I think you guys are missing each other. The body clearly matters more in the digital realm than it did in the film days. However, the fact is that a good sensor is still a lot cheaper than a good lens. I've got a 12MP DSLR from a few years ago and to really get pixel-level resolution I'm going to need a prime lens that costs $400 or so, or a zoom that costs well over $1000. The latter is more than the cost of the body when it was new, and the former is certainly more than it is worth today.
Likewise, if you like a camera on the market today but don't have a lot of cash if you wait a year or two the body will be half the price. However, the lens will cost the same, if not a hair more owing to inflation.
Now, sensors have been improving, and in particular light sensitivity is a LOT better than it used to be. That means you can get a lot further with a zoom at F5 than you could back in the days that it was noise city even at ISO 800. A modern body that takes decent pictures at ISO 6400 gets you 8x as much light compared to the older ones, which means that you need three stops less out of your less. There is a BIG price difference in a sharp zoom at F5 compared to F2, and in action situations the much larger depth of field is an added bonus.
Sure, the sensor matters on camera, just like the CPU frequency matters on a computer. However, both have tended to been over-emphasized such that most cameras (and computers) are fairly unbalanced in favor of selling points and consumers are well-advised to focus on the other attributes if they really want good value.
Yup, and besides, who buys Blackberries anyway? I think they've carried pretty-much every Nexus phone to date, several exclusively I think (especially if you count the ADP1/2 in with that).
Sure, they don't have an iPhone, but people willing to spend the money on one of those probably aren't really their target market in the first place. I love that T-mobile works pretty-much anywhere I go, and it works at 3-4G speeds in virtually all of those places (and I mean I actually GET those speeds consistently). I hear ATT users are lucky to get dial-up speeds.
Are they actually LOSING money, or are they just not raking in a $30/mo premium like some of the other networks? As a customer that's a "feature" I can do without.
I'd say the vast majority fall into this general pattern - at least most organizations that people would associate with the title "church." Many of the "churches" that tend to break away from this pattern tend to avoid using that title anyway, or use it differently.
If you want to go back to the origins of Christianity there was only one "church" - and it didn't meet for an hour on Sunday mornings. In fact, very little in the way of modern church traditions have much resemblance to how Christianity was originally practiced. In its original form it was a very social religion.
Have you even closely examined what happens inside the walls of a church?
M-F it runs like a small business (or in some cases a not-so-small business), complete with meetings and so on - non-employees rarely show up.
Saturday it is a ghost town unless rented out for some purpose like a wedding or funeral, or who knows what.
Sunday morning it is a series of events that involve people sitting in a room lecture style for an hour (maybe standing and singing for part of it - with 30 seconds of token saying hello to the guy next to you). When you're all done you might spend 5-15 minutes milling around in the lobby chatting with people you likely already know before leaving. Chances are you're talking about the same things you'd talk about in a library anyway, except that you can feel good about the fact that the people you're talking to are the right sorts of people.
It is about as much of a community-building activity as the local mall.
Yeah, I know, it's unfair to blame WP for DOS's 640k memory model, but there were enough games that made decent use of XMS and EMS memory that WP should have been able to do somewhat of a better job.
Methinks you are a bit nostalgic - those games were a royal pain to get working with EMS and XMS. I ended up having a config.sys for each one. Some wanted EMS, some wanted XMS, and usually if you configured one or the other the ones that wanted the other wouldn't work. The games that wanted to do their own protected-mode memory management would die out if EMM386 was running, and so on.
And don't get me started on IRQ 5 vs 7 and all that...
Maybe if you had double the system requirements in RAM you could afford to be less picky with the configuration or something, but I usually had to run through my library of config.sys files to find the one that worked, until DOS added all those boot menu support which let me write the equivalent of sysvinit in a batch file.
I hear you on the response times. Probably one of the things I love most about MythTV is that when I hit the button on the remote something happens. There is a slight lag when I actually hit play while it buffers, but less than the cursor movement delay on most DVRs. I can hit the browse-all-shows function and the list loads in about two seconds, with instant cursor movement thereafter. All this on an Atom-based system that pulls all of 10W (granted, just the front end). If they spent an extra 75 cents on the CPU/RAM I imagine that most DVRs could compete.
Oh, and my MythTV system doesn't mess up the schedule if I have more than 15 shows scheduled to record unlike cable company DVR I once had (20 programs sounds like a lot until you have a family of four with different viewing habits).
The point is that it is years later and they're still fighting it out. Chances are that in a decade somebody will be told "don't do that anymore" and most likely they'll just keep on doing something else that is just as bad, since there are no consequences.
Where did I see it stated recently that if the sole penalty for bank robbery was having to return the money you stole every vault in the country would be empty in a week?
After a while I got to be able to recognize certain types of activity, such as a web user browsing to/index.html, based on the sounds the printer made.
Yeah, but try that out with the access log on the Google homepage... The paper would probably catch fire from air friction flying out of the printer...
Until stores figure out how to use a bank teller queue I'll use those self-checkouts every time. Typically those have no line, or they use a bank-teller queue. That means that when I walk up to one I know exactly what kind of experience I'm going to have. Sure, maybe it takes a little longer if something goes wrong, but my time at the register is fairly predictable.
If I queue up in the normal human line inevitably I end up behind the person who is convinced that the sign had a price 3 cents lower than whatever the item rang up. Then they're going to have 387 coupons and argue for 15 minutes about whether one of them is valid for 15 cents off their order. Then they'll pay with food stamps (and fight over whether some item is eligible), and then pay the balance with a check from the bank of Nigeria denominated in Rubles, but they like to keep their register even so they'll throw in 72 cents in change (50% in pennies) as well.
