strange shadow effects where you can perceive over 300 dpi if the shadows land just right. Also there are some strange toner based textural issues where the plastic surface of thinner lines literally looks different. And some 3-d effects of toner on paper. So over 300 dpi is not a complete waste of time, just mostly a waste with average pictures under average conditions. It would be extremely hard to justify over 1200 dpi even in the weirdest corner cases.
They'd probably resolve it by firing me, since there's industry standard press releases about how everyone should love us because we're green, blah blah. Complainers get pounded down, not rewarded. It does depend on the combination of your local corporate culture plus your ability to properly identify it, times how important of an issue it is.
For example, recently making it much harder to use the laser printers, such that people don't print stuff out before meetings, because the printers are far away, not configured on the desktops, and usually out of paper, has saved the earth a lot more than properly recycling huge amounts of paper in the olden days. Exploding the price of soda in the vending machines, crashing sales, has saved the earth a lot more than properly recycling soda cans.
They are offloading multi-threading, multiple robots in a maze doing the same task.
They are not offloading multi-tasking, like I'm currently simultaneously thinking about: 1)/. 2) a really slow data importer I should be fixing, but my brain needs to decompress to unconsciously determine the solution. 3) pr0n, of course.
Moving forward, Solovey wants to investigate other cognitive states that can be reliably detected using fNIRS. Imagine a computer that increases the size of buttons...
Um, WRT to those "buttons", all successful technology is first rolled out for pr0n...
Thats only a square 11000 pixels on a side. A 300 dpi laserprinter would make a roughly one yard/one meter printout.
At a slightly higher resolution that would be a metric A0 paper size. printers that big do exist but are kinda expensive. Best upload it to your local printer/office store and let them print it instead of do it yourself.
Ah this old idea that pops up on/. every month or so.
This is where I relate that you can get ALMOST the same experience temporarily with stretchy gloves and little magnets in the fingertips. Obviously buy gloves with fabric fingers just a little longer than your fingers. In the frozen north, in season, this is not much of a challenge... Florida/.ers are probably all like, gloves, what are those?
It is fun for at least a couple hours. Try touching the body of a big motor or old fashioned xfrmr power supply. Variable speed AC/DC "universal" motors were not as much fun as I expected. Waving in front of a CRT screen is entertaining.
You want gloves thick/strong enough to not tear, but not so thick as to lose sensation.
GOOG for "magnetic wedding ring" and you'll find lots of crackpot "magnets cured mah arthritis" pure BS, but this is probably the ultimate in non-invasive experimentation. Unlike the Magnet-in-glove thing, I've not tried magnetic wedding rings.
The biggest problem with "magnets/hands" is what happens when it inevitably cracks. Sharp little ceramic shards pinching slashed up tissue. So don't go giving steel plates a "high five".
Right now China, India, and Europe are trying to get the USA to play along on this issue. Or at least come to the table.
There is a MUCH bigger problem here. So we pass a law... whats the result?
China: lower levels simply ignore it, higher levels pay a bribe and ignore it.
India: Not quite as corrupt, but pretty much it'll be ignored.
Europe: Move the polluting industry to the EU member state promising the most lax enforcement or a tax break making up for the costs. OR simply move to China. Some net effect but lots of social upheaval.
USA: lower levels will have to follow it and/or go out of business, higher levels pay a bribe (re-election donation) and ignore it.
Coming to the table seems pointless... so our big polluters will ignore whatever's being done, and the small polluters will simply close shop and pollute even more in China...
hostile towards green initiatives solely because "fuck you"
I'm hostile toward them because I'm too smart. Most (all?) of those initiatives are meaningless feel good frippery with no real world effect, or NEGATIVE real world effect. They require stupid people. Not gonna work on me.
