Car analogy: You bought the BMW 325 to impress your friends while driving with the collar of your polo shirt up. It turns out that thieves can steal your muffler for the precious precious platinum in the catalyctic converted. The brand new M3 model developed after the news broke out has the muffler protected by the body. Do you expect a free upgrade from BMW?
+1 for the car analogy. And as far as my ancient Ford truck goes, I don't think they'd issue a recall for anything other than a safety issue. But a BMW? I would indeed expect a product recall from BMW, where they would freely install some "catalytic converter locks" that would be nearly as effective as the body redesign solution you hypothesized.
The replacement boards slide right into the existing locks, which the competitors product will not do.
Yet.
There seems to be a market opportunity here for a vendor who can provide a trustworthy replacement board at a reasonable price. Of course, that means replacing the programming station as well, but it would get a hotel to a potentially better engineered solution, especially if the system was Open Source and scrutinized by the public eye for vulnerabilities.
And I was not arriving at conclusions as to "why" the climate is changing. I'm pointing out that it "is" changing, and that the evidence supports that. The GP was denying that change is happening, which is a completely unsubstantiated claim.
Could this be natural and cyclic? Could this be human-driven? Could it be C: All of the above? That's the real question. And that's just "cause". For "effect", we get into a different set of debates: will the seas rise? Will the increased warmth increase the intensity and/or frequency of calamitous weather? Will certain species die off?
The biggest question is how do we test any of these theories? We can observe the retreat of the polar ice caps, but can't yet prove that the cause is directly related to weather, or to CO2 reflectance, or to any of these global phenomena. Small scale experiments and some studies suggest such things might be the cause, but I agree that there is not yet compelling proof.
What there is plenty of, though, is urgency. The current rates of change are abnormally high, and holding remarkably steady. And people want to know if these changes are permanent, and ultimately what they might mean. Better crops and more food? Bigger storms and more disasters? Both? Neither?
I agree that the answers are yet to be found. But I also agree that we need those answers sooner, rather than later.
An intentionally lethal trap is illegal. But an electric fence is not necessarily lethal. 7,000 volts is lower than a typical cattle fence.
However, the thief may end up merely pissed off and vengeful. If he has a lighter in his pocket, this could turn from burglary to arson, and backfire completely on the property owner.
Make it clear that you are a tech conference, and have no interest in any attribute of a speaker that doesn't directly relate to their technical competence or knowledge in their field. Then make it equally clear that they are obviously the experts in diversity, and as such you will expect them to provide a list of approved ethnicities, genders, sexual orientations, and belief systems, and that they need to give you a percentage of speakers that you will hire to meet the demand. Make them assume the mantle of bigotry. Promise to publish their name in the program as the Diversity Consultant. Go over the top.
If they get where you're going, offer them the choice to drop the whole matter.
But if they insist on going forward, find a local comedian who is a member of an approved diverse group, and hire him or her (herm?) to give a talk on "Diversity in Technology Related Fields".
Make it clear that you are running a tech conference, and that you don't have any preference to speaker attributes that are not relevant to the topic of your conference. But since they are the experts on the topic of "diversity", make them take that role. Have them tell you exactly how many diverse speakers you must have, and have them provide the approved ethnicity list, what percentage of genders, what sexual orientations they must have, and which belief systems need to be represented in order to receive their Stamp of Official Diversity. Go over the top. Paint them to be the ones wearing the mantle of bigotry, and if they are unwilling to participate in the role of Clown General, then offer them the option to drop the whole thing and agree that they will not complain if you hire the most technically competent speakers.
On its surface that sounds like a testable hypothesis. But measuring "belief" is one of those things that's going to be error prone in the extreme no matter what you do, and the wording of the questions would be really hard to keep neutral. I could easily picture such a study being executed very badly. And no matter how it's run, I suspect the results would be used and abused by people with opposing viewpoints to each validate their current opinions.
Finally, the people who don't accept physical evidence from scientific studies are by fiat not going to be swayed by evidence conflicting with their beliefs. If you're looking to change illogical people's minds, logic is not the tool for the job.