Now, one thing I'll never do is queue up at an individual self-checkout behind virtually anybody else unless I see clear evidence of a pocket protector or something. I want to rip my hair out watching people try to figure out how to use the things. Just do what the screen tells you to do whether you think it makes sense or not - no doubt the design is to prevent some form of theft that you're clearly not clever enough to pull off in the first place.
As far as produce goes - I know my local store has pre-weigh stations all over the place where you can print out a barcode to speed checkout. All you need to do is put RFIDs on the bags and pre-weigh their contents, and then you can just walk out with a cart full. They could weigh you on the way in and out and make sure the net weight is close enough, or maybe spot check 1:50 carts which is a lot faster than checking them all out.
Don't forget the 30 pages of requirements revised 12 times (written in the abstract without any regard for the API, so that in the end you build 3000 lines of code where you could have had 2 API calls and a stock widget or whatever, because the stock widget isn't available in green or whatever the requirement says). Don't forget the RFP response before you even get the contract. Don't forget the unit tests, system tests, acceptance tests, change management, and all that. Inevitably the customer will want three things tweaked when it is almost done, so that means rewriting half that documentation, getting it all signed off on paper again, and doing a full set of testing just in case changing the text from green to yellow (which would have worked with the stock widget) broke something, since otherwise you have to write up 300 pages of FMEAs to justify not running all the tests.
Oh, and it is a felony to mischarge your time so make sure those developers spend about 15% of their time micro-categorizing how they spent the other 85%.
And the boss needs to be paid - probably to go schmooze in Washington with the guy who gave him the contract so that he can get the next contract.
This is why governments and most 3rd-generation corporations (ie two or more CEOs since the founder) can't write mobile apps. Too many people with decision rights that just can't be bothered to actually understand the technology.
They don't care if you write the app, or if it works. They care that you can show them documentation proving that you wrote the app, and proving that it works. Whether it actually works or not is just a by-product of the process.
Quality is measured in pounds of paperwork. Oh, and chances are you will in fact be printing it on paper, since how else do you intend to prove that you reviewed it all without signatures? Oh, electronic signatures? Great, can you show me the documentation proving that you have a working electronic signature system?
It would cost $100k to build hello world to government standards...
Adding tax? So iPod will cost - 500 USD as usual paid to Chinese, and then 500 USD you-are-bad tax paid... by the American customer to American government? Well yes, THAT will show them evil commies! They will not loose any cent on it, only USA citizens will be able to afford less goods.
Who is going to spend $1000 on an iPod when they can buy some competing phone built in Kansas for $600? And what idiot would send the work to China when they could send it somewhere else and make the same profit on a price tag $500 cheaper?
The Chinese would lose lots of money on such a tax, in the form of lost sales.
Plus, you don't think it really costs $500 to build an iPod, do you? Products are sold based on what people are willing to pay, not what they cost to make. If it costs more to make than people would be willing to pay then you don't bother making it in the first place. If Apple lowered their cost to produce an iPod by $50 you wouldn't see any change in the price. Likewise, if competition made people less willing to buy them then the price would drop without any regard to the cost of production.
d(Cost*Volume)/d(Price)=0 -> solve for Price.
So, if the ideal selling price of an iPod is $500, then that is what the final price tag will be, tarrifs or not. That just means that Apple has to only collect $250 itself so that it can hand the other $250 to Uncle Sam. Of course, they would instead either clean up their plants (so fast you wouldn't believe it), or send the work elsewhere.
Companies always whine about how they can't do this or that until you tell them they can't sell their product until it is fixed. Then suddenly all those problems seem to disappear. Those CEOs are just doing their jobs - making money at the expense of anyone or anything else, and it is time government started doing theirs - keeping them from doing it..
Well, on my PC rebooting means losing two minutes in the middle of a TV show being recorded by my DVR and splitting it in two, while restarting X doesn't really cause me much hassle at all.
I think you're looking at it the wrong way. We shouldn't be asking "is it the job of the government to ensure that Avatar is created?" We should be asking "do we want our society to be one that creates things like Avatar?" or "do I want to have the option of seeing another movie like Avatar?" (Ok, let's neglect that the visuals were great but the plot was pretty lame.)
Copyright is a way of allowing a movie like Avatar to exist, and people who don't watch it don't have to pay much if anything to let it exist (court costs are about it, and until recently they weren't much), but the people who do want to watch it can pay for it. 20 years ago the system pretty-much worked fine - like it, pay for it, don't like it, don't pay for it.
Now the system is breaking down, but I'm not sure the solution is to go back to patronage. Do we really want to completely wipe out large-scale expensive productions of any kind that rely on copyright currently to survive, or move them completely to a model of government sponsorship? Do we really think that some corporation is going to donate a few hundred million dollars so that somebody can make a fancy movie, with no real return on that investment? And, will directors/producers really have an incentive to make a movie successful if they are paid in advance and have no real skin of their own in the game?
Sure, the world can go on without movies, without large commercial software applications, without drugs, and so on. I think that books and music would probably survive at some level without copyright, because these are works that can be done by a few people at relatively little tangible cost. The problem comes with works that require armies of people to put together. Sure, there will still be linux and apache, but even those products are largely the spare-time creations of people who make their living working for companies that rely on copyright to make money.
