Ok, time tu curn out the old favorites.:) Seriously, you can use Professor Nash's work (for which he won a Nobel Prize, so it had better be right!:) which demonstrates a cooperative economy is a viable solution, and that a competitive economy MUST fail to reach an optimal level of performance. That alone won't do anything, but it is the cornerstone on which you would need to build an anti-DRM argument.
How does this work? Well, DRM costs money to make and to maintain. It takes resources away from the DRM'ed device, reducing the amount of product you are capable of selling. However, the DRM has no intrinsic value in itself. Thus, it must reduce point-of-sale profits and it must increase the costs of producing the product in the first place. (Programmers and hardware engineers with the expertise to produce any meaningful DRM won't come cheap.) It also means you are taking staff away from producing more material, thus reducing what you have to sell, which will also reduce profits.
How does the Prisoner's Dilemma fit in to this? Well, you want to have stronger DRM than competitors, so the pirates will target them and not you. At the same time, you have to be working with them and with hardware vendors, so that your DRM is still readable on other people's systems. Unreadability kills sales - just ask Sony's Betamax division. Oh, and at the same time as providing all this information, you can't provide so much that an outsider can break your DRM scheme.
The solution? There isn't one. The constraints are mutually exclusive. You cannot simultaneously be open and closed, different and the same, restricted and unrestricted. This is the problem that corporations have faced with tape copying, floppy disk copying, or the copying of any physical medium whatsoever. The moment you aim for one ideal, you eliminate any possibility of achieving another, and all of the ideals surrounding DRM are contrary to the ideals surrounding sales.
So how do we deal with piracy, if not with DRM? Not having DRM will maximize sales of those who are interested, but piracy has the potential for reducing that number. (I'm playing devil's advocate here, as there is no data to support that claim, but as it is widely believed, you may as well have an argument that works with those who believe it.)
The first thing you need to consider is shelflife. Most products really have a very short shelflife. Sure, there are some classics, but there really aren't many of them. There are far, far fewer old products that still sell than there are new products that sell. Furthermore, there are even fewer old products that sell AS WELL as new products, even after taking into account that no development needs to be done on something that has been around forever and a day.
Widespread piracy takes time. Most distribution channels are slow, error-prone and unreliable. They work - they work very well - but only over a long timeframe. Over a short timeframe, they are often pretty useless. Most don't scale well and rely on self-propagation to get any kind of performance. This doesn't happen overnight... unless linked to from Slashdot.
What is wanted, then, is increased turnover, shorter software lifecycles, less bundling, more fluidity. Instead of slowing the pirates down, speed the market up. It's all about relative performance, so the two have the same effect on how many will be interested in buying. What does this do to the bottom line, though? Well, now you eliminate the Dilemma altogether. By relying on high dexterity, rather than heavy armour, you don't need any extended cooperation from others, you are limited only by what the media readers are capable of. Nothing else.
What does this do to the division of labour? Well, those who would have been involved in PRing the DRM can now be selling the product. Those involved in researching, developing and implementing the DRM can now be making new products. You have the same number of people
In order to determine the correct amount of virtual memory, you need to decide what contingencies virtual memory is being used for. Don't worry with what you are normally doing, because you can always buy enough RAM to do whatever you do normally, on modern systems. On old systems, where physical RAM was capped at a very small amount, you didn't have that option. So, virtual memory today is what you need when there are abnormal conditions.
The 2.5x case comes from two simple rules of thumb. Firstly, you need enough to be able to hold the whole of what is in RAM now, plus everything you want to swap in, plus enough to minimise fragmentation and cover overheads. Secondly, the more swap space you have, the more metadata you need to manage it AND the greater the latency to perform any kind of swap AND the more swapping you need to do to run all active processes. Too much virtual memory is a Bad Thing. Having 2.5 x RAM was considered a good compromise and it is one I use to this day.
Today, both rules of thumb still hold. The largest single object you can have is one that fills ALL of RAM after the kernel, and you absolutely must have sufficient swap space to be able to dump that object to disk. If you don't, then the kernel will either panic, kill the process or cause any other activity to behave unpredictably. It won't have the resources to behave correctly. Any number of these objects could, in theory, be swapped out - but remember that they don't run when on disk, only when in memory, so the more you have, the smaller the timeslice each will get - and the sum of those timeslices will go down, as you need to allow time for the swap to take place.
However, today isn't quite the same as yesterday. The difference in performance between hard drives and RAM has changed. There is better caching on the drive. The swap algorithms are smarter and there is more understanding of what metadata is useful and what really has no value. Process handling is also smarter, so processes aren't necessarily run in order - round-robin scheduling is used for some time-critical stuff on Linux, but most applications use a more relaxed system.