Retailers have used cameras hidden in mannequin eyes for over 20 years. And they have been using facial recognition on CCTV feeds since the technology became available to them. This just puts the two products together inside one package.
Not at all. They are taking measurements at a very isolated place, and are sampling the atmosphere at 13,000 feet, which is essentially the upper atmosphere over the pacific. it is about as far as they can be from significant local human sources of carbon dioxide.
Mauna Loa's data also corroborates almost perfectly with the global averages taken from the (now) 66 watch stations, few others of which are located on active volcanoes.
And if the volcano itself was a significant contributor of CO2, why would the graph show an unblemished steady line passing right through the 1984 eruption without so much as a twitch?
There is no way to gather a significant amount of data to suggest that we're doing anything "bad" or that anything "good" we do is working without comparing to some 10,000+ year cycle that we've never observed. Best records for such things go back a few hundred years, and beyond that the data is very sketchy and specific only to specific areas (e.g. ice cores, etc.).
Paleoclimatology is not exactly "sketchy". It studies global trends over millennia, and provides far more information than just a book of thermometer readings from the measly few years humans have been recording such things. There is indeed a large body of data. They use ice core data to determine temperatures and atmospheric composition. They have calibrated those readings based on the few hundred years of written records available. They also corroborate the data with other evidence, such as archaeological and fossil data, and even historical accounts of weather related events. No one piece of data tells the whole story, which is why they have gone to such great lengths to collect as much as possible from a wide variety of sources. Put together, the current body of evidence is scientifically acceptable.
The data is available, it's validated, and it's significant. Instead of continuing to deny climate change is happening, and appearing foolish to people who know better, why not put forth some plausible hypotheses about why you think the climate change that is happening now has natural causes at its core, and offer some tests to validate your theories?
Only because it's not actually a tax. A real tax would return to the public at large, and do some benefit to society as a whole. This is profit that goes straight into the pockets of a few 1%ers, and at best a portion of it trickles down to someone in the form of labor paid for staining a piece of wood on their yacht.
Slot machines are a penalty on stupid people. It doesn't make them smarter, it just deprives them of a few of their meager resources in a way they are too stupid to understand.
Only in a case where the lottery is run by the state, and the proceeds go to benefit the state operating the lottery, is it actually a tax on stupid people (as well as a penalty.)
Nobody would argue that competition benefits the consumer. But how do you think they're going to effectively compete? What are the attributes where they could provide value that nobody else is delivering? Perceived quality of OS: iOS +++; Android ++; Windows Phone ++ Perceived user OS experience: iOS +++; Android ++; Windows +++ Perceived user App experience: iOS +++; Android (variable); Windows (unknown) Cost: iOS ---; Android -; Windows -- App store monopoly: iOS ---; Android +; Windows --- App availability: iOS +++; Android ++; Windows - Coolness factor: iOS +++; Android +; Windows ? Compatibility with previous phone apps (can I transfer my Angry Birds high scores from iOS to Android?): (variable, depends on vendor support for the platforms) Vendor lock in (how much do I have invested in my existing phone that I would throw away if I switched?): (varies over duration of ownership.) Compatibility with friends phones: (entirely dependent on circle of friends)
And I'm sure you can insert a dozen other factors here.
Note that I'm not trying to start a flame war over the coolness of iOS vs Android vs Windows, I'm listing it as a user decision factor. If you want to give Android +++, iOS ---, and Windows +++++, go ahead.
After you list these all out, you have to come up with weights. How heavily does the coolness factor come into the buying decision? What about cost? What do people think about ease of use? And it's not like these are solid numbers. At best, they're educated guesses based on market penetration studies and user surveys.
So after all that analysis, where would Firefox OS fit? How would it break into the market? So far Microsoft has spent more money on advertising Windows Phone than Mozilla has ever had in total, yet they aren't exactly taking over the market. Are they counting on a very loyal user base? Are they going to provide fully open phones, yet have some magically strong UI guidelines that keep third parties in line for providing consistent interfaces? Are they going to give away the OS for free to phone makers?