Thinking about alternatives one way to reform the whole system is to dramatically shorten the duration of IP protection. I'd probably establish it per-industry so that there is a good opportunity to profit on the initial investment. For low-investment stuff like books/music I'd make the duration pretty short - maybe a year. For rapidly moving industries like software I'd do similar things with patents - go ahead and let people get a six month or one year lead out of multitouch or whatever, but then that's it. There would no longer be endless litigation since everything would quickly become moot. For things that have very long development cycles like pharmaceuticals I'd probably shorten the duration only a little, and maybe even subdivide the category based on annual revenue or something (many drugs can recoup their initial investment in only a few years, but there are also many drugs that take 10+ years to do so if they treat more rare conditions). I think that pharmaceuticals is also a sticky issue since it runs into issues with social responsibility and all that - maybe try to move that industry to one that is patent free but well-funded by bounties or whatever so that the costs aren't borne by patients.
Lots of people want to "stick it to the man" - and I'm all for reforming corporate governance and all that. However, I think that reform is the key here and not simply burning down the structures of society around us. Companies make way to much money right now, but I don't think the solution to that is to move to a world where only tradesman have a chance of making an income, and only then if they can outperform a robot or somebody in a country that pays $5/day.
I do tend to agree with this. Right now if my display is borked but I can get in via ssh I'm still pretty-much SOL. With a microkernel design I could probably reboot the entire display stack (just tell those drivers that the computer is booting up - unload them from memory and load them back in and all that). It would wipe out all my video buffers and maybe kill my X11 session, but it isn't a reboot. If you go a step further you could have a watchdog of some kind or some other way of triggering this and maybe do that proactively.
It would be nice if a video driver hang was like an apache hang.
Now, since you're talking about hardware there is still some risk - if your video driver tells the video card to hit the reset line or something and that resets the computer, well, there isn't much you can do about that but re-architect the bus so that the hardware itself is more isolated.
I'm sure there is an assembly company somewhere in the world that doesn't abuse its workers. For starters try looking in the US or Europe.
You don't start with the assumption that you have to outsource and then ask whether you can do that and still manage to follow the law and be ethical. You start with the assumption that you have to follow the law and behave ethically and then ask whether you can do that and still manage to outsource.
When a pharmaceutical company tests a drug on a room full of rats it probably costs a few thousand dollars. When they do a phase 3 trial on thousands of people the costs are in the tens of millions of dollars. Trust me, the people employing these scientists have EVERY incentive to make the animal tests as accurate as they can, and figure out how to avoid doing the tests on people only to have the drugs fail in the end. Plus, beyond wasting tens of millions of dollars a company spends years going down the wrong path and then needing to backtrack.
People have been talking about making animal tests more accurate for decades now. No doubt progress has been made.
The guy might as well go on about all those lousy scientists who can't be bothered to find a cure for cancer despite it being vitally important to society that they do so.
1. A battery of algorithmic tests are done to predict any likely toxicity (likely before they even bothered making the compound in the first place). 2. The compound is subjected to 40 bazillion in-vitro experiments to try to predict what it will do to a person. That includes all kinds of exotic tests that even simulate some aspects of structures present in organs like the liver. 3. The compound is tested on cheap and less-regulated animals like mice. 4. Often the compound is tested on a few more expensive and heavily regulated animals like dogs and chimps, often with the animal model being based on how the drug is expected to interact with a person or to explore data gathered in earlier tests.
Only then will the drug actually go into people, first to assess basic safety, and then dosing and efficacy and general trials.
Laws force it to work this way to ensure patient safety. Companies have incentive to do this anyway, since it is much harder to experiment in people (you don't get that many subjects) so you need to know what you're expecting to find before you do the experiment. Plus, testing in people gets way more expensive - so better to find out that it kills dogs early-on.
As far as different levels of animal regulations go - basically the cuter the animal is, the harder it will be to experiment on it. You could probably test all the sea kittens you like and nobody would blink an eye...
So, how is this different from growing a corn plant for the express purpose of putting food in my stomach?
That's the whole problem with biology - you can come up with any arbitrary line in the sand about what is and isn't OK, and chances are somebody can find one of a bazillion life forms that makes you eat your words, and if they can't they can probably create one.
What makes a paramecium or radish different from a mouse? And what is stopping us from making a mouse that specifically lacks that quality that makes it different?
If you're willing to live in a bubble and eat nothing but nutrients chemically synthesized from sterile-filtered CO2, N2, and H2O then most will be able to concede that you at least have ethical consistency.
Well, the MBAs apply Six Sigma to all kinds of stuff that it really doesn't fit. However, the definition of six sigma is pretty straightforward:
A six sigma process is one whose specification limits are at least six standard deviations away from the mean.
So, if a space shuttle part needs to be 1 meter long, and if bad things happen if it is more than one cm off, then a six sigma process would need to produce parts that are 1 meter long with a standard deviation of 1/6th of a centimeter (and a normal distribution of sizes). If the process can do that then the chances of a part ever coming off the line that is off by a centimeter is VERY low - so low that you don't really need to check them all.
Statistical process control is how the Japanese clobbered US industry after WWII. The US was stuck on outdated models where you test every part and reject the bad ones. The pioneer of SPC (an American) realized that you could ditch the testing, and take all that saved money and instead put it into improving your manufacturing process so that you don't make the bad parts to begin with. Modern process control is about randomly monitoring critical parameters and ensuring they all stay in range so that the final product is VERY likely to have the desired attributes.
Alpha is used in hypothesis testing - as in, we can all be sure that 5% of all the clinical trial conclusions ever reached are downright wrong, and most likely those that are actually are reported are wrong much more often than that.