Example, the people who just don't give a F about recycling at the office. We are pure, refined evil, right? Where I work, they purchased extra recycling trash cans and distributed them all over and the idiots blathered on about how we're all going to save the planet by sorting our trash, printed posters hanging everywhere. No one, including myself, noticed we still only have one trash dumpster, not a trash dumpster and a recycling dumpster, how... interesting. As a tech guy I work odd hours, and I get to see the illegal alien cleaning crew pushing a big rolling trash collector around the office and dumping both trash and recycling into the same rolling collector, and that collector dumps mixed trash and recyclables into the same dumpster, and odd mornings that I'm here early I see that dumpster emptied into one trash truck. I know they're not sorting at the landfill, either. So the idiots think they're saving the world, but I know the real world effect is we turned a lot of crude oil into plastic recycling cans, waste lots of money buying those cans, waste time and money proselytizing to people, waste valuable time and money sorting trash that is going to be commingled by the cleaning crew anyway, and finally waste time emptying twice as many trash cans. What a huge amount of environmental damage to fool people into thinking they're preventing environmental damage. I'm sure some minor drone got a nice promotion out of it. So, yeah, I'm one of the insane number of "F you" people who throws my empty soda cans into the regular trash can because apparently I don't want to save the planet.
I have special knowledge about the recyclables/trash issue because of weird working hours. I would assume the same scammers running the same psychological confidence scam in other areas are also not doing anything useful for the environment. Maybe accidentally once in a while they occasionally do something useful by mistake, but on average the environment would be better served by those kind of folks if they merely piled up cash in my backyard and set it on fire. So, yeah, in general, F those people and their goals, they're all scammers.
What does work is financial. So I'm painting my house.. I could use oil and have to buy "bad for the environment" mineral oil solvent to clean the brushes, or I could use latex paint and clean the brushes with practically free water... Thats how I save the environment from hydrocarbon vapor / ozone pollution, by saving money. The actual cost of gasoline is about $8/gallon and thats what it should be at the pump, instead of being cheaper and the balance paid by "stealth" taxation. Higher gas prices would fix a lot of environmental problems.
Given assumption (which may be wrong) climate change causes $1B damage resulting in eventual death of 1M
Scenario 1) do nothing. Result $1B in hole results in 1M deaths.
Scenario 2) Social engineering has never failed in the past, so we'll give it a try. Result $10B in hole results in 10M deaths.
I mean, come on. People are asking the.gov to accomplish something. This is the.gov people, not GOOG. All they're going to do is accept money from donors, send contract money to donors, repeat, plus or minus some feel good legislation.
This is only going to be fixed by individual social pressure working on individuals. Not federal social engineering. Not solving social problems using tech.
The funny part is that this is a "tech" scandal... Supposedly anyone older than 25 is "over the hill" WRT tech and is already obsolete, yet to have lived thru the original watergate scandal you'd have to be at least 50. I mean, yeah, technically I was alive when nixon resigned, but I was only a couple months old so I didn't care too much. I figure you'd have to be at least 50 to have been paying attention.
I guess the tech connection is if you're "in tech" then ask your grandfather about putting -gate at the end of every scandal.
It could be worse, we could be going thru history reporting on water-contra, water-resume or water-yahoo or whatever.
It raises the question... given the much better featureset, why are the major Linux distributions still using GCC to compile the kernel and all the OS packages?
LOL. As if those features matter very much to almost anyone. At this time, if you're developing C, it would be a really wise idea to verify code compiles under both and actually ship GCC compiled binaries because they're better (faster). Lack of optimization under clang is a bug, not a feature, unless you're doing human-powered-optimization like refactoring etc etc.
Its unlikely you'll "ever" be able to totally drop gcc... there are features that clang architecturally opposed to. Specifically fortran/C stuff, and the alternative GCC languages in general. Also its not a binary either/or. This is hardly the first time a OS has had to deal with an alternatives system. On debian, the alternatives system is called the alternatives system. I suppose other OS might have a different named system, and it might work somewhat differently, but the concepts will all be the same.
About 1 in 10 packages fail compilation. That percentage is steadily and constantly decreasing but it will never drop to zero obviously because at least some "OK on GCC" failure are due to two bugs correcting themselves, weird corner cases that the clang guys will never accept as a bugfix and the upstream will never tolerate, etc.
