There are two kinds of people in the world: those who know where they are going, and those who don't. The second group of people are easily engaged by the first oasis of flashing lights and bright sensations they stumble upon, and they probably bumble through a few "pay per click" turnstiles in the proces.
When I seriously looking for something, I can almost always tell from the result page synopsis which sites have rank they didn't deserve, and pass over these quickly.
Right now the ones that are killing me are the SpringerLink spams that seem to offer direct access to a full-text PDF, but you actually wind up at a coin-op with nothing more than an abstract. There are other research journal sites (I don't have the culprits at the tip of my tongue) where you can get to a page of full-text from the Google search results, but if you try to navigate to any other page once you reach that site, it demands a registration/login. But you can go back to Google and get the next page with a clever search. Call it the Elsevier bait and switch. If they didn't invent this, they probably wish they had.
That's a very strange optimization. A site makes its full-text content available to Google to attract people from the outside, but you have to pay a price to navigate the facility from the inside. Imagine a mall where every store is open to the street, but to access the arcade that connects them internally, you have to buy a membership.
I hate Google for putting [PDF] beside links to these nasty academic journal coin-ops. Like I'm going to pay $90 for a fifteen year old paper on some obscure protocol demonstrating a potential drug side effect in rats. There was a new result published this summer on "fast acting" antidepressants acting as 5-HT4 agonists. Turns out that 5-HT4 agonists are also the primary target of a class of drugs promoting GI motility in IBS patients, among which was a drug named Propulsid, later withdrawn due to possible cardiac arrhythmias. That gives the class of "fast acting" antidepressants several different spins. s/depressed/scared witless/
Google has become extremely lax about allowing the likes of JSTOR and Esevier to either mutate their presentation on referrer tag, or report something different to the Googlebot that what the end user sees. In my opinion, these bastions of intellectual shakedown are a little too rich for Google not to notice, or not to care.
Even if I wished to purchase a fifteen-year-old five page article for $90, I certainly wouldn't do so without first exhausting the information I can freely obtain. I can't count the number of times that JSTOR has tried to hit me up for a paper that the author has published freely on his own web site, which I locate a half dozen clicks further down the road. Yet Google is apparently making no effort to return the free full-text version ahead of the abstract behind lexan edition.
So far, Google has self-policed well enough that a Google-result-fixer has not yet become the number one Firefox plug-in, but I can see that day coming soon. It sure would be nice to have a plug-in to automatically substitute any coin-op JSTOR/Esevier link in the Google result set with a link to the free full-text, when that also exists just for the looking. My pain level has not quite reached the point of rolling my own.
I once posted a comment that what Slashdot needs is a way to pre-moderate the foregone contribution. OTOF, I suppose it was a lot more fun for everyone involved to race off to Mordor to destroy the ring *after* the ring woke up.
I term this reverse confirmation bias: if many people have tried and failed, it must be impossible.
But what credit is there to that? Many were the claims to transmute lead into gold. What proved impossible by chemical means was by no means impossible within the framework of the right technology. I think you need to study the "Four Colour Corollary". This theorem states that the truth or falsity of the theorem is entirely independent of the number of bozos who publish unfounded and incorrect speculations disguised as purported proofs. Furthermore, we still don't have a proof that could possibly have been discovered before the computer era, so the deck was stacked towards impossible... until it wasn't.
The same thing happened within the field of AI. This still annoys me. A lot of grand claims were put forward in the 1960s, and it all fell far short of what was promised. Nevertheless, there has been an unbroken stream of solid and important results, if not yet worth writing home about. Weren't the smart people silently expecting it to play out this way all along?
Guess what? Computers are now checking computerized proofs. Does this series converge, or not?
As for this new blood test, the human genome was sequenced a scant seven years ago, the explosive shock wave of proteomics is expanding almost at the limiting wave velocity, and we are now beginning to disentagle some of the fundamental neurochemistry involved. If there are any correlates in the blood whatsoever, it would be shocking to not find them at the present time, or in very short order.
Concerning percentage prediction rates, have we learned nothing? If you have a population of size N which you wish to classify into two distinct groups, given prior p and (1-p), the information required to achieve this is N * H(p), using Shannon's information measure. If this test provides any additional information beyond the prior, one can formally determine the ratio of the unknown information this test provides. If the test is worthless, the ratio will be zero. If the test is perfect, the ratio will be one. If the ratio comes out negative, you just assume the water goes the other direction (by metaphor with electrochemistry), and substitute the absolute value.
The interesting term is the cross entropy between what the experts can determine and what this test can determine. If the cross entropy is 100%, then either test gets you to exactly the same place, and it will probable come down to a matter of economics, which the cheaper approach prevailing. If the cross entropy is significantly less than 100%, then one will likely employ both tests, possibly using the cheaper test to screen the more expensive test, depending on tolerance rates for false negatives and false positives.
Given that they have included 18 elements in this test given a small positive sample size (they don't state their negative sample size), it's almost certain that some of these 18 factors are bogus, and will be eliminated as the sample size increases. If this test is bogus, the factors remaining will dwindle to zero, as the predictive rate also dwindles to nothingness. If the test is fundamentally predictive (to some ratio of the information content) as the bogus factors are pared out, the predictive ratio will likely improve by some marginal amount, maybe enough to be worth doing, maybe not.
In the 1970s one could make easy sport of predicting that any given claimant of the "four colour proof" was wrong and pat yourself on the back for an unbroken chain of confirmations. Great work: you've managed to predict that the world is full of de
I've made substantial contributions on a variety of fronts, both technical and idiosyncratic, and I would say 75% of my contributions remain largely unscathed.
I think there is a need to delete stuff, but they go about it the wrong way. There needs to be a place you can go to find out what was previously deleted, so you don't make the mistake of adding it again, and then having it immediately deleted again. It's a preposterous system as it now stands. Every article needs to maintain a permanent record of material decreed non-notable, or not yet sufficiently notable. A lot of the pre-babble could accumulate there.
This material would not be searched by search engines, or by default from the Wikipedia search bar, but it needs to be searchable by anyone desiring to contribute material not found on the "front page" of an article.
I have to wonder if this downtick in Wikipedia activity corresponds with the end of the Sopranos. It could be just the pop-culture where there isn't much left to add.
For technical content, every third article I visit strikes me as having major holes, usually because the article was written from a single vantage point. For example, Wikipedia contains extensive information on prescription medications, but the vast majority of the pages don't provide even the date of discovery or commercial introduction. Historical information is far too thin in most articles. I also think that additional intellectual property status should be incorporated in the majority of pages where the subject matter intersects with commercial interests.
Most of my contributions tend to be better referenced than the existing text of the article I modify. Sometimes I wonder if anyone ever checks those references. Of the material that I've had immediately deleted, there have been two situations. The first is that the page is "owned" by someone who thinks the page should be a certain way. With those people, you often find yourself in and edit conflict within the first half hour. Early on I learned to pack up my bags and edit elsewhere at the first sign of territoriality.
The more interesting case is where I go to the trouble to improve the exposition so that someone with less expertise in the subject area can put things into a proper perspective. This is similar to the problem with obviousness in the patent system. Some things have no documented prior art, because everyone in the field considers the matter beneath mention. My querulous other experienced a similar frustration doing a research project on government resources in the area of ecology and the environment. Once a program runs out of money, the government web site becomes instantly frozen in electronic amber, with no edit since the previously elected government. If you are in the industry (of helping the government waste taxpayer money), you know which programs have lived and died. If you are a random citizen searching with Google, you don't, and you waste a lot of time navigating the deadwood before the penny drops.
There are plenty of Wikipedia pages with a lot of formal content about computer buzzwords and architectural classifications from past eras that have entirely outlived their usefulness now that computers are 1000 times faster, give or take a few Cheerios, but the article fails to point this out. Show up at a job interview within the IT industry, you would come across as a clueless dust bunny spouting some of that terminology. Yet I've had several efforts to add text to the effect that the industry no longer thinks in these terms instantly deleted as unspeakable POV. But hey, the patent office granted the "one click" patent because apparently no one got around to writing down that this was bloody obvious, so why should the average Wikipedian be any smarter?
One thing people consistently fail to recognize is that the growth of content has vastly outstripped the ability of the MediaWiki software to evolve in pace. Many of the political problems that now exist in Wikipedia c
She did things that were actively counter to the culture at HP. Possible because she got options based on when she left, maybe she did it on purpose for personal gain. Then what you are saying is the shareholders/board aligned her incentive package to encourage those behaviours. Perhaps the board's strategy in appointing Carly to oversee these disagreeable actions is that she makes such an attractive effigy. Amazing. From what I've read in this discussion, not a single person remaining at HP (executive, shareholder, board member) after Carly was forcibly strapped into an ejection seat has been named as bearing any portion of responsiblity for what happened there under her watch.
Everyone is busy documenting how polarizing Carly was during her time at HP and Lucent, and then wondering why she's made the jump to FOX.
The fact is, the same emotional reflex that personalizes all the changes at HP as being the result of one person will drive people to watch all the polarizing drivel on FOX.
Recently physiologists have uncovered the hormonal basis for such behavior. Animals and people subjected to attack or threat experience "subordination stress," as a result of which their adrenal hormones go up, along with blood pressure and the probability of developing ulcers. But -- and this is crucial -- when given the opportunity to "take it out" on someone else, victims show no sign of stress. By passing along their pain, they modulate their own internal distress while generating trouble for the next ones down the line.
When an individual suffers pain, he most often responds by passing it on to someone else. When possible, that "someone else" is the perpetrator, the original source of the pain. But if this cannot be achieved, then others are liable to be victimized, regardless of innocence.
People need to think hard about whether Carly is a cause or a symptome of an unpleasant economic reality, and then consider directing their emotional reponse at an appropriate target. Or not. Perhaps we could tweak the moderation system to assign points for "ulcer relief" when the culpability of the target is dubious or irrelevent. I suspect some people would configure "ulcer relief" as a bonus score, because any victim is better than no victim. Even better if the "victim" is rich and famous, because then we can all deny that this is what we are engaged in doing.