But this argument all seems tied up around a very extremist idea that one and only one match of a few digits in a database means the courts will find him guilty and hang him. One match will only lead the police to the door of a suspect, where they will not only look for other evidence, but also perform another series of DNA tests. Any errors in collection would then be exposed, and if the suspect is the wrong person, he will be exonerated as a result.
In the U.S. in the case of "forensic evidence-to-registered offenders" matches, the CODIS database only provides a list of candidate matches, and is not acceptable by itself as evidence. Once the candidate is identified, a qualified expert has to analyze the quality of the match. CODIS offender samples are collected one of two ways: via blood draw from convicted felons and prisoners; or by a cheek swab from a person indicted for certain kinds of felonies (kidnapping, sex crimes, violent crimes, burglary). Both require trained personnel to collect; the data in the database is considered reliable.
In some cases, DNA can only establish the likely presence of an offender or his property at or near the crime scene. Blood is possible to explain, as it's not uncommon to accidentally spill, and may be present on surfaces like knives. But in the case of rape or other sex crimes it's especially hard to mistake the sample DNA's source from other contaminated sources. Even so, they can and will perform the tests multiple times on multiple samples taken from the victim, (and from a new sample collected from a cooperative suspect), to ensure that they didn't pick up something contaminated.
But it's not a one-match-and-you're-guilty system.
First, it depends on your latitude. The further north you live, the more you need to prepare for disappointment in the winter. Where I live, bright sunshine delivers a peak of only about 80 watts/sq meter at local solar noon on the winter solstice, while the summer solstice provides me with almost 1500 watts/sq meter at noon. The same panels in a tropical region would deliver more. Obviously, you don't need the A/C when it's 0 degrees out, but some light to fill out the day would be useful.
If you follow the rough estimating guide of 10 watts per square foot, assuming you have a 1500W A/C unit, I think it would take 150 square feet, or about $7800 worth of solar panels. Sorry, but I don't know how much the batteries or controllers you would need would cost. I have to assume they are not free.
Crime scene samples of DNA are of two types: incidental and direct. Incidental might be the DNA you leave by sneezing, exhaling, or shedding dead skin flakes. Direct would be that collected from body fluids deposited by the attacker. It would be hard to claim contamination of a direct sample - not that many non-rapists are accidentally going to leave that kind of evidence behind. They don't just run one cell's worth of DNA. What they do is to run sequences on several different bits of the sample - if they all match, it's not contaminated.
No, they don't run a full genome sequence, but they don't have to. The more alleles they compare, the higher the certainty. Statistics quickly take over and at some point there is no longer the possibility of a chance match, and it simply does not take all that many markers. The US Combined DNA Index System (CODIS) tracks 13 different markers, and is considered strong evidence. Matching 8,000 would be crazy strong proof that either the suspect or his zygotic twin brother committed the crime.
No, you're missing how DNA testing works. One chromosome has millions of nucleotides in a unique sequence, but for Law Enforcement purposes, they don't want to fully sequence everyone's complete genome, because that takes many thousands of dollars and a very long time. Instead, they look at only a dozen or so points along the sequence, and if enough of them match, they go back and do a more thorough comparison of more points in the sequence.
The more closely you are related to the target DNA, the better your sequences match. So if your DNA was a close match to that found at a crime scene, the suspect might be your parents, your children, or your siblings. A more distant relative (a cousin, a grandparent) will have less of a match.
What they were trying to do in this town was to find someone who was distantly related to the suspect. Then they would check that family tree looking for cousins, uncles, brothers. Perhaps by testing that guy's uncle they would discover that he was the father of the suspect. That narrows down the suspect pool to a very finite set.
And the Birthday Paradox doesn't enter into it at all. To solve a crime, they are looking only for one very specific individual: the guy who left his DNA all over the crime scene. They don't care if Alice and Bob share common markers, they only care if Alice matches the suspect or Bob matches the suspect.