Now, compiles without error does not mean the binaries actually work. You'd need a full unit testing setup. Compiles but insta-segfaults is not success. Compiles but fails to output correct data is not success. Compiles but anything that uses it fails in some odd way (contemplate a library that compiles, but does not play well with others) So the actual failure rate is obviously somewhat in excess of 10%.
The next issue is thats 10% of Debian packages. Obviously installing clang and messing with the symlinks and compiling Eclipse (written in java) proves very little to nothing about the clang c and c++ compiler, because its not used for java. I would guess that only a large fraction of Debian packages use C or C++. So the actual failure rate for clang when rubbed up against "real world" source code is apparently far in excess of 10%.
Certain general application classes probably have higher or lower failure rates. The failure rate for C operating systems is probably not going to be similar to the failure rate for high performance numerical computing.
So the scales of justice weigh:
GCC: Positives: Free-er society/culture, much faster binaries, and it actually works. Negatives: fully/partially automatic refactoring and source analysis tool writers have a somewhat more difficult time
clang: Positives: People writing source analysis, refactoring type tools prefer working with clang. Negatives: BSD license encourages (at least permits) anti-social behavior, well over a tenth of your code will not compile without problems.
Hmm I think I'll be sticking with GCC for a good long time. Predictions about a decade in the future are pointless, could go either way. For ideological purposes I'd much prefer a GPL / GCC world, much healthier and free-er and more social.
Aside from the more or less irrelevant licensing issue, clang is all about the source analysis tools, refactoring, rewriting support. Uses less memory and time. Both caused by lack of optimization.
Slashdot Mirror
Here's a "eh" solidworks PCB. Give it a bit of cleanup, but it on a background like a model's palm/hand, put it right side up so you can read (how did that slip thru?), work a little on the textures, and its all good.
http://bdm.cc/projects/zigboard/zigboard3d.jpg
Here's a "eh" real photo of a PCB which may or may not link properly. Lets list the faults. Bad lighting leads to shadows, half of each resistor is illuminated and half in shadow. bright camera flash reflection in the center and off some solder joints. The focus plane is obviously near the camera... look at the solder joints holding the two PCBs together, you can see whats going on but they're not in sharp focus.
http://t2.gstatic.com/images?q=tbn:ANd9GcRX2GcAILgBYtnWR-Lc8zHzMjLZthhM-m4knUhtKxZEs82qOQlbuFqHXXCLqw
Now you have to realize the solidworks example is at least an attempt at a marketing mock up, and the genuine pic is merely trying to show a manufacturing fault to someone else, not create a work of art. Still, they're fairly typical examples of what you easily get with CGI vs pics.
CGI is replacing actual photos in the stock picture business, and in the catalog business. I haven't seen an office catalog in years that uses actual pictures. It's all semi-competent CGI.
I know for a fact, err, trustworthy heresay... that almost all electronic catalog "pictures" are 3-d CAD renderings, sometimes with a bit of photoshop. I'm talking about real EE component catalogs, not best buy consumer catalogs.
From trying to take pictures of things I've built, its an unholy PITA and depth of field and reflections and lighting are agonizing. You can look at the pic of a PCB, lets say a stereotypical switching power supply module, and try to figure out how I could get that depth of field and lighting without reflection issues and suddenly realize, this was done in Solidworks not a camera.
If you want to see how bad "real pictures" of electronic devices/components look, try trashy photos of that stuff on ebay. Some of those guys are obviously not even wiping the human grease off the cellphone camera lens first.
Video is a big problem for photographers and not in the way you specify.
A big part of being a photographer is getting just the right moment for the shot at the perfect angle. An instant in time.
Now you just run a high def video camera, walk around and wave the camera, and pick the best individual frame later.
Aside from weddings, this is also causing license chaos because people used to purchase and pay separate photo and video rights at sporting events... why pay for photo rights if you can just use a single video frame, and why miss the action if you can just use a video camera. So enforcement is all confused about that. The fairest way to charge for "rights" at sporting events seems to be by ounce of gear hauled into the venue. I know there have already been court cases over video camera guys selling single frames to newspapers, but I don't know how they've turned out.