One would think it would be better for slashdot discussion if TFA was not a moving target. To think slashdot is ten years old. That's one hell of a slow clue train.
Here's where we finally get to the nub of the matter. What it really depends upon is your definition of "get it working 100%". By several of my criteria, gettings Windows working 100% is an unachievable task. There are some things I don't much like about Linux, either, but even the worst of these could be fixed on the order of a few tens of man years. I happen to rate a few dozen man years as less than infinity, the expected cost of addressing my worst grievances with Windows.
My choice to view several dozen as being less than infinity might seem obvious, but in fact, it is not the popular perspective. If you read thinkers such as Danny Kanheman you will recognize that for the most part people don't think the way they claim to think. By the reflex of learned helplessness, people tend to discount the impossible, exactly as my parent poster has done. Subtracting the impossible, one can get Windows working 100% in a fairly short time period, with respect to a learned helplessness definition of 100%.
Learned helplessness wouldn't be so deeply embedded into the human psyche if it wasn't pragmatic.
It's a fairly substantial investment of time, energy, and talent to buck a mainstream trend. For any professional, I think you can only open so many fronts. My LH relative to IT is quad-CT to zero (that's an APL joke, to thoroughly date myself). On the income tax front, my LH would be closer to 7/10. I'm not motivated to win every possible battle. The last thing any nation wants is legions of empowered individuals, so the barriers are substantial.
The general public tends to constitute 100% largely in terms of instant gratifications: can I watch the newest YouTube video straight out of the box? Terms such as "will I still be able to access my personal data ten years from now after all my current software is obsolete?" rarely carries as much weight.
Nor do people stop to think much about why it is that media formats are directly tied to running specific operating systems, as if OS capabilities has much to do with it.
The other point to note is that engaging in LH has a tendency to also invoke the psyche's PR department, which isn't keen to admit any such thing, so people who have the deepest investment in the pragmatism of LH have the strongest rhetorical reflex to promote their choices as "the one true way".
Apple has historically been very good at exploiting this reflex. They do a great job of enhancing the pragmatic value of LH, and correspondingly their infinities are more infinite than most. With the brutal cooperative multitasking and virtual memory subsystem, no Apple OS prior to OS/X was within orbital radius of "100%" by any criteria I've ever accepted. The LH retort: well, you don't need that. But this PR philosophy leads Apple to more truly embarrassing reversals than most, such as their recent concession that the technical advantages of RISC over CISC in the era of 100 million transistor CPUs are commercially negligible.
One of the main terms that holds Linux back is the instant gratification bondage. Full technical disclosure of video card internals would constitute one large step toward playing iNextSonyGoobTube videos right out of the box. If the college age demographic would simply refute their instant gratification ways, and refuse to view any video encoded in proprietary media encodings, this battle could be won in less time than a Peter Jackson post-production cycle. But it will never happen. Public empowerment? Who needs that? Maybe 5% of college age people include public empowerment in their personal definition of "100%".
BTW, I'm quite conscious that posting on slashdot values my time at $0. It's less of a detriment than it might appear.
I had an idea for a major improvement to slashcode, and I've been waiting for my best shot. This is the perfect forum.
The problem with the moderation system is that you have to wait for the post before you can moderate it. This is a serious design mistake. It's quite obvious with a story like this one. An enterprising moderator could have moderated half the jokes here "-1 obligatory" *before* the jokes were posted.
This single feature would go a long way toward rebalancing the force. The chuckleheads could continue their race to first post the obvious lines, while the chucklehead CDC could escalate their counterespionage in lockstep to strangle as many of the chucklehead jokes as possible in the interval between when the blowhard lemmings depart the cliff and when they impact the ground with the inevitable dull thud of exploding whale funny bones.
What exactly is modal about sizing history to available resources? It's been an accepted human practice since the invention of the palimpsest.
If you want a non-modal design, it's an easy constraint to satisfy: have Photoshop maintain a complete undo history to the beginning, all of the time. What could be less modal than that? It might thrash like a pig on 90% of projects within 30 minutes of enaging in any serious work, but it won't be modal.
Responding to another post, if the average design artist doesn't wish to wade into discussions about technical tweakage, then that's the way Photoshop should be configured by default. Why are the software people busy implementing arbitrary tradeoffs, such as a twenty step limit on the undo history, when the artists themselves regard this as beneath their notice?
Let's get serious about non-modal design. No arbitrary limits on undo history, or internet disk cache, or colour palette depth, or anything at all that the user doesn't explicitly enable themselves. If the user complains about sluggish performance we can explain that the UI design is the least modal design possible, is full compliance with the mother of all GUI manifestos concerning ease of use. However, being open minded as we are, we also added some user controls to activate modal parameters, for those users who don't find it beneath their diginity to adjust them.
It's worse than that. The Nash equilibrium poker strategy is one that can be beaten, but in order to play this strategy, you must never take advantage of your opponent's mistakes. As soon as you adapt to your opponent's mistakes, your strategy is no longer certain to be unbeatable.
Secondly, any use of random numbers creates the possibility of an insider using cooked randomness to establish a covert channel. You want to absolutely certain that your randomness is provably random, and not cooked by anyone involved. It's almost impossible to distinguish cooked randomness from true randomness without advance knowledge of the procedure that uncooks the randomness.
It wasn't long ago that the Mickey Mouse Protection Act extended copyright term by 20 years. Meanwhile, what exactly are they doing about the tens of millions of bot infested PCs, of which the vast vast majority run Windows operating systems, and which increasingly constitutes a threat to national security?
In all my life I've never seen a scientific study about what kind of content has the potential to harm children and why. I'm sure most of my adult peers managed to expose themselves to harmful content as children. Only the least enterprising children fail to accomplish this. And what is the end result? We're all convinced we came out fine, by the skin of our teeth, but the next child won't? What exactly was impared? Our gullibility? Our willingness to vote morons into power?
Obviously there are some children who are adversely affected by coverage of the real world on the six o'clock news. But I have a feeling this bill is not targetted at that content.
He obviously hasn't read his ISO Big Bang 1.0 spec. closely enough. The language reads "quantum mechanics behaves *as if* it bifurcates into parallel universes". The actual mechanism is implementation defined. Consult your documentation on Local Universe for more details.
I've stated this humorously, but I mean it quite seriously. The other universes don't necessarily exist merely because your local universe behaves as if they did. To a certain level of accuracy, the orbits of the bodies in the solar system behave as if epicycles exist. What did that prove?
I do not welcome our FUD spewing overlords, either, but I wasn't expecting to find them employed at Slashdot earning troll bonuses at the expense of public perception.
When contention reaches critical mass and a project forks, it often benefits both groups who are now free to do their own thing; or, one camp fails to keep pace, and the dispute is effectively resolved. It didn't take long for pf taking the more aggressive road to establish itself over Darren Reed's IP Filter.
If this unhappy camp feels up to the challenge of forking the Linux kernel and taking a more aggressive approach toward performance tuning, and they can produce results in the near term that make Linus and his established approach look old and tired, then I applaud their initiative. Chances are, however, it won't be as easy to pull off as the end run around IP Filter.
The main risk to Linux as it now stands is failure to give this initiative a new name. Maybe they should call it Smoothy. I look forward to future reports of the amazingly stable frame rate in the next release of GNU/Smoothy which builds upon the newly released AMD/ATI video specifications, while all the greybeards among us cluck and moan. The one thing I don't want to see is teenagers texting Smoothy complaints about stuttering artifacts in their VR sex appliances to LKML. In other words, be off with you, best of luck, and good riddance. If you can make it work, more power to you.
In human history, the young and insane often prosper. Half of the world's major cities are built on flood plains yet civilization hasn't ended.
I stumbled across an appropriate quote by Henri Poincaré last night.
To doubt everything or to believe everything are two equally convenient solutions; both dispense with the necessity of reflection.
My view of C++ is that it invokes the necessity of reflection more than any other. The people who hate C++ the most are the same people most invested in convenient solutions, for any x, where convenience == x. For almost any x, C++ is terribly inconvenient language. Which is precisely the reaons that C++ excels at solving terribly inconvenient problems, or it would have no reason to exist in the first place.
Due to the exponential nature of computing power increases, at any given time about one third of the surface area of the problems people are most highly motivated to solve are terribly inconvenient problems: the original open-ended evolution of HTML, the recent transition to Web 2.0, scaling a search engine to 500,000 compute nodes, navigating a Mars rover.
Resource management has earned a reputation as being a hard problem only because so many people are bad at it. If you can't correctly manage memory in C++, it's pretty much guaranteed you can't write a sychronization correct multithreaded application.
Most of us have learned to think in terms of preconditions, post-conditions, and invariants for coding a correct iteration that terminates. It's no different to correctly specify object lifetimes in a code base lacking automatic GC. The object comes into existence under this condition, while it exists this invariant is true, when the invariant fails, the object is disposed, and the time between the last two steps is as brief as can be practically arranged (ideally, the distance between two successive program statements).
I suspect Firefox leaks so much because the coders have descended to mechanistic thinking: to do this task we need to allocate this object, later when this mechanism is activated, if it appears correct to do so, the object is disposed. Sweet. We completely bypassed the inconvenient step of specifying a rigorous object lifetime invariant expression. Hurray! The project will ship on schedule.
By contrast, the Java programmer is spared any guilt over bypassing this difficult logical step. Only the GC doesn't entirely save you. If by a logical error, you leave a reference to an object intact that is in fact no longer needed, the GC will never free this object. With any luck, the object maintaining this unnecessary reference will get cleaned up later on, and by implication also its orphans. And there won't be much surface evidence of this missed logical step either, aside from the usual observation that the Java app. is generally consuming far more memory than you would like to believe.
One C++ project I happen to like is the monotone SCM. When you set out to write a rigorous SCM tool, you are committing yourself to a life time of high calorie lunches at the reflection buffet. At this point, C++ is the least of your problems. In fact, I always gravitate toward projects where convenience for its own sake is shuffled out the door on day one. Do you think Google built their massively parallel computational architecture by showing up every day asking themselves first of all "what is most convenient?" No, a challenge on that scale demands a renewed investment in hard thinking after every meal.