What's the point of a tag to a library? One use is to enable quick and efficient checkout. Either a barcode or an RFID chip solves that, but is speed of checkout an actual problem for a home librarian? Another point is to prevent theft, which an RFID tag that can be read from a distance helps solve by placing readers at the exits. Again, not a realistic problem for a home library. So checkout activities really shouldn't factor into consideration for the technology.
Probably the most realistic application for the home librarian is to manage the inventory. That might make more sense for a home library, especially one that fills a large room or spans multiple rooms. The activity of taking inventory then involves reading each label. That would require either the barcode to be visible from the spine, or an RFID tag. Affixing the barcode to the spine is hard to do without damaging the book, and not every item in the catalog even has a spine or is stored on a shelf. RFID works much better for those. RFID is also better if you are trying to locate a specific item that isn't on its proper shelf. Set a reader to beep only if tag 12345 is read, then wave the reader across the shelves. That's much faster than scanning barcodes.
I think his best solution is still to use RFID tags. They're available on sticker rolls (for about $0.35 cents each), but there's nothing saying you have to peel off the backing. You could just place the tag between the pages, which wouldn't damage the book at all. Major chain bookstores have done this with EAS tags for decades. You could also stuff it between the binding and the spine, perhaps holding it in place with a loop of thread. An RFID tag is invisible unless you're actually on the page where the tag is kept, so the value of an old book isn't tarnished by a gaudy barcode.
OK, I get what you're saying now. You want to take the mass of the book out of the equation, so that a fragile spine wouldn't be further damaged or even torn in two by the weight of a heavy book straddling a sharp edge, and all the motion of this mechanism. And I agree.
It looks like the high end commercial book scanners are constructed to take that into account too, where the weight of the book is supported by the covers in a cradle, just like the DIY scanners. They use a vacuum mechanism to draw a single page straight up, then scan both sides of it.
I think there's room for both. Most of my books have little intrinsic value, so if I were scanning my collection I'd have no problem risking them in the automated machine. But I have some that are certainly more fragile than others, and those I'd put in the manual scanning pile.
Got an iPhone? There's a personal inventory app called MyStuff2 ($5.00) that I use for a whole lot of things, and it already comes with "Books" as a category. When you're adding items, it can use the camera for a barcode scanner and look up your titles on Amazon; it then automatically populates fields like title, author, publisher, genre, page count, edition, ISBN, publication date, and cover photo. For books that don't have a barcode, you can manually type the ISBN. Or you can enter all the data manually.
Once an item is in the database, there are action codes you can select. Item Lent is built in by default, and records the current date. It's up to you to type in the name and info of the borrower. When you view a loaned item, there's a convenient "return item" button to tap.
You can browse the list of all items in an action state, such as "lent" or "returned".
The program is very flexible. You can modify the database schema, adding other data you might find useful, like price, vendor, condition, notes, or what have you. You can modify the actions as well. For example, I modified the "lent to" field of the Loan Item activity to be an address book contact field instead of a typed name. So when I look at the loaned item list, I can tap on an item, tap on the contact, then tap dial to call them.
The app supports importing and exporting data a few different ways (CSV, Excel, PDF, HTML) so you can work with it on a separate machine. It can use all kinds of tools to back up the databases, including Box, Dropbox, iCloud, or you can FTP it to your own systems. And it's always with you, which is great if you're just out and about and happen into a book store.
I also have other categories of stuff in the database. I keep data about all of our home appliances and electronics in there, with information about warranties, repair history, replacement parts, and maintenance schedules.
I originally bought it to keep track of our orchid collection, but it's proven very useful for all kinds of home inventory needs. Best $5.00 app I ever bought.
Slashdot Mirror
"First, assume we start with a spherical chicken..."
Car analogy: You bought the BMW 325 to impress your friends while driving with the collar of your polo shirt up. It turns out that thieves can steal your muffler for the precious precious platinum in the catalyctic converted. The brand new M3 model developed after the news broke out has the muffler protected by the body. Do you expect a free upgrade from BMW?