Ah but those licenses were (probably not) paid by and for the previous regime... A reasonable argument could be made for the US to buy a country sized site license for all PCs in Iraq ...
Historically, in the arms dealing arena, this was how "foreign assistance" worked. Congress can't hand a billion bucks to General Dynamics just for fun. But you can hand a billion bucks to Israel on the agreement that they hand that same billion bucks to General Dynamics in exchange for a couple fighter airplanes at list price.
That is probably how an Iraq re licensing deal would work. MS pays congress $1M to send $1B to Iraqi govt on the nod and handshake agreement that they spend that $1B on licenses for windows as part of "economic development" or some such B.S.
MS doesn't have to actually provide or "do" anything, they just need to convince people to send them money, which can be pretty cheap.
No, once a store is available, all you need is to have the devs behind adblock plus only distribute thru the store. I have exactly 13 addons. All I need is a couple to decide to switch from free mode to store mode and that makes FF much less useful to me.
The danger is making it necessary to switch addons makes it not look all that much harder to switch to Chrome.
For example, say somebody tries to monetize noscript or firebug... I'm sure there is a "almost as good" replacement that'll remain free on FF... but if I'm going to do the effort to substitute, I'll find a "almost as good" replacement on Chrome.
responded promptly with an interim solution for this problem [avira.com]
Minor technical error remains. You meant to write
responded promptly with an interim solution for this problem
How does the interim solution get implemented if the machine won't boot?
In ye olden days, if you compiled a new linux kernel, deleted the old one, forgot to run lilo, it doesn't reboot, then the solution was to boot and run lilo, which was a task that separated the men from the boys, err it wasn't that difficult, maybe separated the 7 year olds from the 6 year olds or whatever. Anyway...
Also a note to the editors, that link would have been a million times funnier if it pointed to a ubuntu.com live cdrom/dvd/usb image, or maybe the android-x86 project, which is really quite usable.
"Most people" don't need much more than a working web browser. Even at work, "they" were recently f-ing around with a firewall and managed to block ssh, so I installed ajaxterm to work around that, and I mostly use the webmail page anyway, so all I really need at work is a working browser. I would not want to use ajaxterm on a regular basis 10 hours a day, but its survivable in a pinch until the firewall guys get their stuff together. For example I really enjoy how it floods the logs unless you do extra work....
Ballmer may not be special, maybe even lousy, but worst EVAR!!!? I would pick some of the CEOs around the world that lead us into this global recession - who not only did so but (distinguishing them from their counterparts in government) personally took home millions of tens of millions of dollars for doing so and are living lavishly to this day.
As quislings, I think the authors definition of "worst" might not quite match yours.
But even given that many of the banks would have imploded if not for the bail out, GM would be gone if not for the bail out and plenty of marginally successful companies have gone through quite a bit of economic turmoil that MS has avoided
That is the failure mode. The way "leadership" is defined, MS should have dumped a multiple of their net worth into the Zune, then paid millions in campaign contributions to politicians to get billions in bailout funds. Heads we win, tails you lose.
The landscape is quite a bit different for big companies operating under a corporate owned government, than it is for, say, a cupcake store. In the world of cupcake stores, the MS strategy IS more intelligent than the GM strategy, but this article was talking about the big companies that own the govt and order it around, not a scrapbooking supplies store or other small operators like that.
MS could have paid millions to politicians to force the .mil to buy MS licenses for every Iraqi owned PC in Iraq, that would probably have a pretty good profit. Or they could have purchased politicans to declare linux distros as hacking tools and have border control sieze any laptop with linux installed, or sieze any linux install media. So many kleptocratic ways to turn billions into trillions, and instead they ... failed.
In a way, its bad news, if they get rid of Ballmer and but someone competent in his place, then the public will suffer greatly.
Seriously, what does a "web app marketplace" have to offer that isn't already done better through one of the above resources?
Monetize that, so you get to pay for noscript, adblock plus, etc. I'm thinking this is not going to turn out well.