What's to stop me from registering a.com through zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz.com? At least a token fee creates some kind of barrier to entry to make people think twice before they gain permanent control of some part of the namespace. The $7/yr fee isn't that bad and at least serves that purpose.
Also - nobody has to pay anything for the new TLD. There is no reason that ford.com and ford.xxx have to be owned by the same company. In fact, I'd like to see pre-registration banned entirely. What is the point in expanding the namespace if it just turns into a clone of the.com namespace? Why can't Ford Prefect run his escort service on the new TLD?
Polio vaccine was developed back when products like aspirin were able to make it onto the market. The costs were very different back then as much less was spent post-discovery. However, this is exactly the kind of scenario I'd envision.
I'm all for trying to reform the way companies work, but unfortunately the current system makes the kind of behavior you decry inevitable for any company that is held in majority by the public. The shareholders vote for the board, the board appoints the CEO, and the CEO keeps his job by keeping the shareholders happy. The only companies that really can avoid this trap are those that are still controlled by their founders. Family-owned business (after the founder's death) are hit-and-miss - sometimes they are generous, and sometimes they're just out for money.
Most of the stock in a typical pharmaceutical company is held by institutions from pensions and mutual funds and such. When you pick the mutual fund to put your 401k money into, chances are that the only thing you look at is the rate of return. That means the fund manager is under intense pressure to squeeze every cent out of your money that they can, and that just gets passed along to the CEOs who ultimately control the companies the money gets invested in. The only way that can change is if you fundamentally change how corporations are controlled, or get rid of them.
Sure they can - just inspect them regularly and switch vendors if they get out of line. Before you even sign a long-term contract with a vendor you observe them operating in an ethical manner. It doesn't cost that much to just keep one guy down there to keep an eye on things full-time.
If they have trouble keeping themselves honest the US can just tell companies that source work overseas that they have a burden of proof to demonstrate that local laws are being followed, and if not they will be fined.
If you hold companies blameless for what their business partners do then all you're going to do is sabotage your own job market in favor of markets in countries that have more lax laws. That is just economic suicide. Companies are mostly in it to make money, so if they can dump sludge in the creek in Mozambique then that is where they'll send their production.
The patented drugs would be much more expensive than the publicly funded ones. I think you just mis-read my sentence since we're saying the same thing.
Now, the R&D component of drug costs isn't actually less expensive with socialized medicine than without (at least, I don't think it would be - it never really has been tried). The difference is just who pays those costs. With a patented drug the costs of R&D are amortized over all the pills made (along with a ton of other costs that would go away with public funding). With a publicly developed drug the costs of R&D are paid by the taxpayers, and then most likely would not be recovered through drug prices.
Privately funded drugs pretty-much only have one model since the price of the product is the only way a company can recover its costs. Publicly funded drugs actually could embrace varying levels of socialism. The NIH could develop a drug and then STILL charge $5 per pill or whatever in which case taxpayers would pay relatively little of the cost (as taxpayers - obviously they pay if they take the drug). Or, Congress could appropriate tax funds for the drug R&D and the pills might cost 10 cents each.
The obvious advantage of funding drugs with taxes is that you can be much more flexible with how those costs impact society. Instead of charging people lots of money to quit smoking you could make the cessation drugs free and charge an extra tax on cigarettes to pay for them. And so on... Drug costs are by their nature a relatively regressive way of collecting money.
Another advantage of cheap drugs is compliance. When drugs are properly used (and not over-advertised/prescribed) the fact is that they save money compared to other therapies. If you can push off surgical interventions on arteries to once every 20 years instead of once every 5 years you save a fortune in medical bills (at a cost that is comparatively cheap even with the current astronomical drug prices). However, when drugs are expensive a poor person is more likely to not take them, and then go to the ER and have that surgery done anyway.
Frontline covered some other aspects of health care, and it really does create good debate points about the whole cradle-to-grave thing. They tracked down a case of some guy who had been in and out of ERs numerous times at a cost of tens of thousands of dollars (I think medicare was paying that one). It turns out he has asthma attacks, and they inspected his home and found that the ceiling was crumbling and the poor condition of the home almost certainly was a big contributing factor. However, the guy couldn't afford to have it properly fixed up. So, we have a case where society is willing to pay $50k in six months or whatever on medical bills, but there is no way that society would pay a contractor $10k to repair and clean up the house. On the other hand, if you really did open that door then the whole USA would become one big homeowner's association.
Like I said, healthcare is complicated when you get beyond the sound bites and political posturing. My feeling is that if you're going to let the sick people die in the streets then do that and quit wasting money on them, but if you're going to pretend to care about them and keep them from dying than you might as well get your hands dirty and improve their lives so that it is more cost-effective, even if your only interest is your own wallet.
Education has the same sorts of problems - we stick kids in classrooms but we can't force them to learn. Now, if you're going to let people who can't hold down a job die in the streets from starvation then do it and quit wasting money on ineffective welfare, but if you are going to pretend to care then you need to get your hands dirty and force the kids to get an education such that most of them actually can be productive (there will always be those who will need more care). And, just as with healthcare our institutions are highly dysfunctional so there are a myriad of problems that need to be solved before that can really happen.
Just another example of the kinds of logic that brought us stuff like:
"This product contains a chemical known to the State of California to cause cancer."
"Water - 99.99999%. In accordance with the New Jersey right to know act, contents partially unknown." Yup, there just might be a carbon atom floating around somewhere in there...