+1 for the car analogy. And as far as my ancient Ford truck goes, I don't think they'd issue a recall for anything other than a safety issue. But a BMW? I would indeed expect a product recall from BMW, where they would freely install some "catalytic converter locks" that would be nearly as effective as the body redesign solution you hypothesized.
The replacement boards slide right into the existing locks, which the competitors product will not do.
Yet.
There seems to be a market opportunity here for a vendor who can provide a trustworthy replacement board at a reasonable price. Of course, that means replacing the programming station as well, but it would get a hotel to a potentially better engineered solution, especially if the system was Open Source and scrutinized by the public eye for vulnerabilities.
And I was not arriving at conclusions as to "why" the climate is changing. I'm pointing out that it "is" changing, and that the evidence supports that. The GP was denying that change is happening, which is a completely unsubstantiated claim.
Could this be natural and cyclic? Could this be human-driven? Could it be C: All of the above? That's the real question. And that's just "cause". For "effect", we get into a different set of debates: will the seas rise? Will the increased warmth increase the intensity and/or frequency of calamitous weather? Will certain species die off?
The biggest question is how do we test any of these theories? We can observe the retreat of the polar ice caps, but can't yet prove that the cause is directly related to weather, or to CO2 reflectance, or to any of these global phenomena. Small scale experiments and some studies suggest such things might be the cause, but I agree that there is not yet compelling proof.
What there is plenty of, though, is urgency. The current rates of change are abnormally high, and holding remarkably steady. And people want to know if these changes are permanent, and ultimately what they might mean. Better crops and more food? Bigger storms and more disasters? Both? Neither?
I agree that the answers are yet to be found. But I also agree that we need those answers sooner, rather than later.
An intentionally lethal trap is illegal. But an electric fence is not necessarily lethal. 7,000 volts is lower than a typical cattle fence.
However, the thief may end up merely pissed off and vengeful. If he has a lighter in his pocket, this could turn from burglary to arson, and backfire completely on the property owner.
Go re-watch Robocop, and keep an eye out for the "Magnavolt" car theft deterrent commercial. Best commercial on TV!
It'll be a lot of fun to see the guy's face when they steal his electric fence wire.
Make it clear that you are a tech conference, and have no interest in any attribute of a speaker that doesn't directly relate to their technical competence or knowledge in their field. Then make it equally clear that they are obviously the experts in diversity, and as such you will expect them to provide a list of approved ethnicities, genders, sexual orientations, and belief systems, and that they need to give you a percentage of speakers that you will hire to meet the demand. Make them assume the mantle of bigotry. Promise to publish their name in the program as the Diversity Consultant. Go over the top.
If they get where you're going, offer them the choice to drop the whole matter.
But if they insist on going forward, find a local comedian who is a member of an approved diverse group, and hire him or her (herm?) to give a talk on "Diversity in Technology Related Fields".
Make it clear that you are running a tech conference, and that you don't have any preference to speaker attributes that are not relevant to the topic of your conference. But since they are the experts on the topic of "diversity", make them take that role. Have them tell you exactly how many diverse speakers you must have, and have them provide the approved ethnicity list, what percentage of genders, what sexual orientations they must have, and which belief systems need to be represented in order to receive their Stamp of Official Diversity. Go over the top. Paint them to be the ones wearing the mantle of bigotry, and if they are unwilling to participate in the role of Clown General, then offer them the option to drop the whole thing and agree that they will not complain if you hire the most technically competent speakers.
Yes, but that's a whole year away. In just ten months the iPhone++ comes out and we can all start dreaming again!
On its surface that sounds like a testable hypothesis. But measuring "belief" is one of those things that's going to be error prone in the extreme no matter what you do, and the wording of the questions would be really hard to keep neutral. I could easily picture such a study being executed very badly. And no matter how it's run, I suspect the results would be used and abused by people with opposing viewpoints to each validate their current opinions.
Finally, the people who don't accept physical evidence from scientific studies are by fiat not going to be swayed by evidence conflicting with their beliefs. If you're looking to change illogical people's minds, logic is not the tool for the job.
