Heads are a bit large for birth, but you're missing a critical point. Without that point you might logically decide that we are dangerously close to not being able to birth children.
The bones in a newborn's skull have not fused together, they are more like floating plates. This permits them to be rearranged somewhat and for he head to be squeezed into a longer conic shape, allowing birth.
So perhaps heads are not too large for birth, perhaps we just get born a different way that someone's opinion of the "right" way.
As far as the "only" creature to experience pain in child birth, I'll address that with the common species wide misconception that humans are super special and have exclusive domain over acts that are associated with living.
We haven't defined pain well in Humans, so it would not be a surprise (especially in Sagan's time frame) to believe that it is not present in other animals. The human hubris for exclusivity of actions found amongst all life is immense, and even with all of the recent findings, it is only eroding slowly. Remember, back then:
Only humans could conceive, build, and use tools. Only humans could transfer knowledge to their offspring via teaching. Only humans have a soul. Only humans could cognate. Only humans alter their environment to make up for shortcomings in natural selection based adaptation. and so on...
In truth some of these statements were determined to be 100% false, others have no basis to test, and a few require definitions conveniently borrowed from religion.
As far as I know, there's little to no ability to subjectively measure pain, which is reported differently from individual to individual. When you don't trust someone, you think they're not in as much pain as they report. When someone is raised with a background of expecting pain as a part of life, they're not really hurting from obviously broken bones.
That's why doctors ask you on a 1 to 10 scale, they're trying to normalize the pain against your own perception. It doesn't really tell us anything subjective, but it attempts to normalize objective findings against a scale that isn't really there.
We all know the bible has been translated, yet many of the religious insist that it's the exact word of God; somehow God inspired the original authors and all of the translators who came afterwards.
When the translations fall short, it's time to redefine the words or look for alternate meanings that somehow prevent literal interpretation, yet such analysis only replaces one literal interpretation with another. So days turns into 2000 year periods, but only in this context, not in these other passages where they mean 24 hour periods.
No one suggests the obvious; it is poorly translated with respect to common English and should be fixed. That's because any religion worth its salt will quickly bombast any common tongue translation as non-traditional, misinterpreted, twisting the word of God, etc.
In other words, the circle of complaints is complete. They like it how it is since it provides plenty of opportunity to build cohesion by decrying everything else as foreign and wrong, and it provides plenty of opportunity to adapt to common situations by reinterpreting passages to mean things they likely never intended. Playing this game of contextually sensitive interpretation and redefining reminds me of the excellent film Alphaville.
For those who don't know the film. Words are defined in dictionaries that are obviously edited and constantly updated to limit the populace's ability to think outside of the accepted dogma. It's really a small part of the story, but a huge part of the ideas behind the film.
I don't know where the "farmers like DST" meme can from, but it should go back there very quickly.
Farmers don't benefit from DST, because farmers are like construction workers; they work outside. That means that they maximize their use of daylight, not that the daylight gets shifted around to suit their needs.
When the sun rises, it rises the next day which is about 24 hours from the last time plus or minus some number of seconds depending on latitude. Setting your clocks to whatever time you wish won't alter this behaviour, and if you all agree that today you'll ignore an hour or some other day you'll duplicate an hour, then that's fine. However, it has nothing to do with the actual observations of the world you live in, which is the farmer's realm.
Farmers typically hate DST because it means they have to alter their schedule because the banks honour DST while the farmers honour the setting and rising of the sun.
Kill the farmer DST meme, it's misinformation at its finest.
After following your instructions, I noticed that I noticed that my thumb and index finger formed a R, and I now have two right hands, with one right hand a bit more to the right than the other.
More weighing on my mind is the interaction between the following forces:
1. Reduction in manpower costs spur adoption by the military. 2. Reduction in manpower spurs the deployment of under supervised robot teams. 3. Trapping and reprogramming of robots becomes a more viable means of using the expensive hardware against their (former) owners. 4. Cheap ways of disabling these robots will provide a financial model of warfare that allows countries with weak budgets and manpower to do billions of dollars of damage.
We do know that CO2 traps radiation from reflecting back off the earth into space. Radiation is energy, and that energy trapped on the planet will eventually translate to heat (the apparent fate of all energy).
Now if the magnetosphere shifts, or if God's global warming alarm clock goes off, it's not going to make this little bit of physical phenomenon (CO2 traps radiation in the atmosphere) go away. So instead of brining a bunch of red herrings to the table, either prove that any of your other ideas have a bigger impact than CO2 in the atmosphere (which would expand the realm of findings to date) or address the problem of CO2 in the atmosphere (which all of the findings to date show is the main, but not sole, culprit).
We've already done geo-engineering by putting the greenhouse gasses into the atmosphere in the first place. It requires less creative engineering to stop putting them up there, and we know that greenhouse gasses from (whatever) source raise ambient temperature. Therefore, not putting greenhouse gasses in the atmosphere is a generally plausible solution, even if it means we have to change our lifestyle.
Stuff like sprinkling the ocean with reflective material doesn't have a very well known effect because we haven't tinkered with the planet in that way. I'm just a lowly ex-Biologist, but immediately after reading the description, alarm bells are going off like wild.
These particles will be exposed to one of the world's largest food chains, possibly poisoning one of the greatest stores of bio-mass in existence. Life will probably manage to struggle on, but even a reduction in bio-mass in the ocean has a very profound impact on the land dwelling population of the world.