Slashdot Mirror
I think you guys are missing each other. The body clearly matters more in the digital realm than it did in the film days. However, the fact is that a good sensor is still a lot cheaper than a good lens. I've got a 12MP DSLR from a few years ago and to really get pixel-level resolution I'm going to need a prime lens that costs $400 or so, or a zoom that costs well over $1000. The latter is more than the cost of the body when it was new, and the former is certainly more than it is worth today.
Likewise, if you like a camera on the market today but don't have a lot of cash if you wait a year or two the body will be half the price. However, the lens will cost the same, if not a hair more owing to inflation.
Now, sensors have been improving, and in particular light sensitivity is a LOT better than it used to be. That means you can get a lot further with a zoom at F5 than you could back in the days that it was noise city even at ISO 800. A modern body that takes decent pictures at ISO 6400 gets you 8x as much light compared to the older ones, which means that you need three stops less out of your less. There is a BIG price difference in a sharp zoom at F5 compared to F2, and in action situations the much larger depth of field is an added bonus.
Sure, the sensor matters on camera, just like the CPU frequency matters on a computer. However, both have tended to been over-emphasized such that most cameras (and computers) are fairly unbalanced in favor of selling points and consumers are well-advised to focus on the other attributes if they really want good value.
Yup, and besides, who buys Blackberries anyway? I think they've carried pretty-much every Nexus phone to date, several exclusively I think (especially if you count the ADP1/2 in with that).
Sure, they don't have an iPhone, but people willing to spend the money on one of those probably aren't really their target market in the first place. I love that T-mobile works pretty-much anywhere I go, and it works at 3-4G speeds in virtually all of those places (and I mean I actually GET those speeds consistently). I hear ATT users are lucky to get dial-up speeds.
Are they actually LOSING money, or are they just not raking in a $30/mo premium like some of the other networks? As a customer that's a "feature" I can do without.
I'd say the vast majority fall into this general pattern - at least most organizations that people would associate with the title "church." Many of the "churches" that tend to break away from this pattern tend to avoid using that title anyway, or use it differently.
If you want to go back to the origins of Christianity there was only one "church" - and it didn't meet for an hour on Sunday mornings. In fact, very little in the way of modern church traditions have much resemblance to how Christianity was originally practiced. In its original form it was a very social religion.
Have you even closely examined what happens inside the walls of a church?
M-F it runs like a small business (or in some cases a not-so-small business), complete with meetings and so on - non-employees rarely show up.
Saturday it is a ghost town unless rented out for some purpose like a wedding or funeral, or who knows what.
Sunday morning it is a series of events that involve people sitting in a room lecture style for an hour (maybe standing and singing for part of it - with 30 seconds of token saying hello to the guy next to you). When you're all done you might spend 5-15 minutes milling around in the lobby chatting with people you likely already know before leaving. Chances are you're talking about the same things you'd talk about in a library anyway, except that you can feel good about the fact that the people you're talking to are the right sorts of people.
It is about as much of a community-building activity as the local mall.
Yeah, I know, it's unfair to blame WP for DOS's 640k memory model, but there were enough games that made decent use of XMS and EMS memory that WP should have been able to do somewhat of a better job.
Methinks you are a bit nostalgic - those games were a royal pain to get working with EMS and XMS. I ended up having a config.sys for each one. Some wanted EMS, some wanted XMS, and usually if you configured one or the other the ones that wanted the other wouldn't work. The games that wanted to do their own protected-mode memory management would die out if EMM386 was running, and so on.
And don't get me started on IRQ 5 vs 7 and all that...
Maybe if you had double the system requirements in RAM you could afford to be less picky with the configuration or something, but I usually had to run through my library of config.sys files to find the one that worked, until DOS added all those boot menu support which let me write the equivalent of sysvinit in a batch file.
I hear you on the response times. Probably one of the things I love most about MythTV is that when I hit the button on the remote something happens. There is a slight lag when I actually hit play while it buffers, but less than the cursor movement delay on most DVRs. I can hit the browse-all-shows function and the list loads in about two seconds, with instant cursor movement thereafter. All this on an Atom-based system that pulls all of 10W (granted, just the front end). If they spent an extra 75 cents on the CPU/RAM I imagine that most DVRs could compete.
Oh, and my MythTV system doesn't mess up the schedule if I have more than 15 shows scheduled to record unlike cable company DVR I once had (20 programs sounds like a lot until you have a family of four with different viewing habits).
The point is that it is years later and they're still fighting it out. Chances are that in a decade somebody will be told "don't do that anymore" and most likely they'll just keep on doing something else that is just as bad, since there are no consequences.
Where did I see it stated recently that if the sole penalty for bank robbery was having to return the money you stole every vault in the country would be empty in a week?
After a while I got to be able to recognize certain types of activity, such as a web user browsing to /index.html, based on the sounds the printer made.
Yeah, but try that out with the access log on the Google homepage... The paper would probably catch fire from air friction flying out of the printer...
Until stores figure out how to use a bank teller queue I'll use those self-checkouts every time. Typically those have no line, or they use a bank-teller queue. That means that when I walk up to one I know exactly what kind of experience I'm going to have. Sure, maybe it takes a little longer if something goes wrong, but my time at the register is fairly predictable.
If I queue up in the normal human line inevitably I end up behind the person who is convinced that the sign had a price 3 cents lower than whatever the item rang up. Then they're going to have 387 coupons and argue for 15 minutes about whether one of them is valid for 15 cents off their order. Then they'll pay with food stamps (and fight over whether some item is eligible), and then pay the balance with a check from the bank of Nigeria denominated in Rubles, but they like to keep their register even so they'll throw in 72 cents in change (50% in pennies) as well.