Retailers have used cameras hidden in mannequin eyes for over 20 years. And they have been using facial recognition on CCTV feeds since the technology became available to them. This just puts the two products together inside one package.
Not at all. They are taking measurements at a very isolated place, and are sampling the atmosphere at 13,000 feet, which is essentially the upper atmosphere over the pacific. it is about as far as they can be from significant local human sources of carbon dioxide.
Mauna Loa's data also corroborates almost perfectly with the global averages taken from the (now) 66 watch stations, few others of which are located on active volcanoes.
And if the volcano itself was a significant contributor of CO2, why would the graph show an unblemished steady line passing right through the 1984 eruption without so much as a twitch?
There is no way to gather a significant amount of data to suggest that we're doing anything "bad" or that anything "good" we do is working without comparing to some 10,000+ year cycle that we've never observed. Best records for such things go back a few hundred years, and beyond that the data is very sketchy and specific only to specific areas (e.g. ice cores, etc.).
Paleoclimatology is not exactly "sketchy". It studies global trends over millennia, and provides far more information than just a book of thermometer readings from the measly few years humans have been recording such things. There is indeed a large body of data. They use ice core data to determine temperatures and atmospheric composition. They have calibrated those readings based on the few hundred years of written records available. They also corroborate the data with other evidence, such as archaeological and fossil data, and even historical accounts of weather related events. No one piece of data tells the whole story, which is why they have gone to such great lengths to collect as much as possible from a wide variety of sources. Put together, the current body of evidence is scientifically acceptable.
The data is available, it's validated, and it's significant. Instead of continuing to deny climate change is happening, and appearing foolish to people who know better, why not put forth some plausible hypotheses about why you think the climate change that is happening now has natural causes at its core, and offer some tests to validate your theories?
Only because it's not actually a tax. A real tax would return to the public at large, and do some benefit to society as a whole. This is profit that goes straight into the pockets of a few 1%ers, and at best a portion of it trickles down to someone in the form of labor paid for staining a piece of wood on their yacht.
Slot machines are a penalty on stupid people. It doesn't make them smarter, it just deprives them of a few of their meager resources in a way they are too stupid to understand.
Only in a case where the lottery is run by the state, and the proceeds go to benefit the state operating the lottery, is it actually a tax on stupid people (as well as a penalty.)
Are they going to give away the OS for free to phone makers?
Well, obviously, if it's open source.
Great. Now put a revenue generating stream in there, or at least an inexhaustible source of funding.
Nobody would argue that competition benefits the consumer. But how do you think they're going to effectively compete? What are the attributes where they could provide value that nobody else is delivering?
Perceived quality of OS: iOS +++; Android ++; Windows Phone ++
Perceived user OS experience: iOS +++; Android ++; Windows +++
Perceived user App experience: iOS +++; Android (variable); Windows (unknown)
Cost: iOS ---; Android -; Windows --
App store monopoly: iOS ---; Android +; Windows ---
App availability: iOS +++; Android ++; Windows -
Coolness factor: iOS +++; Android +; Windows ?
Compatibility with previous phone apps (can I transfer my Angry Birds high scores from iOS to Android?): (variable, depends on vendor support for the platforms)
Vendor lock in (how much do I have invested in my existing phone that I would throw away if I switched?): (varies over duration of ownership.)
Compatibility with friends phones: (entirely dependent on circle of friends)
And I'm sure you can insert a dozen other factors here.
Note that I'm not trying to start a flame war over the coolness of iOS vs Android vs Windows, I'm listing it as a user decision factor. If you want to give Android +++, iOS ---, and Windows +++++, go ahead.
After you list these all out, you have to come up with weights. How heavily does the coolness factor come into the buying decision? What about cost? What do people think about ease of use? And it's not like these are solid numbers. At best, they're educated guesses based on market penetration studies and user surveys.
So after all that analysis, where would Firefox OS fit? How would it break into the market? So far Microsoft has spent more money on advertising Windows Phone than Mozilla has ever had in total, yet they aren't exactly taking over the market. Are they counting on a very loyal user base? Are they going to provide fully open phones, yet have some magically strong UI guidelines that keep third parties in line for providing consistent interfaces? Are they going to give away the OS for free to phone makers?