We already have significant problems with mercury content of many types of edible marine life. They don't eat a lethal dose at any given time, but their bodies accumulate the poison until it presents problems for their predators. Such systems of poison storage causes collapses of the predators first, which then cause blooms of the prey, which then cause mass extinctions of the prey due to starvation. In this respect, animals are like humans, willing to watch the whole species go to hell in a hand bucket as long as they can exploit the environment for everything its got.
Even if they're plastic particles, plastics leech phenols which seem to cause some health problems. Even if they're 100% inert (perhaps ceramic?) small particles are deadly in their own right. Particular atmospheric pollution does it's damage whether you get it from living in a city or other means, some people can't get enough of particular pollution so they take up smoking;)
I wonder if the researchers have considered how easy it would be to live, work, sleep, and eat in a house where every interior surface was covered with a fine layer of glitter.
Re:This is a huge amount of work
on
Linux 2.6.27 Out
·
· Score: 2, Interesting
Your humor is appreciated, but there's a large body of evidence that the best programmers can be more than 10x as effective than the rank-and-file, which can be (more than) 10x as effective as the bad programmers.
So a methodology that boosts output of two programmers 400% isn't really promising the impossible. Just consider that they are likely talking about average programmers, and the new environment keeps them engaged at an above-average attention level.
Peer pressure can make people do incredible things (good and bad). Altering the environment to make it more likely that good code is produced isn't snake oil, provided the results do follow. I've never seen a methodology that denies certain techniques are beneficial; instead they seem to argue over different combinations and inclusions of techniques that were observed to work.
Now, as you pointed out, the acronym spewing masses often don't know what they're saying. For that group, any methodology results in the same thing: changing the appearance of the current methodology without altering how the actual work is done. Months later, they have the perfect scapegoat: the methodology sucks.
As another poster already noted: Nobel prizes are for specific discoveries, not a person's reputation since the discovery.
In another reply mblase mentions Einstein. I'm going to do one better than that and mention Linus Pauling. I say one better, because Linus Pauling won two Nobel prizes, and he might still be the only person to have two Noble prizes to his name. His work laid the foundation for viewing shape as a critical element in determining the effects of molecules in living systems. He also won the Nobel Peace Prize.
The man was brilliant, I heard him talk before he passed, and he went to his grave trying to find the chemistry-level solution for the widely accepted idea that vitamin-C provides additional protection against the common cold. Detractors of his efforts joked about Pauling's vitamin C.
We don't want our best minds weighing the benefits of not working after a major discovery should they be in the running for a Nobel.
Basically, it's a retrofit of a parking garage. You're charging the entire time you are at work, minus the lunch break if you're inclined to take one.
No need to burgeon the best solution for the old problem (pumping gas) with the solution for the new problem. I mean, I can't buy hay at my local gas station either; perhaps that's because they didn't start out as barns.
Whenever you can't disprove something, make it encompass so many other items that it becomes incomprehensible without the services of a complete auditing team. Then extend it just a little to include data that you probably can never collect.
We saw this with TCO studies that tried their hardest to show that Linux might not be cheaper than $800 per seat licenses of Windows. While many people saved money by switching (Sherwin Williams, Fender Gituars, etc.) most complained about the intangible losses that they couldn't prove which justified their non-action.
We saw this with global warming. Most argued about specific point data that didn't follow the average (hint there's outliers in every interesting data set), argued that costs would never be calculable for the issue, and built a model of "let's wait and see" which justified their non-action.
We're seeing this with electric (or semi-electric) vehicles. It should be enough to note that we will mostly be charging the cars during non-peak hours, when the grid's capacity is most likely to meet the demand. It should be enough to note that for the same amount of used energy, the electric company can provide it to us cheaper than the gasoline distribution chain. It should be enough to note that at least a dozen ways in which the vehicles are cheaper to maintain offset the initial costs of not having a mass market's cost structure.
Instead, we have to calculate the exact dollar of every line man, telephone pole, coal miner's life insurance plan, etc, ad infinitum. I'll give you a hint: The power company already does this, and it's called your electric bill, which is still cheaper than your gasoline bill.
Just because you don't have it itemized doesn't mean it's more expensive.
For those that enjoy these sort of games, have you even considered the operational costs of the thousands of oil tankers? How about the costs of all those oil platforms? How about the costs of the fire policies on those platforms? etc... ad infinitum.
Prove to me that my electric bill will be higher than my gas bill, and I'll go with the parent poster's observation. It is the only sensible metric. And oddly enough, it should include all of those "other" costs, because if it didn't the company would be out-of-business before you know it.
Well, to start, can you reproduce the binary? If not, then you'll have to find those missing libraries, classes, and whatever else. In the event that simple searches don't turn up anything, reversing the binary might be your starting point. Remember, in your first pass you don't have to reverse for 100% comprehension, only for equivalent compilation.
From there it's just a lot of refactoring till you have code that looks something like the algorithm described in the documentation. Hopefully by then you can identify where the code deviates from the documentation and identify if that deviation is due to the incorrect model, or other factors that failed to get "updated" in the documentation.
Fixing by rewriting is almost always a doomed proposition for your case. It is rare that code follows the documentation 100% because life-learned lessons in the code don't get reflected in pre-existing documentation. It's easier to say "oh yeah, I forgot about that case, just fix it this way" then to pull out approved documents and go through another round of review to add a trivial (but critical) addendum.