Now, one thing I'll never do is queue up at an individual self-checkout behind virtually anybody else unless I see clear evidence of a pocket protector or something. I want to rip my hair out watching people try to figure out how to use the things. Just do what the screen tells you to do whether you think it makes sense or not - no doubt the design is to prevent some form of theft that you're clearly not clever enough to pull off in the first place.
As far as produce goes - I know my local store has pre-weigh stations all over the place where you can print out a barcode to speed checkout. All you need to do is put RFIDs on the bags and pre-weigh their contents, and then you can just walk out with a cart full. They could weigh you on the way in and out and make sure the net weight is close enough, or maybe spot check 1:50 carts which is a lot faster than checking them all out.
Don't forget the 30 pages of requirements revised 12 times (written in the abstract without any regard for the API, so that in the end you build 3000 lines of code where you could have had 2 API calls and a stock widget or whatever, because the stock widget isn't available in green or whatever the requirement says). Don't forget the RFP response before you even get the contract. Don't forget the unit tests, system tests, acceptance tests, change management, and all that. Inevitably the customer will want three things tweaked when it is almost done, so that means rewriting half that documentation, getting it all signed off on paper again, and doing a full set of testing just in case changing the text from green to yellow (which would have worked with the stock widget) broke something, since otherwise you have to write up 300 pages of FMEAs to justify not running all the tests.
Oh, and it is a felony to mischarge your time so make sure those developers spend about 15% of their time micro-categorizing how they spent the other 85%.
And the boss needs to be paid - probably to go schmooze in Washington with the guy who gave him the contract so that he can get the next contract.
This is why governments and most 3rd-generation corporations (ie two or more CEOs since the founder) can't write mobile apps. Too many people with decision rights that just can't be bothered to actually understand the technology.
This is the government we're talking about.
They don't care if you write the app, or if it works. They care that you can show them documentation proving that you wrote the app, and proving that it works. Whether it actually works or not is just a by-product of the process.
Quality is measured in pounds of paperwork. Oh, and chances are you will in fact be printing it on paper, since how else do you intend to prove that you reviewed it all without signatures? Oh, electronic signatures? Great, can you show me the documentation proving that you have a working electronic signature system?
It would cost $100k to build hello world to government standards...
Adding tax?
So iPod will cost - 500 USD as usual paid to Chinese, and then 500 USD you-are-bad tax paid... by the American customer to American government? Well yes, THAT will show them evil commies! They will not loose any cent on it, only USA citizens will be able to afford less goods.
Who is going to spend $1000 on an iPod when they can buy some competing phone built in Kansas for $600? And what idiot would send the work to China when they could send it somewhere else and make the same profit on a price tag $500 cheaper?
The Chinese would lose lots of money on such a tax, in the form of lost sales.
Plus, you don't think it really costs $500 to build an iPod, do you? Products are sold based on what people are willing to pay, not what they cost to make. If it costs more to make than people would be willing to pay then you don't bother making it in the first place. If Apple lowered their cost to produce an iPod by $50 you wouldn't see any change in the price. Likewise, if competition made people less willing to buy them then the price would drop without any regard to the cost of production.
d(Cost*Volume)/d(Price)=0 -> solve for Price.
So, if the ideal selling price of an iPod is $500, then that is what the final price tag will be, tarrifs or not. That just means that Apple has to only collect $250 itself so that it can hand the other $250 to Uncle Sam. Of course, they would instead either clean up their plants (so fast you wouldn't believe it), or send the work elsewhere.
Companies always whine about how they can't do this or that until you tell them they can't sell their product until it is fixed. Then suddenly all those problems seem to disappear. Those CEOs are just doing their jobs - making money at the expense of anyone or anything else, and it is time government started doing theirs - keeping them from doing it..
Well, on my PC rebooting means losing two minutes in the middle of a TV show being recorded by my DVR and splitting it in two, while restarting X doesn't really cause me much hassle at all.
On a desktop I agree the impact is lesser.
I think you're looking at it the wrong way. We shouldn't be asking "is it the job of the government to ensure that Avatar is created?" We should be asking "do we want our society to be one that creates things like Avatar?" or "do I want to have the option of seeing another movie like Avatar?" (Ok, let's neglect that the visuals were great but the plot was pretty lame.)
Copyright is a way of allowing a movie like Avatar to exist, and people who don't watch it don't have to pay much if anything to let it exist (court costs are about it, and until recently they weren't much), but the people who do want to watch it can pay for it. 20 years ago the system pretty-much worked fine - like it, pay for it, don't like it, don't pay for it.
Now the system is breaking down, but I'm not sure the solution is to go back to patronage. Do we really want to completely wipe out large-scale expensive productions of any kind that rely on copyright currently to survive, or move them completely to a model of government sponsorship? Do we really think that some corporation is going to donate a few hundred million dollars so that somebody can make a fancy movie, with no real return on that investment? And, will directors/producers really have an incentive to make a movie successful if they are paid in advance and have no real skin of their own in the game?
Sure, the world can go on without movies, without large commercial software applications, without drugs, and so on. I think that books and music would probably survive at some level without copyright, because these are works that can be done by a few people at relatively little tangible cost. The problem comes with works that require armies of people to put together. Sure, there will still be linux and apache, but even those products are largely the spare-time creations of people who make their living working for companies that rely on copyright to make money.