Also, programming has changed. There is greater re-use of tools and libraries - well, sometimes - and this means that the largest object you really have to handle at a time is much smaller than the size of RAM. A certain fraction of what's left will be used by shared libraries and shared resources.
Lastly, because hard drives are reasonably cheap and most PCs can handle several at the same time, you are far far better off getting a drive and dedicating it to swap. This is good for many reasons, not least because the drive won't have to move the read heads from data space to swap space and back. You eliminate a vast chunk of seek time, reduce the stress on the drive AND can experiment with different swap sizes without risking losing data.
I would therefore STRONGLY advise using the classic 2.5x and a different hard drive, but if you can't do this for some reason and want an updated formula, here is what I would suggest:
The meaningful RAM will be equal to the total RAM minus the space used by the kernel and vital, non-swappable resources/daemons. Multiply this by three for 7200RPM hard drives or by five for 15000RPM hard drives. Multiply by one and a quarter for basic swap schemes, or by one and an eighth for profiling/intelligent swap schemes. Add the size of the hard disk cache, if the cache uses a high water mark to control operations. Subtract the size of the hard drive cache (unless this takes the size below zero) if the behavior is controlled by a low water mark only. Add one megabyte per simultaneous user. Add one megabyte for each large -or- long-running application likely to be running simultaneously. Subtract the total size of all the shared libraries likely to be loaded in the case just considered.
This is a LOT more complex than 2.5x, so much so that I generally wouldn't bother using it except
The authors have done an excellent job of using technobabble, block transfer computations and HomerSpeak(tm), so I present to you an English translation, courtesy of a copy of Babelfish that fell through a wormhole from the 43rd century.
"We would like an OS that has the following characteristics:
It must be modular, and we must be able to decide what modules are present, and which modules are used when more than one would perform the task
We must be able to add our own modules or modify existing ones to suit our requirements
Modules should not be atomic, but be designed in a sufficiently structured way that in many cases, all we need to supply is the glue layers, not the internals (eg: modules as compilable scripts)
We should be able to connect whatever we like to whatever we like, defining only a translation between the two interfaces (ie: everything should be capable of using a generic I/O system such as pipes and a generic language such as XML)
What can be automagic should always have the option of being automagic (and the user should be able to make corrections when the machine guesses wrong), including any/all aspects of architecture, optimization, topology, remote services, load balancing, fault tolerence and security
Automagic guesses should be incapable of leaving the system in a worse state than prior to the automagic guess, where policies/user-agreed compromises define what is worse
Linux does 2, most of 1 and 3 and some of 4 and 5. Plan 9 seems to be working on other aspects of 4 and 5. In theory, mandatory access controls should provide 6, as automagic functions would be tightly defined, but MAC is not well-understood on general-purposes operating systems to the point where the Fedora defaults of SELinux are almost useless.
I'll go through these cases, as the above list isn't entirely clear:
Linux does not have any real support for modules that can perform the same function, as the autoloader couldn't distinguish and you've a high risk of symbol conflicts.
Linux doesn't support scriptable modules from components, where the components are the physical kernel modules and a script is loaded as data to define how these are glued together.
Neither Linux nor Plan 9 have any means of forcing GUI-based applications to use pipes or similar mechanisms, and there is no enforced generic protocol on such mechanisms anyway.
The user should only be concerned with logical properties (what a service does, what the latency is, what security exists, etc) and not with physical properties (how many CPUs there are, if the system is little-endian, big-endian or PDP-endian, if the system is local or remote, what OS is running, etc) EXCEPT where the user wishes to know that information and/or control things at the physical layer. Non-transparent systems can't do the former. Transparent systems can't do the latter.
Automagic is useful ONLY because it is dangerous. If it were not capable of doing something really bad, it would be incapable of doing any good. Automagic without the user being able to seize the reins is VERY BAD. Automagic without the need for the user being able to do so is pointless. Automagic and the user should be able to mix safely, without causing anything to explode or emit magic blue smoke.
Fault tolerence and automagic only mix if the fault tolerence can detect a bad automagic decision and undo it automagically. (However, allowing this to be recursive would be a Bad Thing.) Fault tolerence should also be able to anticipate certain categories of fault and use automagic mechanisms to reduce the liklihood of the fault actually occuring.
Similar things apply with respect to optimization, load balancing, etc, making use of the automagic features of the OS to both reverse incorrect choices and also improve on current choices.
I can think of a few dozen Open Source-based companies that are offering far more innovative products, far better products and sometimes even both at the same time. I have no problem with an editor saying "these are worth watching", but I do have a problem with an editor astroturfing. I also have a problem with editors making bold claims without any actual evidence. All we're getting is the corporate handout.