Sure, this technique seems slower, but every step forward is measurable, and there are no surprises (you can always reproduce the binary as it already works). Rewriting gets you 80% done much faster, but it's hard to say if you'll finish it more quickly: there will be lots of opportunity for surprises with no guarantee of "current functionality" at any point in the game.
I don't think you expressed yourself clearly, or perhaps you meant to write something other than what you wrote.
"Testing cannot detect error with probability significantly greater than zero, unless the system under test is trivially small".
First, when a function doesn't work, it's not "probably not working" for a specific set of inputs, it's 100% not working for a specific set of inputs. 100% is significantly greater than zero.
Second, the whole concept of a unit test is to test something small, so even though your statement's first clause makes no sense, the second clause makes the first clause not applicable to unit testing.
Third, unit testing tests a single class's behaviour. Testing combinations of classes better falls under integration testing. There's no combinatorial explosion of testing when you test only one item.
Naturally to exploit this advance in software testing (and development) you have to buy into it's philosophy. If your code isn't very object-oriented (meaning that your classes don't have distinct and unshared responsibilities), then it will be nigh-useless to test if the unit (class) handles it's responsibilities.
To test a class, you have to write code. Once you write the code, it makes no sense to run it manually every time you wish to verify the class still meets it's responsibilities. So you create a pass / fail framework, and volia! You have unit testing.
It doesn't mean that other testing isn't important, and if you rearchitect the interfaces between your classes, you'll have to rearchitect the unit tests to match. But hey, there's no silver bullet that only shoots the work involved in changing the interfaces between classes except inaction.
PS. The least cost gun wouldn't come with a safety either, which would be unnecessary in a world where accidental discharges never occur. In the real world, guns get dropped. In the real software development process, changes get made without perfect knowledge of all they ways the class is currently used (or abused).
Don't advocate against rational low cost safety. The cost of unit tests get amortized over each regression test (which should be performed after each build), and are a lot less expensive than shipping buggy code to a client.
Good tests are tests that verify the required functionality of the unit, bad tests are tests that verify things that cannot break or are the functional equivalent of the human appendix.
Exactly how would you preform good unit testing in such a way that you could change the code in a critical way and not have it break? I know it seems like an easy cop-out to say it, but if you could do such a thing, then your unit tests are incomplete as they don't test the required functionality.
That's not to say that you don't still have to do integration testing, acceptance testing, performance testing, usability testing, etc.
The only way a reader can read this book and not come away with a pocketful of gems is lack of experience or so much experience that the book bores him. Experience in this case isn't tied to years of programming, it's tied to having to fix truly horrid messes.
This is one of the best introductions into managing a code base that is foreign, hairy, and incomprehensible. I've been doing maintenance programming on and off for years now, and while it's not as glamorous as clean slate design, it's much more difficult.
Fowler's Refactoring gives you the transformation templates, just like Design Patterns gives you the code structure templates. This book gives you the instruction into code maintenance discipline. It's well worth a read, even if it leaves you wanting in a (very) few places. Don't think of it as "Refactoring for Dummies", it's more like "How to use Fowler's books in an enterprise and not go insane". Naturally there's a little duplication of information, but that's not the point of the message.
There seems to be a bit of an anti-testing crowd reacting to passages in the book. While I understand their gripes, let me assure you that either you're going to test your code or the customer will. If the customer finds too many, then you've squandered your good will in exchange for not writing tests. Not a good exchange in my opinion.
Another book along this line that I also highly recommend is "Software Exorcism" by Bill Blunden. Perhaps his quirky humour and old school usenet references will throw off a few acolytes, but there's plenty of treasure in there.
By the way, the bigger problem is with kids who do the work but don't think. I have lots of students who copy their friends' work, so they have great homework grades, but bomb tests because they have no clue what they're talking about.
How is copying their friends' work "doing the work?"
How can you admonish their lack of thinking when your own example has logic errors like this?
I don't buy the "pay the teachers" solution; it's too self-serving and often comes from those close to teaching.
I know teachers that are making pretty close to 50k a year. While that might not be too much in your corner of the U.S. of A., over here it's pretty good pay. Close to what a police officer makes, an more than the typical fireman.
This pay is up from the previously abysmal 28k a year about a decade ago, but the quality of eduction seems to have gone down. Note that I tutor, so I have been in indirect contact with teachers for years.
To illustrate an example, a pupil of mine had a math answer marked wrong because they considered in an inequality test the possibility that the extreme of the solution was in the solution set. Sure enough, it was and they were "deducted" for including the equals sign. This pupil was irate that they were wrong, and asked me to look at the problem.
Sure enough, the student was right. I asked if they talked to the teacher, and they did. The teacher had replied than in such circumstances they graded according to how the majority of the class interpreted the question. So since +50% of the class read the question to mean "only greater than", my pupil was wrong. I wrote a letter detailing how my student was correct (with a formal proof, hey it is math) and the teacher refused to even read the letter.
I'm sure there are still good teachers out there, but after the example of this one power monger.... I'd say let the lot of them starve. More money hasn't solved the problem, in fact it's nigh impossible for a good teacher to do better than a bad one provided they have the same time in service and are servicing the same grade in the same school district. The teacher's union has made sure of that in the three school districts that I've rubbed elbows with.
Old school solutions to this old school problem have existed for decades (and perhaps centuries).