Thinking about alternatives one way to reform the whole system is to dramatically shorten the duration of IP protection. I'd probably establish it per-industry so that there is a good opportunity to profit on the initial investment. For low-investment stuff like books/music I'd make the duration pretty short - maybe a year. For rapidly moving industries like software I'd do similar things with patents - go ahead and let people get a six month or one year lead out of multitouch or whatever, but then that's it. There would no longer be endless litigation since everything would quickly become moot. For things that have very long development cycles like pharmaceuticals I'd probably shorten the duration only a little, and maybe even subdivide the category based on annual revenue or something (many drugs can recoup their initial investment in only a few years, but there are also many drugs that take 10+ years to do so if they treat more rare conditions). I think that pharmaceuticals is also a sticky issue since it runs into issues with social responsibility and all that - maybe try to move that industry to one that is patent free but well-funded by bounties or whatever so that the costs aren't borne by patients.
Lots of people want to "stick it to the man" - and I'm all for reforming corporate governance and all that. However, I think that reform is the key here and not simply burning down the structures of society around us. Companies make way to much money right now, but I don't think the solution to that is to move to a world where only tradesman have a chance of making an income, and only then if they can outperform a robot or somebody in a country that pays $5/day.
I do tend to agree with this. Right now if my display is borked but I can get in via ssh I'm still pretty-much SOL. With a microkernel design I could probably reboot the entire display stack (just tell those drivers that the computer is booting up - unload them from memory and load them back in and all that). It would wipe out all my video buffers and maybe kill my X11 session, but it isn't a reboot. If you go a step further you could have a watchdog of some kind or some other way of triggering this and maybe do that proactively.
It would be nice if a video driver hang was like an apache hang.
Now, since you're talking about hardware there is still some risk - if your video driver tells the video card to hit the reset line or something and that resets the computer, well, there isn't much you can do about that but re-architect the bus so that the hardware itself is more isolated.
I'm sure there is an assembly company somewhere in the world that doesn't abuse its workers. For starters try looking in the US or Europe.
You don't start with the assumption that you have to outsource and then ask whether you can do that and still manage to follow the law and be ethical. You start with the assumption that you have to follow the law and behave ethically and then ask whether you can do that and still manage to outsource.
When a pharmaceutical company tests a drug on a room full of rats it probably costs a few thousand dollars. When they do a phase 3 trial on thousands of people the costs are in the tens of millions of dollars. Trust me, the people employing these scientists have EVERY incentive to make the animal tests as accurate as they can, and figure out how to avoid doing the tests on people only to have the drugs fail in the end. Plus, beyond wasting tens of millions of dollars a company spends years going down the wrong path and then needing to backtrack.
People have been talking about making animal tests more accurate for decades now. No doubt progress has been made.
The guy might as well go on about all those lousy scientists who can't be bothered to find a cure for cancer despite it being vitally important to society that they do so.
Yup, and long before any of that happens:
1. A battery of algorithmic tests are done to predict any likely toxicity (likely before they even bothered making the compound in the first place).
2. The compound is subjected to 40 bazillion in-vitro experiments to try to predict what it will do to a person. That includes all kinds of exotic tests that even simulate some aspects of structures present in organs like the liver.
3. The compound is tested on cheap and less-regulated animals like mice.
4. Often the compound is tested on a few more expensive and heavily regulated animals like dogs and chimps, often with the animal model being based on how the drug is expected to interact with a person or to explore data gathered in earlier tests.
Only then will the drug actually go into people, first to assess basic safety, and then dosing and efficacy and general trials.
Laws force it to work this way to ensure patient safety. Companies have incentive to do this anyway, since it is much harder to experiment in people (you don't get that many subjects) so you need to know what you're expecting to find before you do the experiment. Plus, testing in people gets way more expensive - so better to find out that it kills dogs early-on.
As far as different levels of animal regulations go - basically the cuter the animal is, the harder it will be to experiment on it. You could probably test all the sea kittens you like and nobody would blink an eye...
So, how is this different from growing a corn plant for the express purpose of putting food in my stomach?
That's the whole problem with biology - you can come up with any arbitrary line in the sand about what is and isn't OK, and chances are somebody can find one of a bazillion life forms that makes you eat your words, and if they can't they can probably create one.
What makes a paramecium or radish different from a mouse? And what is stopping us from making a mouse that specifically lacks that quality that makes it different?
If you're willing to live in a bubble and eat nothing but nutrients chemically synthesized from sterile-filtered CO2, N2, and H2O then most will be able to concede that you at least have ethical consistency.
Well, the MBAs apply Six Sigma to all kinds of stuff that it really doesn't fit. However, the definition of six sigma is pretty straightforward:
A six sigma process is one whose specification limits are at least six standard deviations away from the mean.
So, if a space shuttle part needs to be 1 meter long, and if bad things happen if it is more than one cm off, then a six sigma process would need to produce parts that are 1 meter long with a standard deviation of 1/6th of a centimeter (and a normal distribution of sizes). If the process can do that then the chances of a part ever coming off the line that is off by a centimeter is VERY low - so low that you don't really need to check them all.
Statistical process control is how the Japanese clobbered US industry after WWII. The US was stuck on outdated models where you test every part and reject the bad ones. The pioneer of SPC (an American) realized that you could ditch the testing, and take all that saved money and instead put it into improving your manufacturing process so that you don't make the bad parts to begin with. Modern process control is about randomly monitoring critical parameters and ensuring they all stay in range so that the final product is VERY likely to have the desired attributes.
Alpha is used in hypothesis testing - as in, we can all be sure that 5% of all the clinical trial conclusions ever reached are downright wrong, and most likely those that are actually are reported are wrong much more often than that.