A proposal I made many many years back was to run a league table, where so many points were scored for the release of an open source product, so many points were scored for updating someone else's open source product, so many for closed source products that supported or enhanced the usability of an open source product, etc. The more open source, the more points. Also, the more significant (in terms of power, flexibility, etc) the more points, and the more practical the more points.
The idea was to show who actually was doing work, versus who was merely bragging about it. The idea was also to make major Open Source figures (some of whom have never actually released a product but have contributed to many) just as prominent in the table as major corporations with Really Big Bucks to throw around. It would also show those who are working on making Open Source a key player in the computing world, even if their products are not themselves Open Source.
(Oracle would score points for having put their corporate database on Linux, for example, but it would not be as much as Computer Associates for putting their corporate database - Ingres - not only on Linux but opening up the source as well. Postgresql would score more yet, as it is not only Open Source but regularly maintained.)
Newcomers are at no disadvantage, because whatever REAL added value will show up notwithstanding the newness or the lack of awareness. If there's no added value, then there is nothing there to watch. It's merely a rebadge. If there's added value and this value is constantly added to (which is what a new company should be doing), then it will be a very obvious rapid-riser through the charts.
Proprietary vendors who are wary of opening their high-value major product lines will obviously not score as well, nor should they, but they will be represented as a function of what they have contributed - directly in terms of products, and indirectly in terms of improved usability.
Then, journalists MIGHT have a clue as to what is interesting and what is not. They might also have a clue as to what is significant, what is likely to become "big news" and what is worth the effort of covering.
As it stands, they neither know nor care. Whoever pays the piper calls the tune. Mind you, what I'm aiming for, with this idea, is not to introduce honesty, but rather to give people a copy of the songbook. Let people see for themselves if the tune is any good or is even what it's claimed to be. It would seem to me that an informed userbase will take care of the honesty issue by itself.
Whirlpool is also in mhash, which is now many versions on from the ones supplied by Fedora Core and Gentoo. Oh well. It's also in the Linux kernel's crypto library. Whirlpool is a damn good hash and uses the same principles as the Rijndael cipher, which means that the underlying concepts have been deeply analyzed twice - once in each form - showing the basic ideas are fairly solid. Being long should reduce the risk of collisions so is actually a strength in many cases - particularly as we're talking bytes, not megabytes.
Tiger is another good hash function - faster than Whirlpool, smaller for those embedded cases where even the bytes matter, and I believe it is not known to have any attacks against it. Tiger also appears in mhash, not certain if it's in the kernel but it should be.
I don't see that it is really of any consequence whether anyone has actually demonstrated an attack on SHA - the point of security is to not wait until AFTER the house has been plundered to upgrade. SHA is FIPS-180, but if there is even a theory on how it might become broken, I would urge people to use something stronger. Security that is only certain to be good against skript kiddies is really not very useful security.
I've been told by bosses when working for the DoD that although the code I wanted to use was indeed audited, FIPS-compliant and published by the DoD themselves, it wasn't on an approved list of certified FIPS-compliant software so wasn't acceptable. When I asked how something the DoD wrote and published as an official DoD application could possibly not be acceptable for use by the DoD, I was told that procedures were procedures and had to be followed. The fact that it was GOTS and written by the people who do the certifying was of no consequence.
So, don't imagine that this is going to be an easy one. Open Source projects by IBM might be easier to get past the Great And All-Seeing (but definitely not all-knowing) Pointy Haired Bosses - at least some will be familiar with the phrase "nobody ever got sacked for buying IBM" and may even still believe that. Some of the Apache projects might be workable, provided you use the line of reasoning that since Apache is listed on IBM's website as a project they are working on, it is covered by the "nobody ever got sacked for buying IBM". You don't have to tell them that IBM is only one member of a large consortium, and it might be better not to.
Some projects were connected to IBM and other major corporations but are now independent. Postfix is an example of that. I believe evlog (Enterprise-grade event logging for Linux) is also such a project. Speaking of evlog, I would DEFINITELY suggest using it in any commercial or Government setting. It's not that good and Linux has plenty of other security, but "Enterprise-grade logging" is mandatory in many cases and this provides it. It ticks the right box, even if it doesn't do a whole lot more for you. It's a pure CYA and nothing more.