A student that doesn't like their grade in a class can retake the class in lieu of an elective. The new grade replaces the old one, even if it is lower
This has been used with success at many an "upper crust" university for years. Students that don't / can't / won't apply themselves won't waste time retaking classes. Few people can afford the extra time to game the system for higher grades. In short, no new solution was needed.
So why did they come up with a completely new solution? Because now that the school doesn't have that small percentage of 0% students dragging them down, they probably appear to be a whole letter grade higher (on average) for their review boards. Odds are that other schools will either protest until this one changes policy, or quickly follow suit so they don't appear to be behind in comparison to the "new improvement" of a cooked statistic.
Considering what I fear, I'd wager on them matching the policy. That way they can (in their own minds) appear to have improved too! Don't tell them that when everybody improves equally then there's no change in the ranking. That would just spoil the game for everyone.
Considering that they were previously sliding by with 0%; how does sliding by with 50% motivate them?
I understand the "second start" idea; but in a university that means you have to drop out of school for around ten years and come back. What exactly in high school provided them with the new insight as to the importance of the grading game? Why should they get a "second start" in every class? Why should it be available witout the harsh life lessons that a decade of living with your mistakes can provide?
I would argue that gym should be pass / fail, with pass meaning you participated.
Either that, or you have to do some pretty stupid stuff to "level the playing field" for those that are physically gifted in the select sports that your gym has decided to test.
My brother completely rocked in ping pong back in high school (could play for days without dropping a single game), that should more than make up for his inability to hit a baseball, or his less than stellar basketball skills.
Better yet.
Make the grading percentile distribution more like:
A - 100% - 81%
B - 80% - 61%
C - 60% - 41%
D - 40% - 21%
F - 20% - 0%
At least then they will have coherency between letter grade and percentile of accomplishment. With their current distribution, they have no coherency because a student that performs 50% is equal to one that performs nothing.
As far as the admitting colleges go, they will quickly draft their own plans to adjust for the new grading policy, probably relying even more so on the SAT and other measures to determine their admittance criteria. As far as the school is concerned they just doubled the number of "A" students, even if it was only done by lowering the bar for an A.
If what they were suggesting was padding everyone's score by 50 percentage points, then it would be fair (if awkward). Instead what they are suggesting is padding the worst performer's score by 50 percentage points. In statistics, this would be called "cooking the books", and I'll bet they're cooking the books for more than just "a second chance, whenever the student tries to take it". I'll bet that the new point system is presented to performance boards as equal to those school systems that let a student hit dead bottom zero.
If you want to provide a "second chance" to achieve, do what other institutions have done. Let the student take the course again, with the new grade replacing the old grade. It costs the student an elective and another four months of their life; that makes sure it won't be abused by the student body: time is precious. It maintains the current standard of the school because the course will likely be taught the same way.
What they are doing is unconscionable from a statistics point of view; basically they are taking the numbers they don't like and changing them to 50. The "average" will likewise jump (even thought no corresponding jump in work will be performed). Kudos for them on learning how to lie with statistics. Shame on them for doing it by substituting undesirable values with those more palatable.
If this is hacking, I guess shifting your car into 3rd gear is hacking the system to go a little faster than 2nd gear.
He used the system as it was designed. He didn't exploit any bug or software weakness of the system, the weakness was inherit in the design of the software; what yahoo considered a feature is really a steaming pile of problems-to-be-addressed-later.
Palin wasn't forced to use yahoo, nor did she do any kind of analysis of how suitable yahoo would be for her (admittedly illegal) endeavours. This means that she flunks on my tech ticket. Add to that the use of real world answers which can easily be googled, and it's obvious she's not even considered what security means.
It would have been trivial for her to put down a few dollars a month ($10/mo) and have a better (and supported) email account. Better yet, she could have her own email server set up in a private virtual machine ($20/mo + setup costs).
She should be used to having someone else do stuff for her, being a governor. Oh wait, that's it! She couldn't ask anyone else to help her because they'd likely report her ass for breaking the law!
I don't like what the guy did to get his info, but really this starts too look more like a botched attempt at whistleblowing than an act driven by criminal intent. I mean, where's the damage? He didn't delete anything, nor did he send emails posing as the governor. He even went out of his way to tell her that her account had been accessed by himself. She wasn't denied access to her account, because she could use the same stupid password reset web page he used!
Odds are this guy's going to have the book thrown at him hard. All over an act he performed that didn't cost Palin a dollar and exposed that she was consistently and deliberately violating her own state's law.
Every bit of GPL software is licensed. If your and end user, then it's licensed for you. To use it, you have an End User License Agreement.
The Gentoo bunch have their point, but really they're making a mountain out of a molehill. They should rebrand Firefox and forget it. They shouldn't be moaning about having to display a licence on startup, which is exactly what they don't do for inkscape, ls, evince, cp, evolution, rm, et. al.
If you don't like the idea that someone would trademark their project name, then fine. Remove the Linux kernel as that's trademarked. Or, better yet, please download the newest copy of firefox at http://www.firefax.com/ Yep, that's firefax, the backup domain (to offload the main one). No, no, no!!! We don't embed spyware into OUR browers, cause it's called Firefox! It is (almost) the same one you've been using for years!!! MUAHAHAHAHAHA!!!!
Now, you can still be upset about it, but hopefully you can see a reason or two to enforce a trademark, even when your main product is open source.
Slashdot Mirror
Heads are a bit large for birth, but you're missing a critical point. Without that point you might logically decide that we are dangerously close to not being able to birth children.