I would learn Haskell for the inconvenience it offers, but I already enjoy 90% of the inconvenience that functional programming has to offer by grappling with C++ template instantiations.
To paraphrase Einstein--for those of you who find it tedious and inconvenient to wade through a sustained meditation--your programming language should be as convenient as possible, but never more convenient that the problem your program is attempting to solve.
To update Knuth: premature convenience is the root of all evil.
How about we settle for getting the linguistics right, and ignore history? Who put the idea forward that people gravitate towards a name for the purpose of ego equity? That's not how the human mind works. People tend to fixate on affordance distinctions. A person who is trying to decide which of his favoriate applications will run on a system, at that point in time, probably could infer more from the Linux/BSD distinction than whether the userland was GNU or proprietary. As some posts have pointed out, many proprietary systems had GNU userland tools available. In some cases, such as writing complex shell scripts, one might be primarily concerned about whether the target userland was GNU or not-GNU, but I don't think it was the overriding consideration most of the time.
Secondly, the name of the distro tells you a lot about how the system is managed, and whether you have the right sysadmin toolset: packages/yum/apt, etc. For that matter, I don't recall ever installing a GNU file system. You choice of file system was usually dictated by your distribution, not your userland tools. In fact your distribution goes a long way to determining your politics as well: under Debian certain things would be missing or difficult that would be defaults elsewhere. I don't mind calling Debian GNU/Linux, because Debian explicitly bundles much of the FSF politics. I never regarded the GNU in Debian GNU/Linux as a reference to the GNU userland tools.
I think it would be good if OpenBSD had its own C compiler, even if it only generates reasonable code for x86 architectures. A C compiler aimed more at correctness than performance is not complicated unless your code generator needs to support every known platform. It wouldn't mean that gcc needs to be phased out, it could be used on a complementary basis, especially if it has more rigorous portability and error reporting, suitable for compiling a kernel, if not for every imaginable userland program, or the monstrosity known as C++, which is my preferred language for statistical NLP, but would be an abomination for kernel purposes.
There are many balanced people who accomplish a lot in the world, but you fail to recognize the difference between self determination and self discovery. Is it really true that a person can decide to follow either path (a pure choice) without sacrificing any golden geese either way? Or is it more the case that you figure out that your obsessive fascination with an intellectual problem brings something to the table that an overt attempt to style yourself balanced would sacrifice or destroy?
I think the key perspective on this question is that the majority of deep discoveries in mathematics have come from individuals under the age of twenty-five, even if some of them were published later. There are a few exceptions of course, but much of the history of mathematics was paved by young males (mostly) in their youngest, most obsessive years. Obsessive energy is part of the human condition when we peak as young adults. Especially for males, who face a higher risk of not mating at all.
Later on when we are old and slow and stupid, obsessive behaviour has fewer rewards, while balanced behaviour has more. I had all my best ideas in my twenties. Fortunately, I kept a couple of those ideas to myself, so I can still pretend to be obsessed about something I'm busy inventing, but it's just the geek equivalent of buying a convertible when your hair turns grey.
I've known a number of males whose obsessiveness surpassed any concern over group belonging. I haven't met nearly so many women who regard inclusiveness as negotiable.
I don't think the issue is that male engineers have something that women engineers are denied. I think women wish to enter engineering on terms males never demanded. That's probably a good thing. The great advances in future society are more likely to come from groups than individuals.
I also think that women sometimes set their sights too high. If there has ever been an MRI scan taken of a geek coder in the moment of supreme creation (at the keyboard), I don't think you'd see a lot of blood flow to the chit-chat circuits. I do math rarely, but every time I have a math problem to solve, so much of my brain is recruited, I become inert to my surroundings. Any culture where that kind of departure from life is part of the discipline, is going to develop differently, and maybe in ways that women don't enjoy as much, above and beyond boys being boys, and jerks being jerks.
What I'm suggusting is that if you could magically take geek engineering culture and strip away the discriminatory aspects, I believe there are many women who would continue to complain about the end result as an unfriendly work environment. If you don't have some hard edges in engineering culture, you get the NASA effect, where too many people are checking out the social and political consensus, and not enough people are saying, "uh, excuse me, you can't do that that way" with least regard for affiliation. And not just one guy at the bottom of the chain of command.
Microsoft innovated new business practices, such as submitting simulated video evidence in major court cases without originally designating it as such.
People make too much noise about technical innovation. In the vast majority of cases, 90% of the difficulty is in running a successful company in the first place. Make that 99% if the company employs too many Linux hackers.
Dell's primary innovation was in its distribution model, which suited white box systems for about a decade, until the price came down and the market moved toward laptops, which people expect to touch before purchasing. I respect true business innovation as much as I respect true technical innovation. However, when it comes to business practice, there is a fine line between innovative and sharp. Selling printers with the starter cartridge half full of ink? That's a sharp business practice, not an innovation. Enron's mark-to-market accounting methods? I don't even need to comment on that one.
It would be tough to speculate what proportion of technical innovation in the world these days is embodied as software. It's also unclear whether the global innovation rate in the modern world depends in any way upon patent protection. For every person who says she wouldn't have bothered without patent protection, you can probably find another two to say they would have gone further without the risk of patent interference.
The case where patents seem most important is where the resulting business model requires a large initial capitalisation. I doubt the patent system has any connection with innovation rate, but I suspect it does have a connection in the level of investment available to make those innovations commercially viable.
Perhaps what we need is a two term system. For the initial term, perhaps six years from date of filing, the present system would work, with some of the more glaring stupidities cleared up. The renewal application would be a contested process, in which the innovator needs to justify considerable non-triviality while facing a very broad interpretation of any prior art brought to the attention of the patent office.
Such a system might cut the patent application rate in half. Of those approved, maybe 20% pursue the term extension, and that process costs five to ten times as much as the initial application, because these examiners are supposed to know what they are doing, and they have contesting materials to arbitrate.
Net effect is that the cost of the patent system remains about the same, the number of worthless patents declines, and many of the marginal patents are washed off the books at the conclusion of a relatively short six year initial term.
I don't think we can eliminate software patents altogether, it's far too large a sphere of human intellectual work.
Identity theft is a very small problem. Credential bungling is a huge problem, and it is caused by the institutional structure that this defamation-exemption created. How did that get passed into law?
It strikes me as non-constitutional. How is anyone to succeed in the pursuit of happiness when major credit agencies are divulging damaging falsehoods about you in your significant financial relationships? Truly, in American society the pursuit of happiness must translate to litigating these credential-bungling credit rating agencies all to hell.
I also think the law should support a class action law suit over the design of the credit card security mechanism which created this problem in the first place, towards appealing the common notion that their credential bungling is our identity theft.
One post suggested that by enabled me to obtain better interest rates, I benefit from having my (good) credit history collected and reported.
If that is the case, a capitalist society would enable me to hire third parties to gather and report this information. If I wish this service because it enables me to obtain favourable terms, then I seek it out and I pay for it myself.
I have still have that Nixon tape running through my head after viewing Sicko the other night. Note to the Moore haters: let's ignore the messenger, and let Nixon speak for himself, in his own words, brought to use by the magic of magnetic tape:
For private enterprise, it has worked remarkably well. As Nixon counsel, John Erlichman explains to the then-President,
Edgar Kaiser is running this Permanente deal for profit. And the reason he can do it... is all the incentives are toward less medical care, because the less care they give 'em, the more money they make.... The incentives run the right way.
Nixon: Not bad.
The next day, Nixon announces the establishment of HMOs, saying, "I want America to have the finest health care in the world. And I want every American to have that care when he needs it."
The credit rating agencies enjoy the same favorable legislative tilt, with the same detrimental consequences to the typical American: instead of litigating against credential bungling at the hands of the Enrons of personal record handling, we cower in fear over this ficticious concept of "identity theft" which merely a synonym for the non-constitutional legislative contrivance that "if corporations lie about you, you have no power to stop it".
The real point here is that peer review is not the fine filter it is portrayed as being. The end result of the process might be science, but much of the workings along the way are sub-standard.
The people who wish to replace Wikipedia with a different system of "expert review" are just replacing one flawed group of people with another flawed group of people, only this new flawed group of people has manicured egos, and they have all made a pact with a social system which reasons "no need to point out every flaw in the papers I review, do I wish to draw the same scrutiny to my own papers while my tenure track remains uncertain?" So they all sit around pretending their shit doesn't smell, and hope that the rest of us fall for the ruse.
I detest the invisible nature of peer review. The person doing the review is not accountable and doesn't leave a written audit trail. How is that acceptable to Wikipedia, but the standards of its own community?
If a scientist's own work is discredited, does that discredit any papers he might have reviewed? Fat chance that review is repeated, those papers are already on the books. The only offense that gets formally repealed is plagiarism of one published paper by another published paper, because that is an offense of the record against itself. Failures of the review process itself never leads to formal repeal. Or do people here know otherwise?
It's not that I think the current process in Wikipedia works terribly well, but at least it has the potential to evolve into something we haven't tried yet, that could serve as a useful counterbalance to the multitudinous small offenses that overworked professors are almost obliged to pull off in the cut-throat academic world of publish or perish.
We only have to tear the page out of Freakanomics concerning his analysis of Sumo politics to see what happens when the inmates run the asylum.
America is no better. Witness NBA handling of the corrupt referee Robert Hoyzer. The only principled solution is to ban betting against the spread. A corrupt referee who changes the *outcome* of games will find himself subject to merciless scrutiny among the players themselves, who have a huge vested stake in that outcome. Gambling on the outcome of sporting events can't be manipulated half so easily as the spread. Any punter who bets on the outcome of a meaningless game at the end of a season deserves what he gets.
The kind of person who accepts betting against the spread as something that is adequately policed is the same kind of person likely to argue that the peer review system works well enough. I don't agree in either case.