For years I've been waiting for chrome to have addons as good as FF. Maybe being forced to pay will be the big push.
Its interesting watching the journalist / news release re publishers solely focus on competition from download services and totally ignore mail order.
Download took away all the "I want it on release day" and "I want it now".
Mail order took away all the "I want it cheap" and "I've got metered/slow internet access" and "I want something unusual"
That leaves retail stores with... gift sales? Apparently the book industry has given up on retail except for gift sales, and that works for them, but video games are too fast moving of a target.
Printing a jpeg as a PCB layout? Could happen I suppose.
I wonder if DPI is like "megapixels" or "sears air compressor horsepower" where the engineering definition no longer has any relationship to the marketing definition.
For example, marketing could sell four toner colors at 100 dpi as being "400 dpi" after all its four toners each at 100 dpi, right? Something like that would explain why the OP's printer can't successfully output at better than 100 dpi, despite marketing claiming 600 dpi. Hell I could see marketing claiming 50 dpi at 4 colors and duplex printing (two sided) as being "400 dpi".
Latency can be a huge issue. Sure, we got wifi. No we don't trust sensitive data over the air, so you gotta use the wifi, out to the internet, into the VPN concentrator... the corporate VPN concentrator... on the other side of the country... So to VNC into a local server you get millisecond latency connections over wired, or hundred ms (sometimes more) latency over the wifi. Yuck.
I specified that based on visual acuity limits. There's a lot of optical theory explaining why over 300 dpi is mostly useless for toner on paper. Unless your eyeball lens diameter is 10 times bigger than the average human or your retina cell layout is different than all known humans, it is not optically possible to resolve 3000 dpi or whatever on paper under normal conditions and lighting. Depending on how close you can hold the paper before you can't focus on it anymore, and tangentially depending on how bright the light it (little pinhole camera iris) humans top out around 300 dpi.
Now, projected thru transparencies onto a overhead, higher res works, if you have old fashioned overhead projectors and sit close to the screen. Also there are ugly aliasing and anti-aliasing effects that can be avoided by higher res with real vector scaling. And high res allows better/smoother color mixing, in that bluring together 2**8 pixels of 2**16 color is the same as one 2**24 pixel, more or less. There are also relative brightness/consistency effects where making a "line" that varies from 8 to 9 pixels wide looks a lot less consistent than a line that is 85 or 86 pixels wide at 10 times the res, look at the percentage variation of one pixel. If the lighting is really bad, there are strange shadow effects where you can perceive over 300 dpi if the shadows land just right. Also there are some strange toner based textural issues where the plastic surface of thinner lines literally looks different. And some 3-d effects of toner on paper. So over 300 dpi is not a complete waste of time, just mostly a waste with average pictures under average conditions. It would be extremely hard to justify over 1200 dpi even in the weirdest corner cases.
They'd probably resolve it by firing me, since there's industry standard press releases about how everyone should love us because we're green, blah blah.
Complainers get pounded down, not rewarded.
It does depend on the combination of your local corporate culture plus your ability to properly identify it, times how important of an issue it is.
For example, recently making it much harder to use the laser printers, such that people don't print stuff out before meetings, because the printers are far away, not configured on the desktops, and usually out of paper, has saved the earth a lot more than properly recycling huge amounts of paper in the olden days. Exploding the price of soda in the vending machines, crashing sales, has saved the earth a lot more than properly recycling soda cans.
They are offloading multi-threading, multiple robots in a maze doing the same task.
They are not offloading multi-tasking, like I'm currently simultaneously thinking about: /.
1)
2) a really slow data importer I should be fixing, but my brain needs to decompress to unconsciously determine the solution.
3) pr0n, of course.
Moving forward, Solovey wants to investigate other cognitive states that can be reliably detected using fNIRS. Imagine a computer that increases the size of buttons...
Um, WRT to those "buttons", all successful technology is first rolled out for pr0n...
Thats only a square 11000 pixels on a side. A 300 dpi laserprinter would make a roughly one yard/one meter printout.