The bones in a newborn's skull have not fused together, they are more like floating plates. This permits them to be rearranged somewhat and for he head to be squeezed into a longer conic shape, allowing birth.
So perhaps heads are not too large for birth, perhaps we just get born a different way that someone's opinion of the "right" way.
As far as the "only" creature to experience pain in child birth, I'll address that with the common species wide misconception that humans are super special and have exclusive domain over acts that are associated with living.
We haven't defined pain well in Humans, so it would not be a surprise (especially in Sagan's time frame) to believe that it is not present in other animals. The human hubris for exclusivity of actions found amongst all life is immense, and even with all of the recent findings, it is only eroding slowly. Remember, back then:
Only humans could conceive, build, and use tools.
Only humans could transfer knowledge to their offspring via teaching.
Only humans have a soul.
Only humans could cognate.
Only humans alter their environment to make up for shortcomings in natural selection based adaptation.
and so on...
In truth some of these statements were determined to be 100% false, others have no basis to test, and a few require definitions conveniently borrowed from religion.
As far as I know, there's little to no ability to subjectively measure pain, which is reported differently from individual to individual. When you don't trust someone, you think they're not in as much pain as they report. When someone is raised with a background of expecting pain as a part of life, they're not really hurting from obviously broken bones.
That's why doctors ask you on a 1 to 10 scale, they're trying to normalize the pain against your own perception. It doesn't really tell us anything subjective, but it attempts to normalize objective findings against a scale that isn't really there.
We all know the bible has been translated, yet many of the religious insist that it's the exact word of God; somehow God inspired the original authors and all of the translators who came afterwards.
When the translations fall short, it's time to redefine the words or look for alternate meanings that somehow prevent literal interpretation, yet such analysis only replaces one literal interpretation with another. So days turns into 2000 year periods, but only in this context, not in these other passages where they mean 24 hour periods.
No one suggests the obvious; it is poorly translated with respect to common English and should be fixed. That's because any religion worth its salt will quickly bombast any common tongue translation as non-traditional, misinterpreted, twisting the word of God, etc.
In other words, the circle of complaints is complete. They like it how it is since it provides plenty of opportunity to build cohesion by decrying everything else as foreign and wrong, and it provides plenty of opportunity to adapt to common situations by reinterpreting passages to mean things they likely never intended. Playing this game of contextually sensitive interpretation and redefining reminds me of the excellent film Alphaville.
For those who don't know the film. Words are defined in dictionaries that are obviously edited and constantly updated to limit the populace's ability to think outside of the accepted dogma. It's really a small part of the story, but a huge part of the ideas behind the film.
fire and flour can make a nasty explosive too, which one do you want to regulate?
I don't know where the "farmers like DST" meme can from, but it should go back there very quickly.
Farmers don't benefit from DST, because farmers are like construction workers; they work outside. That means that they maximize their use of daylight, not that the daylight gets shifted around to suit their needs.
When the sun rises, it rises the next day which is about 24 hours from the last time plus or minus some number of seconds depending on latitude. Setting your clocks to whatever time you wish won't alter this behaviour, and if you all agree that today you'll ignore an hour or some other day you'll duplicate an hour, then that's fine. However, it has nothing to do with the actual observations of the world you live in, which is the farmer's realm.
Farmers typically hate DST because it means they have to alter their schedule because the banks honour DST while the farmers honour the setting and rising of the sun.
Kill the farmer DST meme, it's misinformation at its finest.
I tried that, but I just wound up confused.
After following your instructions, I noticed that I noticed that my thumb and index finger formed a R, and I now have two right hands, with one right hand a bit more to the right than the other.
(sorry, couldn't resist)
Everybody likes a Robocop reference.
More weighing on my mind is the interaction between the following forces:
1. Reduction in manpower costs spur adoption by the military.
2. Reduction in manpower spurs the deployment of under supervised robot teams.
3. Trapping and reprogramming of robots becomes a more viable means of using the expensive hardware against their (former) owners.
4. Cheap ways of disabling these robots will provide a financial model of warfare that allows countries with weak budgets and manpower to do billions of dollars of damage.
The USA's military spending is about 25% greater than that of the entire world combined. http://www.globalsecurity.org/military/world/spending.htm Guess who's going to be losing these new, expensive weapons?
You state a falsehood.
We do know that CO2 traps radiation from reflecting back off the earth into space. Radiation is energy, and that energy trapped on the planet will eventually translate to heat (the apparent fate of all energy).
Now if the magnetosphere shifts, or if God's global warming alarm clock goes off, it's not going to make this little bit of physical phenomenon (CO2 traps radiation in the atmosphere) go away. So instead of brining a bunch of red herrings to the table, either prove that any of your other ideas have a bigger impact than CO2 in the atmosphere (which would expand the realm of findings to date) or address the problem of CO2 in the atmosphere (which all of the findings to date show is the main, but not sole, culprit).
We've already done geo-engineering by putting the greenhouse gasses into the atmosphere in the first place. It requires less creative engineering to stop putting them up there, and we know that greenhouse gasses from (whatever) source raise ambient temperature. Therefore, not putting greenhouse gasses in the atmosphere is a generally plausible solution, even if it means we have to change our lifestyle.
Stuff like sprinkling the ocean with reflective material doesn't have a very well known effect because we haven't tinkered with the planet in that way. I'm just a lowly ex-Biologist, but immediately after reading the description, alarm bells are going off like wild.