Sorry, 8000 samples, yes.
But this argument all seems tied up around a very extremist idea that one and only one match of a few digits in a database means the courts will find him guilty and hang him. One match will only lead the police to the door of a suspect, where they will not only look for other evidence, but also perform another series of DNA tests. Any errors in collection would then be exposed, and if the suspect is the wrong person, he will be exonerated as a result.
In the U.S. in the case of "forensic evidence-to-registered offenders" matches, the CODIS database only provides a list of candidate matches, and is not acceptable by itself as evidence. Once the candidate is identified, a qualified expert has to analyze the quality of the match. CODIS offender samples are collected one of two ways: via blood draw from convicted felons and prisoners; or by a cheek swab from a person indicted for certain kinds of felonies (kidnapping, sex crimes, violent crimes, burglary). Both require trained personnel to collect; the data in the database is considered reliable.
In some cases, DNA can only establish the likely presence of an offender or his property at or near the crime scene. Blood is possible to explain, as it's not uncommon to accidentally spill, and may be present on surfaces like knives. But in the case of rape or other sex crimes it's especially hard to mistake the sample DNA's source from other contaminated sources. Even so, they can and will perform the tests multiple times on multiple samples taken from the victim, (and from a new sample collected from a cooperative suspect), to ensure that they didn't pick up something contaminated.
But it's not a one-match-and-you're-guilty system.
First, it depends on your latitude. The further north you live, the more you need to prepare for disappointment in the winter. Where I live, bright sunshine delivers a peak of only about 80 watts/sq meter at local solar noon on the winter solstice, while the summer solstice provides me with almost 1500 watts/sq meter at noon. The same panels in a tropical region would deliver more. Obviously, you don't need the A/C when it's 0 degrees out, but some light to fill out the day would be useful.
If you follow the rough estimating guide of 10 watts per square foot, assuming you have a 1500W A/C unit, I think it would take 150 square feet, or about $7800 worth of solar panels. Sorry, but I don't know how much the batteries or controllers you would need would cost. I have to assume they are not free.
Crime scene samples of DNA are of two types: incidental and direct. Incidental might be the DNA you leave by sneezing, exhaling, or shedding dead skin flakes. Direct would be that collected from body fluids deposited by the attacker. It would be hard to claim contamination of a direct sample - not that many non-rapists are accidentally going to leave that kind of evidence behind. They don't just run one cell's worth of DNA. What they do is to run sequences on several different bits of the sample - if they all match, it's not contaminated.
No, they don't run a full genome sequence, but they don't have to. The more alleles they compare, the higher the certainty. Statistics quickly take over and at some point there is no longer the possibility of a chance match, and it simply does not take all that many markers. The US Combined DNA Index System (CODIS) tracks 13 different markers, and is considered strong evidence. Matching 8,000 would be crazy strong proof that either the suspect or his zygotic twin brother committed the crime.
No, you're missing how DNA testing works. One chromosome has millions of nucleotides in a unique sequence, but for Law Enforcement purposes, they don't want to fully sequence everyone's complete genome, because that takes many thousands of dollars and a very long time. Instead, they look at only a dozen or so points along the sequence, and if enough of them match, they go back and do a more thorough comparison of more points in the sequence.
The more closely you are related to the target DNA, the better your sequences match. So if your DNA was a close match to that found at a crime scene, the suspect might be your parents, your children, or your siblings. A more distant relative (a cousin, a grandparent) will have less of a match.
What they were trying to do in this town was to find someone who was distantly related to the suspect. Then they would check that family tree looking for cousins, uncles, brothers. Perhaps by testing that guy's uncle they would discover that he was the father of the suspect. That narrows down the suspect pool to a very finite set.
And the Birthday Paradox doesn't enter into it at all. To solve a crime, they are looking only for one very specific individual: the guy who left his DNA all over the crime scene. They don't care if Alice and Bob share common markers, they only care if Alice matches the suspect or Bob matches the suspect.