ISO 9000 (or later) compliance is probably the toughest requirement, as it stipulates documenting the process and activities, where the level of documentation depends on how critical the project is, and Open Source projects have neither that type of documentation or any real concept of criticalness as components are freely reusable. Your best bet is to work through vendors that are themselves ISO 900x certified AND supply either the Open Source OR the links to those projects, then argue that by documenting the use of a project that comes from an ISO 900x certified source, you inherit the certification indirectly. Some bosses will buy that easier than others and depending on the structure of the organization, you may have flexibility on who you present the case to. If so, shop.
Nobody but an idiot would contend that Sir Isaac Newton was not an intellectual giant, a true genius, with astonishing brilliance in physics, mathematics, chemistry and music - a range few modern "geniuses" can even hope to compare with, even as they fail to compare even in a single one of them.
Nobody but a wilfull participant in intellectual fraud would - today - contend that he discovered the laws of motion (he plagarised them off Descartes) or calculus (Gottfried Leibniz produced the modern version and published before Newton, Archimedes was the original discoverer although it is very unlikely either Newton or Leibniz was aware of the earlier work).
Nobody but a fool would deny that Newton's pressure to obliterate the wave theory of light set optical science back many centuries, and that his hostility towards others caused serious enough rifts that collaboration between Europe and Britain, or between British scientists, was seriously inhibited for a long time after his death, retarding science in general.
Does that make Newton any less of a great scientist? No. Does it devalue the Principa? No. So why, if this holds true of Newton can it not hold true of Yau? Why, if we can accept that Newton had serious character flaws that led to intellectual fraud, forgery, deceit and political manipulation for his own ends, can we not accept that an equally brilliant but equally troubled intellect would suffer the same flaws?
Furthermore, since a genius HAS exceptional intelligence, should we not hold them to a higher standard of conduct, rather than a lesser one? Particularly in modern times, where most such flaws are easily treatable?
(Incidently, since I am drawing parallels with Newton, does this mean I believe Newton should have his knighthood revoked? Actually, yes. He has no business holding credit where that credit is not due but was acquired through malpractice.)
As for accessibility - that is of no consequence to the matter at hand, which is a matter of who discovered what. Clarifying is not the same as discovering. Clarifying is a continuous process that starts from the moment of discovery and will never stop. Original discovery only happens once - otherwise it ceases to be original - and the issue at hand is the originality of the discovery, not whether it is readable.
Sure, "human nature" is inevitable, being humans. However, that begs the question of what is human nature. Most philosophers over time seem to agree that character flaws and delusions are largely optional. (This does not include chemical imbalances, which are optional today only insofar as treatments exist.) Professor Nash, for example, was fortunate. He suffers from schizophrenia but when he was diagnosed, treatments did exist. Not very good ones, sure, but they worked. Einstein was not so fortunate, predating the ability to diagnose - never mind treat - his disorders. Newton (paranoid schizophrenic, possibly bipolar as well) is a mixed bag - his temper tantrums and disorderly conduct are generally forgiven as the consequences of his condition(s), but neither the conditions he had nor any other conditions known to exist can be used to excuse him of the other charges. Paranoia may make you suspicious of others, but it doesn't make you rob them.
Yau - well, he hasn't the excuse of Einstein or Newton, even as far as such excuses went. If he has a personality disorder, he is electing - entirely of his own free will - not to treat it. Such defects are no longer an excuse, they have become a choice (at least, if you have money) and so cannot be used to justify behaviour. Nobody of Yau's stature can claim ignorance or lack of resources. The alternative - that there is no defect to treat and his actions are concious and deliberate - hardly exonerates him.
The situation here, though, is more complex I believe. Shing-Tung Yau seems to have deliberately turned mathematics from a relatively peaceful subject into a pieceful one, and Perelman seems utterly uninterested in having his emotions mauled for the sole purpose of gratifying a glory-seeker.
I would consider Yau's attitude, if the New Yorker piece is accurate, to be academic fraud, plagarism and the wilfull falsifying of results - any of which are severe enough in academia to warrant the nullification of previous awards, even if these took place afterwards. There have, in fact, been cases where doctorates have been revoked by the awarding University in England as a result of later scholarly abuses. They are certainly sufficiently serious that any professional mathematical society to which Yau belongs should investigate matters for possible disciplinary action should they be true.
(Sure, you can't do much. The military can court-martial, the Government can haul you off to Gitmo, but the mathematician's union is a little more limited. They could probably ban him from conferences they specifically held, and they could probably lean on journals to be more careful in refereeing his work, but that's about it. Well, actually, given his ego, they could probably take out an ad in a major Chinese newspaper, satirizing him. That could probably do him more damage than any official action.)
Personally, I think the Fields medal should have been awarded to Perelman specifically BECAUSE he refused it. They can't make him accept it - but that's what Swiss bank safety deposit boxes are for. On the other hand, they need to make it clear - to him and to everyone else - that mathematics is about truth, and truth has nothing to do with who accepts what. If a proof is correct, then it is correct and that is the end of the matter. Neither politics nor personalities have any say in it.