These particles will be exposed to one of the world's largest food chains, possibly poisoning one of the greatest stores of bio-mass in existence. Life will probably manage to struggle on, but even a reduction in bio-mass in the ocean has a very profound impact on the land dwelling population of the world.
We already have significant problems with mercury content of many types of edible marine life. They don't eat a lethal dose at any given time, but their bodies accumulate the poison until it presents problems for their predators. Such systems of poison storage causes collapses of the predators first, which then cause blooms of the prey, which then cause mass extinctions of the prey due to starvation. In this respect, animals are like humans, willing to watch the whole species go to hell in a hand bucket as long as they can exploit the environment for everything its got.
Even if they're plastic particles, plastics leech phenols which seem to cause some health problems. Even if they're 100% inert (perhaps ceramic?) small particles are deadly in their own right. Particular atmospheric pollution does it's damage whether you get it from living in a city or other means, some people can't get enough of particular pollution so they take up smoking ;)
I wonder if the researchers have considered how easy it would be to live, work, sleep, and eat in a house where every interior surface was covered with a fine layer of glitter.
Your humor is appreciated, but there's a large body of evidence that the best programmers can be more than 10x as effective than the rank-and-file, which can be (more than) 10x as effective as the bad programmers.
So a methodology that boosts output of two programmers 400% isn't really promising the impossible. Just consider that they are likely talking about average programmers, and the new environment keeps them engaged at an above-average attention level.
Peer pressure can make people do incredible things (good and bad). Altering the environment to make it more likely that good code is produced isn't snake oil, provided the results do follow. I've never seen a methodology that denies certain techniques are beneficial; instead they seem to argue over different combinations and inclusions of techniques that were observed to work.
Now, as you pointed out, the acronym spewing masses often don't know what they're saying. For that group, any methodology results in the same thing: changing the appearance of the current methodology without altering how the actual work is done. Months later, they have the perfect scapegoat: the methodology sucks.
Aqueous martinis are out, parsley soda is in!
As another poster already noted: Nobel prizes are for specific discoveries, not a person's reputation since the discovery.
In another reply mblase mentions Einstein. I'm going to do one better than that and mention Linus Pauling. I say one better, because Linus Pauling won two Nobel prizes, and he might still be the only person to have two Noble prizes to his name. His work laid the foundation for viewing shape as a critical element in determining the effects of molecules in living systems. He also won the Nobel Peace Prize.
The man was brilliant, I heard him talk before he passed, and he went to his grave trying to find the chemistry-level solution for the widely accepted idea that vitamin-C provides additional protection against the common cold. Detractors of his efforts joked about Pauling's vitamin C.
We don't want our best minds weighing the benefits of not working after a major discovery should they be in the running for a Nobel.
I can imagine a new type of gas station.
Basically, it's a retrofit of a parking garage. You're charging the entire time you are at work, minus the lunch break if you're inclined to take one.
No need to burgeon the best solution for the old problem (pumping gas) with the solution for the new problem. I mean, I can't buy hay at my local gas station either; perhaps that's because they didn't start out as barns.
Whenever you can't disprove something, make it encompass so many other items that it becomes incomprehensible without the services of a complete auditing team. Then extend it just a little to include data that you probably can never collect.
We saw this with TCO studies that tried their hardest to show that Linux might not be cheaper than $800 per seat licenses of Windows. While many people saved money by switching (Sherwin Williams, Fender Gituars, etc.) most complained about the intangible losses that they couldn't prove which justified their non-action.
We saw this with global warming. Most argued about specific point data that didn't follow the average (hint there's outliers in every interesting data set), argued that costs would never be calculable for the issue, and built a model of "let's wait and see" which justified their non-action.
We're seeing this with electric (or semi-electric) vehicles. It should be enough to note that we will mostly be charging the cars during non-peak hours, when the grid's capacity is most likely to meet the demand. It should be enough to note that for the same amount of used energy, the electric company can provide it to us cheaper than the gasoline distribution chain. It should be enough to note that at least a dozen ways in which the vehicles are cheaper to maintain offset the initial costs of not having a mass market's cost structure.
Instead, we have to calculate the exact dollar of every line man, telephone pole, coal miner's life insurance plan, etc, ad infinitum. I'll give you a hint: The power company already does this, and it's called your electric bill, which is still cheaper than your gasoline bill.
Just because you don't have it itemized doesn't mean it's more expensive.
For those that enjoy these sort of games, have you even considered the operational costs of the thousands of oil tankers? How about the costs of all those oil platforms? How about the costs of the fire policies on those platforms? etc... ad infinitum.
Prove to me that my electric bill will be higher than my gas bill, and I'll go with the parent poster's observation. It is the only sensible metric. And oddly enough, it should include all of those "other" costs, because if it didn't the company would be out-of-business before you know it.
Well, to start, can you reproduce the binary? If not, then you'll have to find those missing libraries, classes, and whatever else. In the event that simple searches don't turn up anything, reversing the binary might be your starting point. Remember, in your first pass you don't have to reverse for 100% comprehension, only for equivalent compilation.
From there it's just a lot of refactoring till you have code that looks something like the algorithm described in the documentation. Hopefully by then you can identify where the code deviates from the documentation and identify if that deviation is due to the incorrect model, or other factors that failed to get "updated" in the documentation.
Fixing by rewriting is almost always a doomed proposition for your case. It is rare that code follows the documentation 100% because life-learned lessons in the code don't get reflected in pre-existing documentation. It's easier to say "oh yeah, I forgot about that case, just fix it this way" then to pull out approved documents and go through another round of review to add a trivial (but critical) addendum.