In other words, you can always put on more clothes, but there's a limit to what you can take off.
What's the point of a tag to a library? One use is to enable quick and efficient checkout. Either a barcode or an RFID chip solves that, but is speed of checkout an actual problem for a home librarian? Another point is to prevent theft, which an RFID tag that can be read from a distance helps solve by placing readers at the exits. Again, not a realistic problem for a home library. So checkout activities really shouldn't factor into consideration for the technology.
Probably the most realistic application for the home librarian is to manage the inventory. That might make more sense for a home library, especially one that fills a large room or spans multiple rooms. The activity of taking inventory then involves reading each label. That would require either the barcode to be visible from the spine, or an RFID tag. Affixing the barcode to the spine is hard to do without damaging the book, and not every item in the catalog even has a spine or is stored on a shelf. RFID works much better for those. RFID is also better if you are trying to locate a specific item that isn't on its proper shelf. Set a reader to beep only if tag 12345 is read, then wave the reader across the shelves. That's much faster than scanning barcodes.
I think his best solution is still to use RFID tags. They're available on sticker rolls (for about $0.35 cents each), but there's nothing saying you have to peel off the backing. You could just place the tag between the pages, which wouldn't damage the book at all. Major chain bookstores have done this with EAS tags for decades. You could also stuff it between the binding and the spine, perhaps holding it in place with a loop of thread. An RFID tag is invisible unless you're actually on the page where the tag is kept, so the value of an old book isn't tarnished by a gaudy barcode.
Here's an interesting paper on the topic of using RFID in libraries: http://www.bic.org.uk/files/pdfs/090109%20library%20guide%20final%20rev.pdf
OK, I get what you're saying now. You want to take the mass of the book out of the equation, so that a fragile spine wouldn't be further damaged or even torn in two by the weight of a heavy book straddling a sharp edge, and all the motion of this mechanism. And I agree.
It looks like the high end commercial book scanners are constructed to take that into account too, where the weight of the book is supported by the covers in a cradle, just like the DIY scanners. They use a vacuum mechanism to draw a single page straight up, then scan both sides of it.
I think there's room for both. Most of my books have little intrinsic value, so if I were scanning my collection I'd have no problem risking them in the automated machine. But I have some that are certainly more fragile than others, and those I'd put in the manual scanning pile.
Got an iPhone? There's a personal inventory app called MyStuff2 ($5.00) that I use for a whole lot of things, and it already comes with "Books" as a category. When you're adding items, it can use the camera for a barcode scanner and look up your titles on Amazon; it then automatically populates fields like title, author, publisher, genre, page count, edition, ISBN, publication date, and cover photo. For books that don't have a barcode, you can manually type the ISBN. Or you can enter all the data manually.
Once an item is in the database, there are action codes you can select. Item Lent is built in by default, and records the current date. It's up to you to type in the name and info of the borrower. When you view a loaned item, there's a convenient "return item" button to tap.
You can browse the list of all items in an action state, such as "lent" or "returned".
The program is very flexible. You can modify the database schema, adding other data you might find useful, like price, vendor, condition, notes, or what have you. You can modify the actions as well. For example, I modified the "lent to" field of the Loan Item activity to be an address book contact field instead of a typed name. So when I look at the loaned item list, I can tap on an item, tap on the contact, then tap dial to call them.
The app supports importing and exporting data a few different ways (CSV, Excel, PDF, HTML) so you can work with it on a separate machine. It can use all kinds of tools to back up the databases, including Box, Dropbox, iCloud, or you can FTP it to your own systems. And it's always with you, which is great if you're just out and about and happen into a book store.
I also have other categories of stuff in the database. I keep data about all of our home appliances and electronics in there, with information about warranties, repair history, replacement parts, and maintenance schedules.
I originally bought it to keep track of our orchid collection, but it's proven very useful for all kinds of home inventory needs. Best $5.00 app I ever bought.
http://www.maddysoft.com/iphone/mystuff/