Furthermore, yes, it does turn him somewhat into a figurehead. And which would YOU prefer to be the role-model for all future mathematicians - the egomaniac or the gentleman? I'd argue that the sciences (and I include maths as a science) need to emphasise honesty, integrity and quality. Most here are computer programmers, or at least familiar with programming, so it would perhaps make sense to liken this to code. Would you rather a program work right (even if it's hard to understand), or be broken and/or stolen (even if it's made easy)? (I'll let you pick which OS' I am referring to, and which one I believe to be inherently superior.)
Perelman's proofs might be "high magic" in the coder's sense of being so hard very few (to none) can understand it, but I fail to see why that should be a problem. If anything, it is proof of the quality of his intellect and instinct. Those who reject that which they cannot understand are superstitious peasents. (Ok, that's a bit of a troll, but it's also true. You cannot learn that which you already understand, so it is only by not understanding that you are capable of learning. Thus, only the intelligent admit ignorance and only the ignorant claim certainty.)
Yau has claimed that he does not understand the proof. So where does the problem lie - with pto proof or Yau? Well, obviously Yau. If the problem was the proof, then Yau could establish where the error was that resulted in the proof being nonsense. The inability to establish such a proof does not mean that Perelman's work is perfect, only that it is beyond Yau to make any claims about it whatsoever. Were I to write a flawless program in raw assembly, would flaws magically apear if someone who could not read assembly state that it was incomprehensible to them? That would be stupid.
This entire dispute cuts to the heart of ALL theoretical and practical sciences and SHOULD be examined in depth by all official bodies with any degree of say in the matter. Perelman should NOT be permitted to walk away and play victim. If he is a victim of academic fraud, then academia has a responsib
I have repeatedly come to the conclusion that all of the software, all of the tools, all of the drivers, all of the everything needed for a quality user experience exists under Linux, but are NOT included in Linux distributions in a way that is trivial.
However, it should not be hard to roll a distribution that supports the devices out there. This might include commercial add-ons for Linux (the commercial version of CUPS adds a LOT of printers to the database, for example), but a hardened, newbie-tolerent, desktop-friendly Linux distribution could be done today.
It would take a bit of time to assemble, a lot of disk space and bandwidth I don't have to distribute it, and some sort of investment to be able to get the utilities out there. If these were all done tomorrow, I could have something ready for newbies within a week that really was closer to "plug and play" than "plug and pray".
As things stand, I don't see enough interest in such a product, or any real chance of the investment needed to fund the compilation. Everything's there, everything's done, but the only way to get past the "some assembly required" stage is for someone to do the assembly. That's not cheap, but it isn't difficult. What's lacking is the green stuff in the wallet, the grey stuff in the heads of programers has already been applied.
It's expedient - if it works, but how often does it work? Companies were trying to produce "legal" DVD players for Linux for a long time, but it was DeCSS that cracked the market. Cryptography was the Royal Preserve of RSA for a long time, and it was only with the re-implementing of RC4 and (especially) RSA's piblic key encryption algorithm that we started seeing crypto products in mainstream use - the European PGP implementation, for example.
I'm not advocating a gigantic crime ring to bust everyone's IP into the open - although reverse-engineering is supposed to be legal in many places in part to prevent illegal leveraging of technological monopolies in other markets. It has also got a track record of working.
Waiting for corporations to supply, even when paid, would very likely be a waste of time. There are many closed-source drivers out there which are rarely (if ever) updated, for commercial products those vendors sell. By not updating, those vendors are voluntarily choosing to put doing nothing over potential earnings. In the case of ARINC drivers, the hardware can easily cost a few thousand dollars for what is little more than an RS-232 device with an aviation rating. Why should they be any more active if you paid them? In fact, how much would you need to pay them before they'd notice? A few hundred thousand, to be economically viable?
For a couple of hundred thousand dollars, you could have a full-time staff dedicated to nothing but writing Linux drivers in one of a number of IP "safe havens". You could then guarantee the drivers would work with the latest kernel, work on all applicable platforms, be bug-free and be open-source. And, personally, I'd rather have a dedicated team of Linux gurus writing such code than have a couple of part-time hacks who really know nothing about Linux but are being paid by a corporate warlord to churn out something that is good enough for the company to keep the cheque.
If the Linux community could pool together enough cash to buy up-to-date usable drivers of any quality, then they could pool together enough cash to pay someone to write vastly superior drivers for a vastly wider range of Linux platforms including the most recent.