On a tangent the size of Jupiter, I once asked myself the question, not in the spirit of sci-fi, but in the spirit of someone who might set out to write some sci-fi (which is not my cup of tea), that if we regard the hacking off of limbs in the American civil war as medically barbaric, what will appear barbaric about our generation to the generations of the future?
To be brutally honest, most people are not equiped to pose this question, much less answer it. If you've seen those end of the year lists, you know what I mean. To begin with, the answer is never the same. We don't need to invent the theory of pathogenesis twice. Nor is it as literal as replacing a D-saw with a fancy collection of salt shakers.
My suspicion is that 100 years from now, society will look back at the 20'th century and gasp over our practice of approving medications on the basis of population studies.
Not only will it be regarded as scientifically suspect, even barbaric in the exposure of the general population to poorly understood chemical agents, but also as a great industry of hucksterism. Take a therapeutic substance that achieves a large effect for a small number of individuals who need precisely that treatment, conduct a large population study with includes a small group of these people, th
Your conceit is slimier than you are making it appear. You are leaving out the step where the duality of the license is conveyed by the original developer. Or are you just assuming that if you come across any source code with both the BSD and GPL licenses pasted inside that this implies a dual license? Presumably you need a statement from the original author, preferably in writing, even better written into the source file itself, that the original author intends the two license texts to function under a dual-license either/or interpretation.
The only way copyright for a work can be verified by a recipient is when the derivation chain can be reconstructed through the derived work(s) back to the original work. If a person goes out of his/her way within the letter of the law to muddy that trail, I would regard the actions of that person as quite vile.
Perhaps you can, as you suggest, legally strip out one of the licenses (and the text granting the dual license, if that was present within the source file). But then if you distribute your derived work, you are handing people half a proof. Sensible projects would not accept the work in that form. They would want to see the derivation chain before taking that risk.
You seem to be suggesting by implication that the derivation chain can be maintained outside the text of the source file by some auxilliary shuffleupagus which most users of the source are unlikely to chance upon. If that wasn't your purpose, why delete the in-place full disclosure of the dual-license status of the original work?
In my view, there is no ethical way to substitute for the attribution "this is a derived work of a work by Bob Dawg who granted permission to do so under a dual-license" to "this is a derived work of an original work by Bob Dawg [and good luck figuring out the terms of his original license grant]" and by grant I mean his *statement* that permission was granted under dual-license terms, not the mere inclusion of two different licenses into the top of a source file, which in the absence of a clear statement, I would regard as ambiguous.
I see no problem with writing "This a derived work of an original work by Old Author dual-licensed under the GPL and BSD licenses. The original work is available at URL. New Author has elected to make changes under the GPL, which is reproduced below."
My point is not about trimming out the text of the BSD license, it's about muddying the terms of the original grant through crafty omission in the wording of your own grant, to the detriment of any future party who comes into contact with the derived work.
Your use of the word legal word "requires" (in the selfish conjugation: that I myself can not be sued by any party) neglects getting along nicely with any other parties to the overall process, upstream or downstream.
I could be wrong. The mere inclusion of two license texts could amount to an implicit grant of dual-license either/or status. So what? The only thing standing upon the letter of the law achieves is breeding more lawyers. Then you have to pay them a lot of money to stop breeding. With luck, and patience, and buckets of cash the legal bugs are resolved. Original authors, if they have any time left to code, learn to paste a more explicit legal text into their document that accomplishes the same thing legally at great cost that common sense and good will could have accomplished in the first place.
Only a dimwit or a lawyer advocates a move in the game tree where the only consequence of the end game is that the lawyers get rich while the loophole is closed. That is after all an explicit function of the legal system: to be so damn expensive and unproductive that it scares us into applying common sense against natural inclinations such as the one you expressed.
Refering to it as the "BitKeeper fiasco" merely serves to reveal your biases. At the time Linus chose to adopt BitKeeper, Linus had become a choke point on patch integration. He claimed the use of BitKeeper greatly improved his efficiency and coordination.
As I understand it, a lot of work was going on behind the scenes in parallel with the adoption of BitKeeper to improve the social dynamics of patch flow among the kernel hacker cabal. Yes, there was a loud and sustained polemical outcry, but show me solid evidence that kernel evolution, growth, maturity, or stability was compromised during this period as a result of all that background noise. Background noise is a fact of life in most open source projects.
When McVoy finally did have his predicted rug-pulling two-year-old temper tantrum, Linus had come to a good understanding of exactly what he needed in a collaborative source code control system and git was not long in arriving. Would it have been possible to implement git correctly three years earlier without the basis of that experience? Would the social systems have been ready for it? These are important questions the BitKeeper yowlers are rarely seen to seriously consider.
What fiasco? Perhaps tempest in a teapot is a better description. If a crappy SCM had been adopted instead of BitKeeper, Linux would have had two camps of yowlers: the hardest working cogs at the center whose work process integrating patches was poorly supported, and the peripheral yowlers with nothing better to do than posit social conspiracy theories about why their patch submissions were so consistently falling through the cracks. Without BitKeeper, Linus insists that less code would have been integrated.
I'll agree with fiasco if you can explain how having two camps of yowlers is better than one. At roughly the same point in time we had the Debian non-fiasco, where all potential yowlers were given their due, and nothing was released for three years. But thank the gods they weren't using BitKeeper. That would have been a real fiasco. Yes certainly, score -1 for Linus foresight, and +1 for Debian foresight on the basis of yowler management, rather than progress achieved.
If humans were birds, we would die in the shell. With our giant noggins we'd be completely unable to weild the egg tooth to crack out of our shell. Birds have a lot more pressure to develop cognitive capacity without also developing cognitive bulk. It's hardly surprising that birds accomplish more with less.
I personally regard Alex as a lot more interesting and significant than any Apple product. However, the topic is hardly worthy of discussion. It could work as a honeypot. -1 for everyone who chimes in with an obligatory parrot joke.
The actual radiation release from TMI was not earth shattering, regardless of Spin at Eleven. However, they released a report following the accident which claimed the accident had a relatively modest risk profile. This "nothing to see here" Kemeny report was published well before the Idaho National Lab finished dismantling the reactor core. What they found at the bottom was shocking. Let's just say the radioactive blob was well on its way to China.
7:45 a.m. By now there are at least 20, perhaps as many as 60, operators, supervisors, and other persons in the control room. Although none is yet ready to believe that the core had been uncovered, radiation levels in the power plant buildings are so high that Nuclear Regulatory Commission regulations require the declaration of a general emergency. While state and federal officials are being informed of elevated radiation levels, unbeknown to all, a molten mass of metal and fuel--some twenty tons in all--is spilling into the bottom of the reactor vessel. The bottom of the reactor vessel is steel, five inches (13 cm) thick. But even that thickness of steel would not be expected to hold up for more than a few hours against such heat.
Note that the information presented here comes *after* they discovered the true magnitude of the molten blob years later. It took INEEL a good while to chisel out twenty tons of highly radioactive material with a remote-controlled jackhammer.
e. There is no indication that any core material made contact with the steel pressure vessel at a temperature above the melting point of steel (2,800F).
Well, they later discovered that twenty tons of material well above that temperature was puddling in that vicinity at an alarming rate: perhaps no longer than episode in the series 24.
The story of TMI is not what was actually released, but how clueless they all were for a long time afterward about how close it came to dumping a Chernobyl-unit of molten goo into the Pennsylvania water table.
All remaining dosimeters had limits of 0.001 R/s and therefore read "off scale". Thus, the reactor crew could ascertain only that the radiation levels were somewhere above 0.001 R/s (3.6 R/h), while the true levels were 5,600 times higher in some areas.
Because of the fallacious low readings, the reactor crew chief Alexander Akimov assumed that the reactor was intact. The evidence of pieces of graphite and reactor fuel lying around the building was ignored, and the readings of another dosimeter brought in by 4:30 a.m. were dismissed under the assumption that the new dosimeter must have been defective. Akimov stayed with his crew in the reactor building until morning, trying to pump water into the reactor. None of them wore any protective gear. Most of them, including Akimov, died from radiation exposure within three weeks.
I suspect he took one look at that reading and thought to himself, "if that reading is correct, my goose is cooked". The Soviet Union never established much of a track record in encouraging the self-preservation of men poured into the breech. Typically, your reward for survival was being shot.
Back in America, the debate centers around 0.5 cancers in the aftermath, rather than the one or two hour window between what actually happened and the China syndrome. I wonder if they made an explicit political calculation: let's rush through publication of the Kemeny report before we learn anything more frightening we'd be obligated to disclose. Under the Bush administration, those obligations have mostly been terminated. They could probably write the accident report today for a future accident that hasn't even happened yet.
The "faster than the speed of light" thing surprises me. Not because of how c functions in relationship to matter and energy, but because the physicists, whose discipline has now had a full 100 years to digest these complexities, and personally, eight or more years of post-secondary education hammering home the need to state things carefully, fail to state that the fact of the violation of the speed of light for an effect can not itself be established at faster than the speed of light.
Two physicists in a similar reference frame measure two entangled particles in different light cones (any interaction would therefore need to travel faster than ligth). The entanglement effect says that if one measures red, the other measures blue. How do they confirm this? The information about their measurements must travel *at the speed of light* until information from the distinct measurements meets up. At *this point in time* they know if the entaglement effect conformed with theory or did not conform with theory. They can't posssibly determine this conclusion faster than the speed of light between the positions where the measurements were taken.
It interests me that the effect can travel faster than light, but the conclusion about the effect can not, yet I've never seen a physicist discuss this. The discussion always goes entanglement, faster than light, spooky, bada bing. It's possible that the entanglement effect doesn't resolve itself until information about the two experimental measurements (which converges in obedience with the speed of light) actually meets up. Perhaps the disentanglement takes place only *after* the results of the two experiments meets up. That would involve the experiment (and experimenters) having become entangled in the experiment. Weird? In the realm of the very tiny, that's never stopped mother nature before.