At a slightly higher resolution that would be a metric A0 paper size. printers that big do exist but are kinda expensive. Best upload it to your local printer/office store and let them print it instead of do it yourself.
Ah this old idea that pops up on /. every month or so.
This is where I relate that you can get ALMOST the same experience temporarily with stretchy gloves and little magnets in the fingertips. Obviously buy gloves with fabric fingers just a little longer than your fingers. In the frozen north, in season, this is not much of a challenge... Florida /.ers are probably all like, gloves, what are those?
It is fun for at least a couple hours. Try touching the body of a big motor or old fashioned xfrmr power supply. Variable speed AC/DC "universal" motors were not as much fun as I expected. Waving in front of a CRT screen is entertaining.
You want gloves thick/strong enough to not tear, but not so thick as to lose sensation.
GOOG for "magnetic wedding ring" and you'll find lots of crackpot "magnets cured mah arthritis" pure BS, but this is probably the ultimate in non-invasive experimentation. Unlike the Magnet-in-glove thing, I've not tried magnetic wedding rings.
The biggest problem with "magnets/hands" is what happens when it inevitably cracks. Sharp little ceramic shards pinching slashed up tissue. So don't go giving steel plates a "high five".
Right now China, India, and Europe are trying to get the USA to play along on this issue. Or at least come to the table.
There is a MUCH bigger problem here. So we pass a law... whats the result?
China: lower levels simply ignore it, higher levels pay a bribe and ignore it.
India: Not quite as corrupt, but pretty much it'll be ignored.
Europe: Move the polluting industry to the EU member state promising the most lax enforcement or a tax break making up for the costs. OR simply move to China. Some net effect but lots of social upheaval.
USA: lower levels will have to follow it and/or go out of business, higher levels pay a bribe (re-election donation) and ignore it.
Coming to the table seems pointless... so our big polluters will ignore whatever's being done, and the small polluters will simply close shop and pollute even more in China...
hostile towards green initiatives solely because "fuck you"
I'm hostile toward them because I'm too smart. Most (all?) of those initiatives are meaningless feel good frippery with no real world effect, or NEGATIVE real world effect. They require stupid people. Not gonna work on me.
Example, the people who just don't give a F about recycling at the office. We are pure, refined evil, right? Where I work, they purchased extra recycling trash cans and distributed them all over and the idiots blathered on about how we're all going to save the planet by sorting our trash, printed posters hanging everywhere. No one, including myself, noticed we still only have one trash dumpster, not a trash dumpster and a recycling dumpster, how... interesting. As a tech guy I work odd hours, and I get to see the illegal alien cleaning crew pushing a big rolling trash collector around the office and dumping both trash and recycling into the same rolling collector, and that collector dumps mixed trash and recyclables into the same dumpster, and odd mornings that I'm here early I see that dumpster emptied into one trash truck. I know they're not sorting at the landfill, either. So the idiots think they're saving the world, but I know the real world effect is we turned a lot of crude oil into plastic recycling cans, waste lots of money buying those cans, waste time and money proselytizing to people, waste valuable time and money sorting trash that is going to be commingled by the cleaning crew anyway, and finally waste time emptying twice as many trash cans. What a huge amount of environmental damage to fool people into thinking they're preventing environmental damage. I'm sure some minor drone got a nice promotion out of it. So, yeah, I'm one of the insane number of "F you" people who throws my empty soda cans into the regular trash can because apparently I don't want to save the planet.
I have special knowledge about the recyclables/trash issue because of weird working hours. I would assume the same scammers running the same psychological confidence scam in other areas are also not doing anything useful for the environment. Maybe accidentally once in a while they occasionally do something useful by mistake, but on average the environment would be better served by those kind of folks if they merely piled up cash in my backyard and set it on fire. So, yeah, in general, F those people and their goals, they're all scammers.
What does work is financial. So I'm painting my house.. I could use oil and have to buy "bad for the environment" mineral oil solvent to clean the brushes, or I could use latex paint and clean the brushes with practically free water... Thats how I save the environment from hydrocarbon vapor / ozone pollution, by saving money. The actual cost of gasoline is about $8/gallon and thats what it should be at the pump, instead of being cheaper and the balance paid by "stealth" taxation. Higher gas prices would fix a lot of environmental problems.