Well, the International Space Station is currently incapable of performing science, is due to be retired before it is due to be complete, and the length of time it has been in space is roughly the same as the expected mean time between significant meteorite strikes on an object of that size in that orbit. Oh, and bits are being held on with duct tape. A scrap metal merchant might offer $10 for it.
The idea that one can obtain information indirectly (messages before/after the encrypted message, the length of message, timing, etc) has been around for some time and is generally accepted as valid. It's one reason I don't altogether trust websites that use SSL only for specific information, as then it's trivial to figure out the format of that information and some of the characteristics, which must surely weaken the protection. The speed penalty of just encrypting everything would likely be less than the cost penalty of compromised security, particularly for some of the major vendors.
My father worked in military communications just prior to the Turkish invasion of Cyprus. Based on the few comments he has made on the subject - and it's harder to get information from him than from a Microsoft manual - the British used one-time pads on tape. Both tapes would be generated in parallel, one would then be delivered by courier, and each tape would have enough garbage at the front to guarantee the tapes were synchronised. (They could not be aligned simply on the Nth word on the tape, the tape machines themselves had to be configured to work at identical speeds, as there was no clock to synchronise with and it was intended to work on a lossy network.) Once the tapes were used, they were suppsed to be destroyed, though it wasn't completely clear to me if that always happened.
This was only one of several cipher machines they used, but my impression was that the bulk of encrypted information went via one-time pads despite the problems with getting the tapes to the other person and despite the enormous overheads of retuning everything in the system.
A lot of encryption schemes use XOR, some use XOR exclusively (which makes sense, as it is an exclusive or, after all), but you are correct that XOR is not an encryption method per-se. It is a mathematical operation that can be used to encrypt, but that is all.
It is arguable that if you had a one-time pad system (which is provably totally secure so long as the key is secure and used no more than once) where each bit/byte of the message was XORed with a totally random bit/byte from start to finish and where (since it's a one-time pad) the total input stream for the message was less than or equal to the length of the random pad it was XORed with, that you could refer to this as "encryption by means of XOR". It is also entirely reasonable, particularly if the phrase is an accepted shorthand amongst a particular group, for this to be shortened to "XOR encryption", although that is not strictly correct. Shorthand rarely is.
Newcomers to encryption may misunderstand such shorthand, or may misunderstand the OTP methodology, or may have examined very common encryption algorithms on the Internet and developed the term themselves on the basis of what they see. Such newcomers deserve commendation for being observent enough to get to that point in the first place, then forwarded to the information needed to complete their understanding. DON'T tell me you've never misused terminology in a subject you were new to, and DON'T tell me you wouldn't profit in such cases by getting forwarded to the necessary information.
So we've experts using shortcuts and confused newcomers, are there other categories? Well, yes, there are the idiots. There are a lot of these, as can be seen from the very large number of methods marked as fatally flawed on the block crypto lounge and hash function lounge. And those only list the better algorithms. The number of unlisted (even if open source) algorithms is an order or two magnitude greater, but nobody bothers to look at them beyond the most trivial of analysis because it's so obvious that the approach is utterly flawed from beginning to end.
Unfortunately, a lot of idiots get to produce professional crypto products (the Quartz public-key system is one example) so it's not even possible to go by appearances.
It would be hard for Microsoft to claim that they were taking Open Source seriously if they had their "open" license officially skewered by a license review board.
Since Microsoft aren't required to use any specific license for any specific product, and since a product can be "open source" and under multiple licenses that include non-open ones (eg: the Windows version of the Qt toolkit, for the longest time) and "non-free" ones (virtually all other dual-licensing cases), the argument that Microsoft would be limited in some way obviously doesn't apply.
Furthermore, since if the license HAD been deemed acceptable by the OSI, Microsoft could have claimed a massive PR coup over competitors who claim it is hostile to opening up code in any kind of meaningful way (as per EU lawsuit), it follows that Microsoft either did not want to claim such openness (which would make no sense, given how busy they are in claiming that they are perfectly open enough) OR they feared a bad PR reaction in the event of a rejection OR the license is deliberately intended NOT to be OSI-compliant but is intended to confer that impression to those corporations, countries, continents, etc, that have demanded greater openness in the software industry.
Those Slashdotters living in Europe whose European MPs are sympathetic to Open Source (or hostile to Microsoft - works in either case) may wish to notify their EuroMPs of Microsoft's request that the OSI not look closely at their terms and conditions. At worst, the EuroMP won't do anything, so it can't hurt. At best, it may be very useful ammunition the EU can use against Microsoft in Microsoft's tedious appeals - it's going to be harder for Microsoft to claim compliance with the EU court ruling on providing technical information for analysis if they won't even provide the license for the technical information to be analysed and prohibit the analysis even when the license is provided anyway.