On a related point, I've never seen a physicist comment on whether it is possible to take two particles of unknown histories and prove they are not entangled. I suspect this can only be done by taking measurements which shuffle the quantum deck. Entangled particles are always introduced as an exceptional state of matter, produced painstakingly only in laboratory equipment for the purpose of conducting this experiment.
Is it not possible that most of the particles in the universe are entangled with most of the other particles of the universe? If there is no physical demonstration that two particles *are not* entangled, on what basis could you answer "no"? As a simpler case, is it possible to construct three particles A, AB, and B where AB is entangled with both A and B?
It just bugs me that the typical account of this effect rarely gets past the word spooky before exposition ceases, as if the very phrase "faster than light" causes some kind of cerebral blood flow trauma in any person who has devoted eight years of higher education in grappling with the consequences of E=mc^2.
Slashdot Mirror
There are two kinds of people in the world: those who know where they are going, and those who don't. The second group of people are easily engaged by the first oasis of flashing lights and bright sensations they stumble upon, and they probably bumble through a few "pay per click" turnstiles in the proces.
When I seriously looking for something, I can almost always tell from the result page synopsis which sites have rank they didn't deserve, and pass over these quickly.
Right now the ones that are killing me are the SpringerLink spams that seem to offer direct access to a full-text PDF, but you actually wind up at a coin-op with nothing more than an abstract. There are other research journal sites (I don't have the culprits at the tip of my tongue) where you can get to a page of full-text from the Google search results, but if you try to navigate to any other page once you reach that site, it demands a registration/login. But you can go back to Google and get the next page with a clever search. Call it the Elsevier bait and switch. If they didn't invent this, they probably wish they had.
That's a very strange optimization. A site makes its full-text content available to Google to attract people from the outside, but you have to pay a price to navigate the facility from the inside. Imagine a mall where every store is open to the street, but to access the arcade that connects them internally, you have to buy a membership.
I hate Google for putting [PDF] beside links to these nasty academic journal coin-ops. Like I'm going to pay $90 for a fifteen year old paper on some obscure protocol demonstrating a potential drug side effect in rats. There was a new result published this summer on "fast acting" antidepressants acting as 5-HT4 agonists. Turns out that 5-HT4 agonists are also the primary target of a class of drugs promoting GI motility in IBS patients, among which was a drug named Propulsid, later withdrawn due to possible cardiac arrhythmias. That gives the class of "fast acting" antidepressants several different spins. s/depressed/scared witless/
Google has become extremely lax about allowing the likes of JSTOR and Esevier to either mutate their presentation on referrer tag, or report something different to the Googlebot that what the end user sees. In my opinion, these bastions of intellectual shakedown are a little too rich for Google not to notice, or not to care.
Even if I wished to purchase a fifteen-year-old five page article for $90, I certainly wouldn't do so without first exhausting the information I can freely obtain. I can't count the number of times that JSTOR has tried to hit me up for a paper that the author has published freely on his own web site, which I locate a half dozen clicks further down the road. Yet Google is apparently making no effort to return the free full-text version ahead of the abstract behind lexan edition.
So far, Google has self-policed well enough that a Google-result-fixer has not yet become the number one Firefox plug-in, but I can see that day coming soon. It sure would be nice to have a plug-in to automatically substitute any coin-op JSTOR/Esevier link in the Google result set with a link to the free full-text, when that also exists just for the looking. My pain level has not quite reached the point of rolling my own.
I once posted a comment that what Slashdot needs is a way to pre-moderate the foregone contribution. OTOF, I suppose it was a lot more fun for everyone involved to race off to Mordor to destroy the ring *after* the ring woke up.
I term this reverse confirmation bias: if many people have tried and failed, it must be impossible.
... until it wasn't.
But what credit is there to that? Many were the claims to transmute lead into gold. What proved impossible by chemical means was by no means impossible within the framework of the right technology. I think you need to study the "Four Colour Corollary". This theorem states that the truth or falsity of the theorem is entirely independent of the number of bozos who publish unfounded and incorrect speculations disguised as purported proofs. Furthermore, we still don't have a proof that could possibly have been discovered before the computer era, so the deck was stacked towards impossible
The same thing happened within the field of AI. This still annoys me. A lot of grand claims were put forward in the 1960s, and it all fell far short of what was promised. Nevertheless, there has been an unbroken stream of solid and important results, if not yet worth writing home about. Weren't the smart people silently expecting it to play out this way all along?
I feel the statistical results are the most important:
http://www.ucl.ac.uk/media/library/robotillusions
And there recent is progress even in the long discredited field of automatic proof:
http://www.maa.org/devlin/devlin_01_05.html
Guess what? Computers are now checking computerized proofs. Does this series converge, or not?
As for this new blood test, the human genome was sequenced a scant seven years ago, the explosive shock wave of proteomics is expanding almost at the limiting wave velocity, and we are now beginning to disentagle some of the fundamental neurochemistry involved. If there are any correlates in the blood whatsoever, it would be shocking to not find them at the present time, or in very short order.
Concerning percentage prediction rates, have we learned nothing? If you have a population of size N which you wish to classify into two distinct groups, given prior p and (1-p), the information required to achieve this is N * H(p), using Shannon's information measure. If this test provides any additional information beyond the prior, one can formally determine the ratio of the unknown information this test provides. If the test is worthless, the ratio will be zero. If the test is perfect, the ratio will be one. If the ratio comes out negative, you just assume the water goes the other direction (by metaphor with electrochemistry), and substitute the absolute value.
The interesting term is the cross entropy between what the experts can determine and what this test can determine. If the cross entropy is 100%, then either test gets you to exactly the same place, and it will probable come down to a matter of economics, which the cheaper approach prevailing. If the cross entropy is significantly less than 100%, then one will likely employ both tests, possibly using the cheaper test to screen the more expensive test, depending on tolerance rates for false negatives and false positives.
Given that they have included 18 elements in this test given a small positive sample size (they don't state their negative sample size), it's almost certain that some of these 18 factors are bogus, and will be eliminated as the sample size increases. If this test is bogus, the factors remaining will dwindle to zero, as the predictive rate also dwindles to nothingness. If the test is fundamentally predictive (to some ratio of the information content) as the bogus factors are pared out, the predictive ratio will likely improve by some marginal amount, maybe enough to be worth doing, maybe not.
In the 1970s one could make easy sport of predicting that any given claimant of the "four colour proof" was wrong and pat yourself on the back for an unbroken chain of confirmations. Great work: you've managed to predict that the world is full of de
I've made substantial contributions on a variety of fronts, both technical and idiosyncratic, and I would say 75% of my contributions remain largely unscathed.
I think there is a need to delete stuff, but they go about it the wrong way. There needs to be a place you can go to find out what was previously deleted, so you don't make the mistake of adding it again, and then having it immediately deleted again. It's a preposterous system as it now stands. Every article needs to maintain a permanent record of material decreed non-notable, or not yet sufficiently notable. A lot of the pre-babble could accumulate there.
This material would not be searched by search engines, or by default from the Wikipedia search bar, but it needs to be searchable by anyone desiring to contribute material not found on the "front page" of an article.
I have to wonder if this downtick in Wikipedia activity corresponds with the end of the Sopranos. It could be just the pop-culture where there isn't much left to add.
For technical content, every third article I visit strikes me as having major holes, usually because the article was written from a single vantage point. For example, Wikipedia contains extensive information on prescription medications, but the vast majority of the pages don't provide even the date of discovery or commercial introduction. Historical information is far too thin in most articles. I also think that additional intellectual property status should be incorporated in the majority of pages where the subject matter intersects with commercial interests.
Most of my contributions tend to be better referenced than the existing text of the article I modify. Sometimes I wonder if anyone ever checks those references. Of the material that I've had immediately deleted, there have been two situations. The first is that the page is "owned" by someone who thinks the page should be a certain way. With those people, you often find yourself in and edit conflict within the first half hour. Early on I learned to pack up my bags and edit elsewhere at the first sign of territoriality.
The more interesting case is where I go to the trouble to improve the exposition so that someone with less expertise in the subject area can put things into a proper perspective. This is similar to the problem with obviousness in the patent system. Some things have no documented prior art, because everyone in the field considers the matter beneath mention. My querulous other experienced a similar frustration doing a research project on government resources in the area of ecology and the environment. Once a program runs out of money, the government web site becomes instantly frozen in electronic amber, with no edit since the previously elected government. If you are in the industry (of helping the government waste taxpayer money), you know which programs have lived and died. If you are a random citizen searching with Google, you don't, and you waste a lot of time navigating the deadwood before the penny drops.
There are plenty of Wikipedia pages with a lot of formal content about computer buzzwords and architectural classifications from past eras that have entirely outlived their usefulness now that computers are 1000 times faster, give or take a few Cheerios, but the article fails to point this out. Show up at a job interview within the IT industry, you would come across as a clueless dust bunny spouting some of that terminology. Yet I've had several efforts to add text to the effect that the industry no longer thinks in these terms instantly deleted as unspeakable POV. But hey, the patent office granted the "one click" patent because apparently no one got around to writing down that this was bloody obvious, so why should the average Wikipedian be any smarter?
One thing people consistently fail to recognize is that the growth of content has vastly outstripped the ability of the MediaWiki software to evolve in pace. Many of the political problems that now exist in Wikipedia c
Everyone is busy documenting how polarizing Carly was during her time at HP and Lucent, and then wondering why she's made the jump to FOX.
The fact is, the same emotional reflex that personalizes all the changes at HP as being the result of one person will drive people to watch all the polarizing drivel on FOX.
Consider Targets of Aggression on redirected aggression.
People need to think hard about whether Carly is a cause or a symptome of an unpleasant economic reality, and then consider directing their emotional reponse at an appropriate target. Or not. Perhaps we could tweak the moderation system to assign points for "ulcer relief" when the culpability of the target is dubious or irrelevent. I suspect some people would configure "ulcer relief" as a bonus score, because any victim is better than no victim. Even better if the "victim" is rich and famous, because then we can all deny that this is what we are engaged in doing.
http://wikileaks.org/wiki/index.php?title=On_the_take_and_loving_it&oldid=6476
One would think it would be better for slashdot discussion if TFA was not a moving target. To think slashdot is ten years old. That's one hell of a slow clue train.