Given assumption (which may be wrong) climate change causes $1B damage resulting in eventual death of 1M
Scenario 1) do nothing. Result $1B in hole results in 1M deaths.
Scenario 2) Social engineering has never failed in the past, so we'll give it a try. Result $10B in hole results in 10M deaths.
I mean, come on. People are asking the .gov to accomplish something. This is the .gov people, not GOOG. All they're going to do is accept money from donors, send contract money to donors, repeat, plus or minus some feel good legislation.
This is only going to be fixed by individual social pressure working on individuals. Not federal social engineering. Not solving social problems using tech.
The funny part is that this is a "tech" scandal... Supposedly anyone older than 25 is "over the hill" WRT tech and is already obsolete, yet to have lived thru the original watergate scandal you'd have to be at least 50. I mean, yeah, technically I was alive when nixon resigned, but I was only a couple months old so I didn't care too much. I figure you'd have to be at least 50 to have been paying attention.
I guess the tech connection is if you're "in tech" then ask your grandfather about putting -gate at the end of every scandal.
It could be worse, we could be going thru history reporting on water-contra, water-resume or water-yahoo or whatever.
It raises the question... given the much better featureset, why are the major Linux distributions still using GCC to compile the kernel and all the OS packages?
LOL. As if those features matter very much to almost anyone. At this time, if you're developing C, it would be a really wise idea to verify code compiles under both and actually ship GCC compiled binaries because they're better (faster). Lack of optimization under clang is a bug, not a feature, unless you're doing human-powered-optimization like refactoring etc etc.
Its unlikely you'll "ever" be able to totally drop gcc... there are features that clang architecturally opposed to. Specifically fortran/C stuff, and the alternative GCC languages in general. Also its not a binary either/or. This is hardly the first time a OS has had to deal with an alternatives system. On debian, the alternatives system is called the alternatives system. I suppose other OS might have a different named system, and it might work somewhat differently, but the concepts will all be the same.
For a good time you can examine
http://clang.debian.net/
About 1 in 10 packages fail compilation. That percentage is steadily and constantly decreasing but it will never drop to zero obviously because at least some "OK on GCC" failure are due to two bugs correcting themselves, weird corner cases that the clang guys will never accept as a bugfix and the upstream will never tolerate, etc.
Now, compiles without error does not mean the binaries actually work. You'd need a full unit testing setup. Compiles but insta-segfaults is not success. Compiles but fails to output correct data is not success. Compiles but anything that uses it fails in some odd way (contemplate a library that compiles, but does not play well with others) So the actual failure rate is obviously somewhat in excess of 10%.
The next issue is thats 10% of Debian packages. Obviously installing clang and messing with the symlinks and compiling Eclipse (written in java) proves very little to nothing about the clang c and c++ compiler, because its not used for java. I would guess that only a large fraction of Debian packages use C or C++. So the actual failure rate for clang when rubbed up against "real world" source code is apparently far in excess of 10%.
Certain general application classes probably have higher or lower failure rates. The failure rate for C operating systems is probably not going to be similar to the failure rate for high performance numerical computing.
So the scales of justice weigh:
GCC: Positives: Free-er society/culture, much faster binaries, and it actually works. Negatives: fully/partially automatic refactoring and source analysis tool writers have a somewhat more difficult time
clang: Positives: People writing source analysis, refactoring type tools prefer working with clang. Negatives: BSD license encourages (at least permits) anti-social behavior, well over a tenth of your code will not compile without problems.
Hmm I think I'll be sticking with GCC for a good long time. Predictions about a decade in the future are pointless, could go either way. For ideological purposes I'd much prefer a GPL / GCC world, much healthier and free-er and more social.
Aside from the more or less irrelevant licensing issue, clang is all about the source analysis tools, refactoring, rewriting support. Uses less memory and time. Both caused by lack of optimization.