Sure, this technique seems slower, but every step forward is measurable, and there are no surprises (you can always reproduce the binary as it already works). Rewriting gets you 80% done much faster, but it's hard to say if you'll finish it more quickly: there will be lots of opportunity for surprises with no guarantee of "current functionality" at any point in the game.
I don't think you expressed yourself clearly, or perhaps you meant to write something other than what you wrote.
"Testing cannot detect error with probability significantly greater than zero, unless the system under test is trivially small".
First, when a function doesn't work, it's not "probably not working" for a specific set of inputs, it's 100% not working for a specific set of inputs. 100% is significantly greater than zero.
Second, the whole concept of a unit test is to test something small, so even though your statement's first clause makes no sense, the second clause makes the first clause not applicable to unit testing.
Third, unit testing tests a single class's behaviour. Testing combinations of classes better falls under integration testing. There's no combinatorial explosion of testing when you test only one item.
Naturally to exploit this advance in software testing (and development) you have to buy into it's philosophy. If your code isn't very object-oriented (meaning that your classes don't have distinct and unshared responsibilities), then it will be nigh-useless to test if the unit (class) handles it's responsibilities.
To test a class, you have to write code. Once you write the code, it makes no sense to run it manually every time you wish to verify the class still meets it's responsibilities. So you create a pass / fail framework, and volia! You have unit testing.
It doesn't mean that other testing isn't important, and if you rearchitect the interfaces between your classes, you'll have to rearchitect the unit tests to match. But hey, there's no silver bullet that only shoots the work involved in changing the interfaces between classes except inaction.
PS. The least cost gun wouldn't come with a safety either, which would be unnecessary in a world where accidental discharges never occur. In the real world, guns get dropped. In the real software development process, changes get made without perfect knowledge of all they ways the class is currently used (or abused).
Don't advocate against rational low cost safety. The cost of unit tests get amortized over each regression test (which should be performed after each build), and are a lot less expensive than shipping buggy code to a client.
Nice comment, care to back it up?
Good tests are tests that verify the required functionality of the unit, bad tests are tests that verify things that cannot break or are the functional equivalent of the human appendix.
Exactly how would you preform good unit testing in such a way that you could change the code in a critical way and not have it break? I know it seems like an easy cop-out to say it, but if you could do such a thing, then your unit tests are incomplete as they don't test the required functionality.
That's not to say that you don't still have to do integration testing, acceptance testing, performance testing, usability testing, etc.
The only way a reader can read this book and not come away with a pocketful of gems is lack of experience or so much experience that the book bores him. Experience in this case isn't tied to years of programming, it's tied to having to fix truly horrid messes.
This is one of the best introductions into managing a code base that is foreign, hairy, and incomprehensible. I've been doing maintenance programming on and off for years now, and while it's not as glamorous as clean slate design, it's much more difficult.
Fowler's Refactoring gives you the transformation templates, just like Design Patterns gives you the code structure templates. This book gives you the instruction into code maintenance discipline. It's well worth a read, even if it leaves you wanting in a (very) few places. Don't think of it as "Refactoring for Dummies", it's more like "How to use Fowler's books in an enterprise and not go insane". Naturally there's a little duplication of information, but that's not the point of the message.
There seems to be a bit of an anti-testing crowd reacting to passages in the book. While I understand their gripes, let me assure you that either you're going to test your code or the customer will. If the customer finds too many, then you've squandered your good will in exchange for not writing tests. Not a good exchange in my opinion.
Another book along this line that I also highly recommend is "Software Exorcism" by Bill Blunden. Perhaps his quirky humour and old school usenet references will throw off a few acolytes, but there's plenty of treasure in there.
By the way, the bigger problem is with kids who do the work but don't think. I have lots of students who copy their friends' work, so they have great homework grades, but bomb tests because they have no clue what they're talking about.
How is copying their friends' work "doing the work?"
How can you admonish their lack of thinking when your own example has logic errors like this?
I don't buy the "pay the teachers" solution; it's too self-serving and often comes from those close to teaching.
I know teachers that are making pretty close to 50k a year. While that might not be too much in your corner of the U.S. of A., over here it's pretty good pay. Close to what a police officer makes, an more than the typical fireman.
This pay is up from the previously abysmal 28k a year about a decade ago, but the quality of eduction seems to have gone down. Note that I tutor, so I have been in indirect contact with teachers for years.
To illustrate an example, a pupil of mine had a math answer marked wrong because they considered in an inequality test the possibility that the extreme of the solution was in the solution set. Sure enough, it was and they were "deducted" for including the equals sign. This pupil was irate that they were wrong, and asked me to look at the problem.
Sure enough, the student was right. I asked if they talked to the teacher, and they did. The teacher had replied than in such circumstances they graded according to how the majority of the class interpreted the question. So since +50% of the class read the question to mean "only greater than", my pupil was wrong. I wrote a letter detailing how my student was correct (with a formal proof, hey it is math) and the teacher refused to even read the letter.
I'm sure there are still good teachers out there, but after the example of this one power monger.... I'd say let the lot of them starve. More money hasn't solved the problem, in fact it's nigh impossible for a good teacher to do better than a bad one provided they have the same time in service and are servicing the same grade in the same school district. The teacher's union has made sure of that in the three school districts that I've rubbed elbows with.