Here's where we finally get to the nub of the matter. What it really depends upon is your definition of "get it working 100%". By several of my criteria, gettings Windows working 100% is an unachievable task. There are some things I don't much like about Linux, either, but even the worst of these could be fixed on the order of a few tens of man years. I happen to rate a few dozen man years as less than infinity, the expected cost of addressing my worst grievances with Windows.
My choice to view several dozen as being less than infinity might seem obvious, but in fact, it is not the popular perspective. If you read thinkers such as Danny Kanheman you will recognize that for the most part people don't think the way they claim to think. By the reflex of learned helplessness, people tend to discount the impossible, exactly as my parent poster has done. Subtracting the impossible, one can get Windows working 100% in a fairly short time period, with respect to a learned helplessness definition of 100%.
Learned helplessness wouldn't be so deeply embedded into the human psyche if it wasn't pragmatic.
It's a fairly substantial investment of time, energy, and talent to buck a mainstream trend. For any professional, I think you can only open so many fronts. My LH relative to IT is quad-CT to zero (that's an APL joke, to thoroughly date myself). On the income tax front, my LH would be closer to 7/10. I'm not motivated to win every possible battle. The last thing any nation wants is legions of empowered individuals, so the barriers are substantial.
The general public tends to constitute 100% largely in terms of instant gratifications: can I watch the newest YouTube video straight out of the box? Terms such as "will I still be able to access my personal data ten years from now after all my current software is obsolete?" rarely carries as much weight.
Nor do people stop to think much about why it is that media formats are directly tied to running specific operating systems, as if OS capabilities has much to do with it.
The other point to note is that engaging in LH has a tendency to also invoke the psyche's PR department, which isn't keen to admit any such thing, so people who have the deepest investment in the pragmatism of LH have the strongest rhetorical reflex to promote their choices as "the one true way".
Apple has historically been very good at exploiting this reflex. They do a great job of enhancing the pragmatic value of LH, and correspondingly their infinities are more infinite than most. With the brutal cooperative multitasking and virtual memory subsystem, no Apple OS prior to OS/X was within orbital radius of "100%" by any criteria I've ever accepted. The LH retort: well, you don't need that. But this PR philosophy leads Apple to more truly embarrassing reversals than most, such as their recent concession that the technical advantages of RISC over CISC in the era of 100 million transistor CPUs are commercially negligible.
One of the main terms that holds Linux back is the instant gratification bondage. Full technical disclosure of video card internals would constitute one large step toward playing iNextSonyGoobTube videos right out of the box. If the college age demographic would simply refute their instant gratification ways, and refuse to view any video encoded in proprietary media encodings, this battle could be won in less time than a Peter Jackson post-production cycle. But it will never happen. Public empowerment? Who needs that? Maybe 5% of college age people include public empowerment in their personal definition of "100%".
BTW, I'm quite conscious that posting on slashdot values my time at $0. It's less of a detriment than it might appear.
I had an idea for a major improvement to slashcode, and I've been waiting for my best shot. This is the perfect forum.
The problem with the moderation system is that you have to wait for the post before you can moderate it. This is a serious design mistake. It's quite obvious with a story like this one. An enterprising moderator could have moderated half the jokes here "-1 obligatory" *before* the jokes were posted.
This single feature would go a long way toward rebalancing the force. The chuckleheads could continue their race to first post the obvious lines, while the chucklehead CDC could escalate their counterespionage in lockstep to strangle as many of the chucklehead jokes as possible in the interval between when the blowhard lemmings depart the cliff and when they impact the ground with the inevitable dull thud of exploding whale funny bones.
What exactly is modal about sizing history to available resources? It's been an accepted human practice since the invention of the palimpsest.
If you want a non-modal design, it's an easy constraint to satisfy: have Photoshop maintain a complete undo history to the beginning, all of the time. What could be less modal than that? It might thrash like a pig on 90% of projects within 30 minutes of enaging in any serious work, but it won't be modal.
Responding to another post, if the average design artist doesn't wish to wade into discussions about technical tweakage, then that's the way Photoshop should be configured by default. Why are the software people busy implementing arbitrary tradeoffs, such as a twenty step limit on the undo history, when the artists themselves regard this as beneath their notice?
Let's get serious about non-modal design. No arbitrary limits on undo history, or internet disk cache, or colour palette depth, or anything at all that the user doesn't explicitly enable themselves. If the user complains about sluggish performance we can explain that the UI design is the least modal design possible, is full compliance with the mother of all GUI manifestos concerning ease of use. However, being open minded as we are, we also added some user controls to activate modal parameters, for those users who don't find it beneath their diginity to adjust them.
It's worse than that. The Nash equilibrium poker strategy is one that can be beaten, but in order to play this strategy, you must never take advantage of your opponent's mistakes. As soon as you adapt to your opponent's mistakes, your strategy is no longer certain to be unbeatable.
Secondly, any use of random numbers creates the possibility of an insider using cooked randomness to establish a covert channel. You want to absolutely certain that your randomness is provably random, and not cooked by anyone involved. It's almost impossible to distinguish cooked randomness from true randomness without advance knowledge of the procedure that uncooks the randomness.
It wasn't long ago that the Mickey Mouse Protection Act extended copyright term by 20 years. Meanwhile, what exactly are they doing about the tens of millions of bot infested PCs, of which the vast vast majority run Windows operating systems, and which increasingly constitutes a threat to national security?
In all my life I've never seen a scientific study about what kind of content has the potential to harm children and why. I'm sure most of my adult peers managed to expose themselves to harmful content as children. Only the least enterprising children fail to accomplish this. And what is the end result? We're all convinced we came out fine, by the skin of our teeth, but the next child won't? What exactly was impared? Our gullibility? Our willingness to vote morons into power?
Obviously there are some children who are adversely affected by coverage of the real world on the six o'clock news. But I have a feeling this bill is not targetted at that content.
He obviously hasn't read his ISO Big Bang 1.0 spec. closely enough. The language reads "quantum mechanics behaves *as if* it bifurcates into parallel universes". The actual mechanism is implementation defined. Consult your documentation on Local Universe for more details.
I've stated this humorously, but I mean it quite seriously. The other universes don't necessarily exist merely because your local universe behaves as if they did. To a certain level of accuracy, the orbits of the bodies in the solar system behave as if epicycles exist. What did that prove?
I do not welcome our FUD spewing overlords, either, but I wasn't expecting to find them employed at Slashdot earning troll bonuses at the expense of public perception.
When contention reaches critical mass and a project forks, it often benefits both groups who are now free to do their own thing; or, one camp fails to keep pace, and the dispute is effectively resolved. It didn't take long for pf taking the more aggressive road to establish itself over Darren Reed's IP Filter.
If this unhappy camp feels up to the challenge of forking the Linux kernel and taking a more aggressive approach toward performance tuning, and they can produce results in the near term that make Linus and his established approach look old and tired, then I applaud their initiative. Chances are, however, it won't be as easy to pull off as the end run around IP Filter.
The main risk to Linux as it now stands is failure to give this initiative a new name. Maybe they should call it Smoothy. I look forward to future reports of the amazingly stable frame rate in the next release of GNU/Smoothy which builds upon the newly released AMD/ATI video specifications, while all the greybeards among us cluck and moan. The one thing I don't want to see is teenagers texting Smoothy complaints about stuttering artifacts in their VR sex appliances to LKML. In other words, be off with you, best of luck, and good riddance. If you can make it work, more power to you.
In human history, the young and insane often prosper. Half of the world's major cities are built on flood plains yet civilization hasn't ended.
My view of C++ is that it invokes the necessity of reflection more than any other. The people who hate C++ the most are the same people most invested in convenient solutions, for any x, where convenience == x. For almost any x, C++ is terribly inconvenient language. Which is precisely the reaons that C++ excels at solving terribly inconvenient problems, or it would have no reason to exist in the first place.
Due to the exponential nature of computing power increases, at any given time about one third of the surface area of the problems people are most highly motivated to solve are terribly inconvenient problems: the original open-ended evolution of HTML, the recent transition to Web 2.0, scaling a search engine to 500,000 compute nodes, navigating a Mars rover.
Resource management has earned a reputation as being a hard problem only because so many people are bad at it. If you can't correctly manage memory in C++, it's pretty much guaranteed you can't write a sychronization correct multithreaded application.
Most of us have learned to think in terms of preconditions, post-conditions, and invariants for coding a correct iteration that terminates. It's no different to correctly specify object lifetimes in a code base lacking automatic GC. The object comes into existence under this condition, while it exists this invariant is true, when the invariant fails, the object is disposed, and the time between the last two steps is as brief as can be practically arranged (ideally, the distance between two successive program statements).
I suspect Firefox leaks so much because the coders have descended to mechanistic thinking: to do this task we need to allocate this object, later when this mechanism is activated, if it appears correct to do so, the object is disposed. Sweet. We completely bypassed the inconvenient step of specifying a rigorous object lifetime invariant expression. Hurray! The project will ship on schedule.
By contrast, the Java programmer is spared any guilt over bypassing this difficult logical step. Only the GC doesn't entirely save you. If by a logical error, you leave a reference to an object intact that is in fact no longer needed, the GC will never free this object. With any luck, the object maintaining this unnecessary reference will get cleaned up later on, and by implication also its orphans. And there won't be much surface evidence of this missed logical step either, aside from the usual observation that the Java app. is generally consuming far more memory than you would like to believe.
One C++ project I happen to like is the monotone SCM. When you set out to write a rigorous SCM tool, you are committing yourself to a life time of high calorie lunches at the reflection buffet. At this point, C++ is the least of your problems. In fact, I always gravitate toward projects where convenience for its own sake is shuffled out the door on day one. Do you think Google built their massively parallel computational architecture by showing up every day asking themselves first of all "what is most convenient?" No, a challenge on that scale demands a renewed investment in hard thinking after every meal.