(Hey, the EU does a lot of crappy things, makes strange drug-induced choices, and is rife with power-plays and nefarious dealings. The least we can do is exploit all that so that one of the few sensible rulings is actually upheld.)
...he'd shown that the Fields Medal award system had no holes and was therefore topologically equivalent to a sphere, as per the Poincare Hypothesis that he had proved earlier. However, as he had proved that since cows are spherical, it follows that the medal must also be topologically the same as a cow, since the transform works both ways. A cow can weight 850 lbs. Clearly this is utterly impractical to carry, never mind wear as a medal. Besides which, it would obviously exceed the maximum weight for carry-on luggage.
Seriously, I do like it when people offer genuine, interesting and downright challenging scenarios, and I'd certainly place yours in that category. Why? Because it's those that define the really interesting boundary conditions, the situations which separate those definitions which will stand the test of time and those which won't.
(Often, scientists like to cheat a little and define small, well-behaved regions in isolation. An example would be Hooke's Law with springs, which is a crude approximation that is only valid over a narrow range but works "well enough" within that. Another would be the tustle between Quantum Mechanics - only valid for the microscopic, Relativity - only valid for the macroscopic or near light-speed, and Newton's Laws - the only useful way of modelling human-scale phenomena. It's only by understanding what happens at the boundaries, when two or more models produce the right answer for provably the wrong reasons or where no simple model works at all, that you get the really interesting discoveries, the real unification of ideas.)
Ok, your examples are obviously sufficient to prove that what an object orbits is irrelevent. We now have to decide what is important.
I suggested on K5 that what it forms from is what's important. A planetary accretion disk will have a composition dependent on the planet, a stellar accretion disk will have a composition dependent on the stellar nursary. A body that forms at a distance D from a planet out of a planetary accretion disk should contain a far narrower range of elements than a body that forms from a stellar accretion disk at the same distance, and the total span of elements should also be far narrower. This means that bodies that form in planetary accretion disks should be far more uniform than in a solar system with a comparable number of bodies within it. Also, because you have different sources for the material, because the spacing determines the relative abundances, and because the gravitational fields are going to be very different, the composition of a body B should be unique to a specific orbit and an accretion disk of specific type. No other possible combination of starting conditions should be able to produce the same overall composition.
To me, that says that the composition is a key factor. It can pinpoint exactly what sort of body the object formed around and at what distance. Ok, what other key factors are there?
The internal structure (eg: the number of cores) identifies the WAY it formed. It's not totally unique, as bodies will form and unform in an early system, but certain generalities exist. A body with no core or one core is probably an object that has accumulated mass over time. An object with multiple cores was, at one time, multiple objects of comparable size that have adhesed together but are still distinct. The structure also limits the timeframe over which it formed. An object with no core probably formed quite quickly - no strata and no migration. An object with one or more cores probably formed over a much longer time.
So let's go back to your example of a body orbiting three stars in the way described - and such systems are probably quite common, as stars tend to form in twos or threes, so it's a very useful test to apply. Well, here we have a problem. My definition won't work, as the scenario doesn't tell me what the body is made of, where it formed, how it formed, what its structure is, etc. As such, my definition is clearly of no value for pure observation. It's totally the wrong perspective. (In the same way, since e=mc^2, and since specific heat is measured as joules per mole per kelvin, it would be entirely correct to state your weight in degrees fahrenheit. One can be directly converted to the other, so neither is "wrong". Which is useful depends on what the weight is needed for.)
Your definition is good for observational purposes, in that it would allow an astronomer 100 light-years away to tell if an object is a planet or not. 100 l
Not sure it needed verbatim cut & paste here, though. I shouldn't complain too much, having done the same thing a few times myself. As I posted there - and have posted here several times - I do not believe in external "symptomatic" definitions. Those identify the object by effects rather than causes, and by appearance rather than mechanics. I have offered here, there (and just about everywhere) a definition scheme which is extensible, flexible and structural, which allows you to categorize planets, stars, asteroids, comets, black holes and just about any other phenomena you are likely to encounter as a simple set of tests that all observers can universally agree upon.
I am firmly of the belief that, in a subject like Astronomy, isolated definitions which have no broader context and no reason that can be backed by some combination of pure maths and/or pure physics are fine as personal labels but useless as categories in a science.