Domains couldn't be free today.
What's to stop me from registering a.com through zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz.com? At least a token fee creates some kind of barrier to entry to make people think twice before they gain permanent control of some part of the namespace. The $7/yr fee isn't that bad and at least serves that purpose.
Also - nobody has to pay anything for the new TLD. There is no reason that ford.com and ford.xxx have to be owned by the same company. In fact, I'd like to see pre-registration banned entirely. What is the point in expanding the namespace if it just turns into a clone of the .com namespace? Why can't Ford Prefect run his escort service on the new TLD?
Polio vaccine was developed back when products like aspirin were able to make it onto the market. The costs were very different back then as much less was spent post-discovery. However, this is exactly the kind of scenario I'd envision.
I'm all for trying to reform the way companies work, but unfortunately the current system makes the kind of behavior you decry inevitable for any company that is held in majority by the public. The shareholders vote for the board, the board appoints the CEO, and the CEO keeps his job by keeping the shareholders happy. The only companies that really can avoid this trap are those that are still controlled by their founders. Family-owned business (after the founder's death) are hit-and-miss - sometimes they are generous, and sometimes they're just out for money.
Most of the stock in a typical pharmaceutical company is held by institutions from pensions and mutual funds and such. When you pick the mutual fund to put your 401k money into, chances are that the only thing you look at is the rate of return. That means the fund manager is under intense pressure to squeeze every cent out of your money that they can, and that just gets passed along to the CEOs who ultimately control the companies the money gets invested in. The only way that can change is if you fundamentally change how corporations are controlled, or get rid of them.
Sure they can - just inspect them regularly and switch vendors if they get out of line. Before you even sign a long-term contract with a vendor you observe them operating in an ethical manner. It doesn't cost that much to just keep one guy down there to keep an eye on things full-time.
If they have trouble keeping themselves honest the US can just tell companies that source work overseas that they have a burden of proof to demonstrate that local laws are being followed, and if not they will be fined.
If you hold companies blameless for what their business partners do then all you're going to do is sabotage your own job market in favor of markets in countries that have more lax laws. That is just economic suicide. Companies are mostly in it to make money, so if they can dump sludge in the creek in Mozambique then that is where they'll send their production.
The patented drugs would be much more expensive than the publicly funded ones. I think you just mis-read my sentence since we're saying the same thing.
Now, the R&D component of drug costs isn't actually less expensive with socialized medicine than without (at least, I don't think it would be - it never really has been tried). The difference is just who pays those costs. With a patented drug the costs of R&D are amortized over all the pills made (along with a ton of other costs that would go away with public funding). With a publicly developed drug the costs of R&D are paid by the taxpayers, and then most likely would not be recovered through drug prices.
Privately funded drugs pretty-much only have one model since the price of the product is the only way a company can recover its costs. Publicly funded drugs actually could embrace varying levels of socialism. The NIH could develop a drug and then STILL charge $5 per pill or whatever in which case taxpayers would pay relatively little of the cost (as taxpayers - obviously they pay if they take the drug). Or, Congress could appropriate tax funds for the drug R&D and the pills might cost 10 cents each.
The obvious advantage of funding drugs with taxes is that you can be much more flexible with how those costs impact society. Instead of charging people lots of money to quit smoking you could make the cessation drugs free and charge an extra tax on cigarettes to pay for them. And so on... Drug costs are by their nature a relatively regressive way of collecting money.
Another advantage of cheap drugs is compliance. When drugs are properly used (and not over-advertised/prescribed) the fact is that they save money compared to other therapies. If you can push off surgical interventions on arteries to once every 20 years instead of once every 5 years you save a fortune in medical bills (at a cost that is comparatively cheap even with the current astronomical drug prices). However, when drugs are expensive a poor person is more likely to not take them, and then go to the ER and have that surgery done anyway.
Frontline covered some other aspects of health care, and it really does create good debate points about the whole cradle-to-grave thing. They tracked down a case of some guy who had been in and out of ERs numerous times at a cost of tens of thousands of dollars (I think medicare was paying that one). It turns out he has asthma attacks, and they inspected his home and found that the ceiling was crumbling and the poor condition of the home almost certainly was a big contributing factor. However, the guy couldn't afford to have it properly fixed up. So, we have a case where society is willing to pay $50k in six months or whatever on medical bills, but there is no way that society would pay a contractor $10k to repair and clean up the house. On the other hand, if you really did open that door then the whole USA would become one big homeowner's association.
Like I said, healthcare is complicated when you get beyond the sound bites and political posturing. My feeling is that if you're going to let the sick people die in the streets then do that and quit wasting money on them, but if you're going to pretend to care about them and keep them from dying than you might as well get your hands dirty and improve their lives so that it is more cost-effective, even if your only interest is your own wallet.
Education has the same sorts of problems - we stick kids in classrooms but we can't force them to learn. Now, if you're going to let people who can't hold down a job die in the streets from starvation then do it and quit wasting money on ineffective welfare, but if you are going to pretend to care then you need to get your hands dirty and force the kids to get an education such that most of them actually can be productive (there will always be those who will need more care). And, just as with healthcare our institutions are highly dysfunctional so there are a myriad of problems that need to be solved before that can really happen.
Just another example of the kinds of logic that brought us stuff like:
"This product contains a chemical known to the State of California to cause cancer."
"Water - 99.99999%. In accordance with the New Jersey right to know act, contents partially unknown." Yup, there just might be a carbon atom floating around somewhere in there...