I would learn Haskell for the inconvenience it offers, but I already enjoy 90% of the inconvenience that functional programming has to offer by grappling with C++ template instantiations.
To paraphrase Einstein--for those of you who find it tedious and inconvenient to wade through a sustained meditation--your programming language should be as convenient as possible, but never more convenient that the problem your program is attempting to solve.
To update Knuth: premature convenience is the root of all evil.
How about we settle for getting the linguistics right, and ignore history? Who put the idea forward that people gravitate towards a name for the purpose of ego equity? That's not how the human mind works. People tend to fixate on affordance distinctions. A person who is trying to decide which of his favoriate applications will run on a system, at that point in time, probably could infer more from the Linux/BSD distinction than whether the userland was GNU or proprietary. As some posts have pointed out, many proprietary systems had GNU userland tools available. In some cases, such as writing complex shell scripts, one might be primarily concerned about whether the target userland was GNU or not-GNU, but I don't think it was the overriding consideration most of the time.
Secondly, the name of the distro tells you a lot about how the system is managed, and whether you have the right sysadmin toolset: packages/yum/apt, etc. For that matter, I don't recall ever installing a GNU file system. You choice of file system was usually dictated by your distribution, not your userland tools. In fact your distribution goes a long way to determining your politics as well: under Debian certain things would be missing or difficult that would be defaults elsewhere. I don't mind calling Debian GNU/Linux, because Debian explicitly bundles much of the FSF politics. I never regarded the GNU in Debian GNU/Linux as a reference to the GNU userland tools.
I think it would be good if OpenBSD had its own C compiler, even if it only generates reasonable code for x86 architectures. A C compiler aimed more at correctness than performance is not complicated unless your code generator needs to support every known platform. It wouldn't mean that gcc needs to be phased out, it could be used on a complementary basis, especially if it has more rigorous portability and error reporting, suitable for compiling a kernel, if not for every imaginable userland program, or the monstrosity known as C++, which is my preferred language for statistical NLP, but would be an abomination for kernel purposes.
There are many balanced people who accomplish a lot in the world, but you fail to recognize the difference between self determination and self discovery. Is it really true that a person can decide to follow either path (a pure choice) without sacrificing any golden geese either way? Or is it more the case that you figure out that your obsessive fascination with an intellectual problem brings something to the table that an overt attempt to style yourself balanced would sacrifice or destroy?
I think the key perspective on this question is that the majority of deep discoveries in mathematics have come from individuals under the age of twenty-five, even if some of them were published later. There are a few exceptions of course, but much of the history of mathematics was paved by young males (mostly) in their youngest, most obsessive years. Obsessive energy is part of the human condition when we peak as young adults. Especially for males, who face a higher risk of not mating at all.
Later on when we are old and slow and stupid, obsessive behaviour has fewer rewards, while balanced behaviour has more. I had all my best ideas in my twenties. Fortunately, I kept a couple of those ideas to myself, so I can still pretend to be obsessed about something I'm busy inventing, but it's just the geek equivalent of buying a convertible when your hair turns grey.
I've known a number of males whose obsessiveness surpassed any concern over group belonging. I haven't met nearly so many women who regard inclusiveness as negotiable.
I don't think the issue is that male engineers have something that women engineers are denied. I think women wish to enter engineering on terms males never demanded. That's probably a good thing. The great advances in future society are more likely to come from groups than individuals.
I also think that women sometimes set their sights too high. If there has ever been an MRI scan taken of a geek coder in the moment of supreme creation (at the keyboard), I don't think you'd see a lot of blood flow to the chit-chat circuits. I do math rarely, but every time I have a math problem to solve, so much of my brain is recruited, I become inert to my surroundings. Any culture where that kind of departure from life is part of the discipline, is going to develop differently, and maybe in ways that women don't enjoy as much, above and beyond boys being boys, and jerks being jerks.
What I'm suggusting is that if you could magically take geek engineering culture and strip away the discriminatory aspects, I believe there are many women who would continue to complain about the end result as an unfriendly work environment. If you don't have some hard edges in engineering culture, you get the NASA effect, where too many people are checking out the social and political consensus, and not enough people are saying, "uh, excuse me, you can't do that that way" with least regard for affiliation. And not just one guy at the bottom of the chain of command.
Microsoft innovated new business practices, such as submitting simulated video evidence in major court cases without originally designating it as such.
People make too much noise about technical innovation. In the vast majority of cases, 90% of the difficulty is in running a successful company in the first place. Make that 99% if the company employs too many Linux hackers.
Dell's primary innovation was in its distribution model, which suited white box systems for about a decade, until the price came down and the market moved toward laptops, which people expect to touch before purchasing. I respect true business innovation as much as I respect true technical innovation. However, when it comes to business practice, there is a fine line between innovative and sharp. Selling printers with the starter cartridge half full of ink? That's a sharp business practice, not an innovation. Enron's mark-to-market accounting methods? I don't even need to comment on that one.
It would be tough to speculate what proportion of technical innovation in the world these days is embodied as software. It's also unclear whether the global innovation rate in the modern world depends in any way upon patent protection. For every person who says she wouldn't have bothered without patent protection, you can probably find another two to say they would have gone further without the risk of patent interference.
The case where patents seem most important is where the resulting business model requires a large initial capitalisation. I doubt the patent system has any connection with innovation rate, but I suspect it does have a connection in the level of investment available to make those innovations commercially viable.
Perhaps what we need is a two term system. For the initial term, perhaps six years from date of filing, the present system would work, with some of the more glaring stupidities cleared up. The renewal application would be a contested process, in which the innovator needs to justify considerable non-triviality while facing a very broad interpretation of any prior art brought to the attention of the patent office.
Such a system might cut the patent application rate in half. Of those approved, maybe 20% pursue the term extension, and that process costs five to ten times as much as the initial application, because these examiners are supposed to know what they are doing, and they have contesting materials to arbitrate.
Net effect is that the cost of the patent system remains about the same, the number of worthless patents declines, and many of the marginal patents are washed off the books at the conclusion of a relatively short six year initial term.
I don't think we can eliminate software patents altogether, it's far too large a sphere of human intellectual work.
It strikes me as non-constitutional. How is anyone to succeed in the pursuit of happiness when major credit agencies are divulging damaging falsehoods about you in your significant financial relationships? Truly, in American society the pursuit of happiness must translate to litigating these credential-bungling credit rating agencies all to hell.
I also think the law should support a class action law suit over the design of the credit card security mechanism which created this problem in the first place, towards appealing the common notion that their credential bungling is our identity theft.
One post suggested that by enabled me to obtain better interest rates, I benefit from having my (good) credit history collected and reported.
If that is the case, a capitalist society would enable me to hire third parties to gather and report this information. If I wish this service because it enables me to obtain favourable terms, then I seek it out and I pay for it myself.
I have still have that Nixon tape running through my head after viewing Sicko the other night. Note to the Moore haters: let's ignore the messenger, and let Nixon speak for himself, in his own words, brought to use by the magic of magnetic tape:
From http://whatisthemessage.blogspot.com/2007/06/michael-moores-sicko.html
The credit rating agencies enjoy the same favorable legislative tilt, with the same detrimental consequences to the typical American: instead of litigating against credential bungling at the hands of the Enrons of personal record handling, we cower in fear over this ficticious concept of "identity theft" which merely a synonym for the non-constitutional legislative contrivance that "if corporations lie about you, you have no power to stop it".
The real point here is that peer review is not the fine filter it is portrayed as being. The end result of the process might be science, but much of the workings along the way are sub-standard.
The people who wish to replace Wikipedia with a different system of "expert review" are just replacing one flawed group of people with another flawed group of people, only this new flawed group of people has manicured egos, and they have all made a pact with a social system which reasons "no need to point out every flaw in the papers I review, do I wish to draw the same scrutiny to my own papers while my tenure track remains uncertain?" So they all sit around pretending their shit doesn't smell, and hope that the rest of us fall for the ruse.
I detest the invisible nature of peer review. The person doing the review is not accountable and doesn't leave a written audit trail. How is that acceptable to Wikipedia, but the standards of its own community?
If a scientist's own work is discredited, does that discredit any papers he might have reviewed? Fat chance that review is repeated, those papers are already on the books. The only offense that gets formally repealed is plagiarism of one published paper by another published paper, because that is an offense of the record against itself. Failures of the review process itself never leads to formal repeal. Or do people here know otherwise?
It's not that I think the current process in Wikipedia works terribly well, but at least it has the potential to evolve into something we haven't tried yet, that could serve as a useful counterbalance to the multitudinous small offenses that overworked professors are almost obliged to pull off in the cut-throat academic world of publish or perish.
We only have to tear the page out of Freakanomics concerning his analysis of Sumo politics to see what happens when the inmates run the asylum.
America is no better. Witness NBA handling of the corrupt referee Robert Hoyzer. The only principled solution is to ban betting against the spread. A corrupt referee who changes the *outcome* of games will find himself subject to merciless scrutiny among the players themselves, who have a huge vested stake in that outcome. Gambling on the outcome of sporting events can't be manipulated half so easily as the spread. Any punter who bets on the outcome of a meaningless game at the end of a season deserves what he gets.
The kind of person who accepts betting against the spread as something that is adequately policed is the same kind of person likely to argue that the peer review system works well enough. I don't agree in either case.
On a tangent the size of Jupiter, I once asked myself the question, not in the spirit of sci-fi, but in the spirit of someone who might set out to write some sci-fi (which is not my cup of tea), that if we regard the hacking off of limbs in the American civil war as medically barbaric, what will appear barbaric about our generation to the generations of the future?
To be brutally honest, most people are not equiped to pose this question, much less answer it. If you've seen those end of the year lists, you know what I mean. To begin with, the answer is never the same. We don't need to invent the theory of pathogenesis twice. Nor is it as literal as replacing a D-saw with a fancy collection of salt shakers.