Old school solutions to this old school problem have existed for decades (and perhaps centuries).
A student that doesn't like their grade in a class can retake the class in lieu of an elective. The new grade replaces the old one, even if it is lower
This has been used with success at many an "upper crust" university for years. Students that don't / can't / won't apply themselves won't waste time retaking classes. Few people can afford the extra time to game the system for higher grades. In short, no new solution was needed.
So why did they come up with a completely new solution? Because now that the school doesn't have that small percentage of 0% students dragging them down, they probably appear to be a whole letter grade higher (on average) for their review boards. Odds are that other schools will either protest until this one changes policy, or quickly follow suit so they don't appear to be behind in comparison to the "new improvement" of a cooked statistic.
Considering what I fear, I'd wager on them matching the policy. That way they can (in their own minds) appear to have improved too! Don't tell them that when everybody improves equally then there's no change in the ranking. That would just spoil the game for everyone.
Considering that they were previously sliding by with 0%; how does sliding by with 50% motivate them?
I understand the "second start" idea; but in a university that means you have to drop out of school for around ten years and come back. What exactly in high school provided them with the new insight as to the importance of the grading game? Why should they get a "second start" in every class? Why should it be available witout the harsh life lessons that a decade of living with your mistakes can provide?
I would argue that gym should be pass / fail, with pass meaning you participated.
Either that, or you have to do some pretty stupid stuff to "level the playing field" for those that are physically gifted in the select sports that your gym has decided to test.
My brother completely rocked in ping pong back in high school (could play for days without dropping a single game), that should more than make up for his inability to hit a baseball, or his less than stellar basketball skills.
Better yet. Make the grading percentile distribution more like:
A - 100% - 81%
B - 80% - 61%
C - 60% - 41%
D - 40% - 21%
F - 20% - 0%
At least then they will have coherency between letter grade and percentile of accomplishment. With their current distribution, they have no coherency because a student that performs 50% is equal to one that performs nothing.
As far as the admitting colleges go, they will quickly draft their own plans to adjust for the new grading policy, probably relying even more so on the SAT and other measures to determine their admittance criteria. As far as the school is concerned they just doubled the number of "A" students, even if it was only done by lowering the bar for an A.
If what they were suggesting was padding everyone's score by 50 percentage points, then it would be fair (if awkward). Instead what they are suggesting is padding the worst performer's score by 50 percentage points. In statistics, this would be called "cooking the books", and I'll bet they're cooking the books for more than just "a second chance, whenever the student tries to take it". I'll bet that the new point system is presented to performance boards as equal to those school systems that let a student hit dead bottom zero.
If you want to provide a "second chance" to achieve, do what other institutions have done. Let the student take the course again, with the new grade replacing the old grade. It costs the student an elective and another four months of their life; that makes sure it won't be abused by the student body: time is precious. It maintains the current standard of the school because the course will likely be taught the same way.
What they are doing is unconscionable from a statistics point of view; basically they are taking the numbers they don't like and changing them to 50. The "average" will likewise jump (even thought no corresponding jump in work will be performed). Kudos for them on learning how to lie with statistics. Shame on them for doing it by substituting undesirable values with those more palatable.
If this is hacking, I guess shifting your car into 3rd gear is hacking the system to go a little faster than 2nd gear.
He used the system as it was designed. He didn't exploit any bug or software weakness of the system, the weakness was inherit in the design of the software; what yahoo considered a feature is really a steaming pile of problems-to-be-addressed-later.
Palin wasn't forced to use yahoo, nor did she do any kind of analysis of how suitable yahoo would be for her (admittedly illegal) endeavours. This means that she flunks on my tech ticket. Add to that the use of real world answers which can easily be googled, and it's obvious she's not even considered what security means.
It would have been trivial for her to put down a few dollars a month ($10/mo) and have a better (and supported) email account. Better yet, she could have her own email server set up in a private virtual machine ($20/mo + setup costs).
She should be used to having someone else do stuff for her, being a governor. Oh wait, that's it! She couldn't ask anyone else to help her because they'd likely report her ass for breaking the law!
I don't like what the guy did to get his info, but really this starts too look more like a botched attempt at whistleblowing than an act driven by criminal intent. I mean, where's the damage? He didn't delete anything, nor did he send emails posing as the governor. He even went out of his way to tell her that her account had been accessed by himself. She wasn't denied access to her account, because she could use the same stupid password reset web page he used!
Odds are this guy's going to have the book thrown at him hard. All over an act he performed that didn't cost Palin a dollar and exposed that she was consistently and deliberately violating her own state's law.
Every bit of GPL software is licensed. If your and end user, then it's licensed for you. To use it, you have an End User License Agreement.
The Gentoo bunch have their point, but really they're making a mountain out of a molehill. They should rebrand Firefox and forget it. They shouldn't be moaning about having to display a licence on startup, which is exactly what they don't do for inkscape, ls, evince, cp, evolution, rm, et. al.
If you don't like the idea that someone would trademark their project name, then fine. Remove the Linux kernel as that's trademarked. Or, better yet, please download the newest copy of firefox at http://www.firefax.com/ Yep, that's firefax, the backup domain (to offload the main one). No, no, no!!! We don't embed spyware into OUR browers, cause it's called Firefox! It is (almost) the same one you've been using for years!!! MUAHAHAHAHAHA!!!!
Now, you can still be upset about it, but hopefully you can see a reason or two to enforce a trademark, even when your main product is open source.