My suspicion is that 100 years from now, society will look back at the 20'th century and gasp over our practice of approving medications on the basis of population studies.
Not only will it be regarded as scientifically suspect, even barbaric in the exposure of the general population to poorly understood chemical agents, but also as a great industry of hucksterism. Take a therapeutic substance that achieves a large effect for a small number of individuals who need precisely that treatment, conduct a large population study with includes a small group of these people, th
Your conceit is slimier than you are making it appear. You are leaving out the step where the duality of the license is conveyed by the original developer. Or are you just assuming that if you come across any source code with both the BSD and GPL licenses pasted inside that this implies a dual license? Presumably you need a statement from the original author, preferably in writing, even better written into the source file itself, that the original author intends the two license texts to function under a dual-license either/or interpretation.
The only way copyright for a work can be verified by a recipient is when the derivation chain can be reconstructed through the derived work(s) back to the original work. If a person goes out of his/her way within the letter of the law to muddy that trail, I would regard the actions of that person as quite vile.
Perhaps you can, as you suggest, legally strip out one of the licenses (and the text granting the dual license, if that was present within the source file). But then if you distribute your derived work, you are handing people half a proof. Sensible projects would not accept the work in that form. They would want to see the derivation chain before taking that risk.
You seem to be suggesting by implication that the derivation chain can be maintained outside the text of the source file by some auxilliary shuffleupagus which most users of the source are unlikely to chance upon. If that wasn't your purpose, why delete the in-place full disclosure of the dual-license status of the original work?
In my view, there is no ethical way to substitute for the attribution "this is a derived work of a work by Bob Dawg who granted permission to do so under a dual-license" to "this is a derived work of an original work by Bob Dawg [and good luck figuring out the terms of his original license grant]" and by grant I mean his *statement* that permission was granted under dual-license terms, not the mere inclusion of two different licenses into the top of a source file, which in the absence of a clear statement, I would regard as ambiguous.
I see no problem with writing "This a derived work of an original work by Old Author dual-licensed under the GPL and BSD licenses. The original work is available at URL. New Author has elected to make changes under the GPL, which is reproduced below."
My point is not about trimming out the text of the BSD license, it's about muddying the terms of the original grant through crafty omission in the wording of your own grant, to the detriment of any future party who comes into contact with the derived work.
Your use of the word legal word "requires" (in the selfish conjugation: that I myself can not be sued by any party) neglects getting along nicely with any other parties to the overall process, upstream or downstream.
I could be wrong. The mere inclusion of two license texts could amount to an implicit grant of dual-license either/or status. So what? The only thing standing upon the letter of the law achieves is breeding more lawyers. Then you have to pay them a lot of money to stop breeding. With luck, and patience, and buckets of cash the legal bugs are resolved. Original authors, if they have any time left to code, learn to paste a more explicit legal text into their document that accomplishes the same thing legally at great cost that common sense and good will could have accomplished in the first place.
Only a dimwit or a lawyer advocates a move in the game tree where the only consequence of the end game is that the lawyers get rich while the loophole is closed. That is after all an explicit function of the legal system: to be so damn expensive and unproductive that it scares us into applying common sense against natural inclinations such as the one you expressed.
Refering to it as the "BitKeeper fiasco" merely serves to reveal your biases. At the time Linus chose to adopt BitKeeper, Linus had become a choke point on patch integration. He claimed the use of BitKeeper greatly improved his efficiency and coordination.
As I understand it, a lot of work was going on behind the scenes in parallel with the adoption of BitKeeper to improve the social dynamics of patch flow among the kernel hacker cabal. Yes, there was a loud and sustained polemical outcry, but show me solid evidence that kernel evolution, growth, maturity, or stability was compromised during this period as a result of all that background noise. Background noise is a fact of life in most open source projects.
When McVoy finally did have his predicted rug-pulling two-year-old temper tantrum, Linus had come to a good understanding of exactly what he needed in a collaborative source code control system and git was not long in arriving. Would it have been possible to implement git correctly three years earlier without the basis of that experience? Would the social systems have been ready for it? These are important questions the BitKeeper yowlers are rarely seen to seriously consider.
What fiasco? Perhaps tempest in a teapot is a better description. If a crappy SCM had been adopted instead of BitKeeper, Linux would have had two camps of yowlers: the hardest working cogs at the center whose work process integrating patches was poorly supported, and the peripheral yowlers with nothing better to do than posit social conspiracy theories about why their patch submissions were so consistently falling through the cracks. Without BitKeeper, Linus insists that less code would have been integrated.
I'll agree with fiasco if you can explain how having two camps of yowlers is better than one. At roughly the same point in time we had the Debian non-fiasco, where all potential yowlers were given their due, and nothing was released for three years. But thank the gods they weren't using BitKeeper. That would have been a real fiasco. Yes certainly, score -1 for Linus foresight, and +1 for Debian foresight on the basis of yowler management, rather than progress achieved.
If humans were birds, we would die in the shell. With our giant noggins we'd be completely unable to weild the egg tooth to crack out of our shell. Birds have a lot more pressure to develop cognitive capacity without also developing cognitive bulk. It's hardly surprising that birds accomplish more with less.
I personally regard Alex as a lot more interesting and significant than any Apple product. However, the topic is hardly worthy of discussion. It could work as a honeypot. -1 for everyone who chimes in with an obligatory parrot joke.
http://americanhistory.si.edu/tmi/tmi03.htm
7:45 a.m. By now there are at least 20, perhaps as many as 60, operators, supervisors, and other persons in the control room. Although none is yet ready to believe that the core had been uncovered, radiation levels in the power plant buildings are so high that Nuclear Regulatory Commission regulations require the declaration of a general emergency. While state and federal officials are being informed of elevated radiation levels, unbeknown to all, a molten mass of metal and fuel--some twenty tons in all--is spilling into the bottom of the reactor vessel. The bottom of the reactor vessel is steel, five inches (13 cm) thick. But even that thickness of steel would not be expected to hold up for more than a few hours against such heat.
Note that the information presented here comes *after* they discovered the true magnitude of the molten blob years later. It took INEEL a good while to chisel out twenty tons of highly radioactive material with a remote-controlled jackhammer.
From the rather tame Kemeny report
e. There is no indication that any core material made contact with the steel pressure vessel at a temperature above the melting point of steel (2,800F).
Well, they later discovered that twenty tons of material well above that temperature was puddling in that vicinity at an alarming rate: perhaps no longer than episode in the series 24.
The story of TMI is not what was actually released, but how clueless they all were for a long time afterward about how close it came to dumping a Chernobyl-unit of molten goo into the Pennsylvania water table.
Concerning Chernobyl:
All remaining dosimeters had limits of 0.001 R/s and therefore read "off scale". Thus, the reactor crew could ascertain only that the radiation levels were somewhere above 0.001 R/s (3.6 R/h), while the true levels were 5,600 times higher in some areas.
Because of the fallacious low readings, the reactor crew chief Alexander Akimov assumed that the reactor was intact. The evidence of pieces of graphite and reactor fuel lying around the building was ignored, and the readings of another dosimeter brought in by 4:30 a.m. were dismissed under the assumption that the new dosimeter must have been defective. Akimov stayed with his crew in the reactor building until morning, trying to pump water into the reactor. None of them wore any protective gear. Most of them, including Akimov, died from radiation exposure within three weeks.
I suspect he took one look at that reading and thought to himself, "if that reading is correct, my goose is cooked". The Soviet Union never established much of a track record in encouraging the self-preservation of men poured into the breech. Typically, your reward for survival was being shot.
Back in America, the debate centers around 0.5 cancers in the aftermath, rather than the one or two hour window between what actually happened and the China syndrome. I wonder if they made an explicit political calculation: let's rush through publication of the Kemeny report before we learn anything more frightening we'd be obligated to disclose. Under the Bush administration, those obligations have mostly been terminated. They could probably write the accident report today for a future accident that hasn't even happened yet.
The "faster than the speed of light" thing surprises me. Not because of how c functions in relationship to matter and energy, but because the physicists, whose discipline has now had a full 100 years to digest these complexities, and personally, eight or more years of post-secondary education hammering home the need to state things carefully, fail to state that the fact of the violation of the speed of light for an effect can not itself be established at faster than the speed of light.
Two physicists in a similar reference frame measure two entangled particles in different light cones (any interaction would therefore need to travel faster than ligth). The entanglement effect says that if one measures red, the other measures blue. How do they confirm this? The information about their measurements must travel *at the speed of light* until information from the distinct measurements meets up. At *this point in time* they know if the entaglement effect conformed with theory or did not conform with theory. They can't posssibly determine this conclusion faster than the speed of light between the positions where the measurements were taken.
It interests me that the effect can travel faster than light, but the conclusion about the effect can not, yet I've never seen a physicist discuss this. The discussion always goes entanglement, faster than light, spooky, bada bing. It's possible that the entanglement effect doesn't resolve itself until information about the two experimental measurements (which converges in obedience with the speed of light) actually meets up. Perhaps the disentanglement takes place only *after* the results of the two experiments meets up. That would involve the experiment (and experimenters) having become entangled in the experiment. Weird? In the realm of the very tiny, that's never stopped mother nature before.
On a related point, I've never seen a physicist comment on whether it is possible to take two particles of unknown histories and prove they are not entangled. I suspect this can only be done by taking measurements which shuffle the quantum deck. Entangled particles are always introduced as an exceptional state of matter, produced painstakingly only in laboratory equipment for the purpose of conducting this experiment.
Is it not possible that most of the particles in the universe are entangled with most of the other particles of the universe? If there is no physical demonstration that two particles *are not* entangled, on what basis could you answer "no"? As a simpler case, is it possible to construct three particles A, AB, and B where AB is entangled with both A and B?
It just bugs me that the typical account of this effect rarely gets past the word spooky before exposition ceases, as if the very phrase "faster than light" causes some kind of cerebral blood flow trauma in any person who has devoted eight years of higher education in grappling with the consequences of E=mc^2.