In science, you produce a hypothesis that can be tested in a manner that is repeatable by anyone. A test for gravitational rounding is problematic, as it is a continuum and not an on/off switch. A test of orbiting a star is just as problematic, as it immediately prohibits planets that reach escape velocity from remaining planets. The object hasn't changed, only the relative velocity. If such a planet is recaptured by another star, without slowing down, then it becomes a planet again. This time, all that has changed has been the force exerted by another object entirely.
When the definition of A depends on the characteristics of B, you've a lousy definition. A cat remains a cat, even if a dog walks down a street backwards.
The IAU's definition, then, does NOT meet any quantifiable standard whatsoever, it is merely political window-dressing. The choice of Pluton, on the grounds that it wasn't in a spell-checker, is further evidence that the IAU themselves never took the definition seriously. Do you think the IAU would consider it acceptable for an astronomer to claim something was peer-reviewed, if Grammatik or some other grammer tool didn't complain? Of course not! If it's not a refereed paper in a credible journal, then they would use the astronomer's guts for fishing bait. Then why accept any less from the IAU themselves?
It is a pity that there has been so little adverse reaction to the decision, and that those of us who are opposed to the new definition have no voice in the issue. It is not acceptable, IMHO, for a major international organization to behave in a manner that is clearly harmful to the credibility of astronomy in particular but also the scientific community in general. I doubt they'll be ashamed of themselves - I doubt they'll even care until they run into so many contradictions that the definition is totally untennable, and then they'll whine that they could not possibly have forseen it happening, and how were they to know?
That's one reason I do it - another, related, reason is that invariably someone at work will want to do something outrageously strange but not know what options exist. A quick search through the installed documentation files will usually tell me exactly what's needed, what the commandline options are, etc, ad nausium. Alternatively, and this has happened a few times, someone who knows I'm a computer geek will tell me about their latest pet project which invariably involves something they have no idea how to do but they do know that it can be done by computer. I can flat-out guarantee it'll either require something I installed in the last 48 hours -or- something I regularly update the Freshmeat record for.
If you start stationary, then apply a force such as to move around a circle, you will end up with kinetic energy. What an interesting observation.
My greater fear is that this will lead to both the filching of investors and the feeding stereotypes amongst the less tolerent of Britain - a far more likely outcome than the elimination of energy costs, methinks.
So because the dolphin isn't brainless enough to jump out of its tank and beach itself and die in the process, that makes them stupid? I suppose by comparison the child that plays away from road isn't as smart as the kid that plays in traffic, you know, the one that's seeking to "enlarge his environment" by becoming road pizza.
A far more interesting behaviour is when dolphins deliberately chase fish up onto river banks to die. The dolphins then use the river banks as a gigantic seafood buffet, picking what fish they want before rolling/sliding back into the water. So they certainly do make very good use of the beach...
Slashdot Mirror
For Windows, omit the word "either" and replace "or" with "and".
How does this work? Well, DRM costs money to make and to maintain. It takes resources away from the DRM'ed device, reducing the amount of product you are capable of selling. However, the DRM has no intrinsic value in itself. Thus, it must reduce point-of-sale profits and it must increase the costs of producing the product in the first place. (Programmers and hardware engineers with the expertise to produce any meaningful DRM won't come cheap.) It also means you are taking staff away from producing more material, thus reducing what you have to sell, which will also reduce profits.
How does the Prisoner's Dilemma fit in to this? Well, you want to have stronger DRM than competitors, so the pirates will target them and not you. At the same time, you have to be working with them and with hardware vendors, so that your DRM is still readable on other people's systems. Unreadability kills sales - just ask Sony's Betamax division. Oh, and at the same time as providing all this information, you can't provide so much that an outsider can break your DRM scheme.
The solution? There isn't one. The constraints are mutually exclusive. You cannot simultaneously be open and closed, different and the same, restricted and unrestricted. This is the problem that corporations have faced with tape copying, floppy disk copying, or the copying of any physical medium whatsoever. The moment you aim for one ideal, you eliminate any possibility of achieving another, and all of the ideals surrounding DRM are contrary to the ideals surrounding sales.
So how do we deal with piracy, if not with DRM? Not having DRM will maximize sales of those who are interested, but piracy has the potential for reducing that number. (I'm playing devil's advocate here, as there is no data to support that claim, but as it is widely believed, you may as well have an argument that works with those who believe it.)
The first thing you need to consider is shelflife. Most products really have a very short shelflife. Sure, there are some classics, but there really aren't many of them. There are far, far fewer old products that still sell than there are new products that sell. Furthermore, there are even fewer old products that sell AS WELL as new products, even after taking into account that no development needs to be done on something that has been around forever and a day.
Widespread piracy takes time. Most distribution channels are slow, error-prone and unreliable. They work - they work very well - but only over a long timeframe. Over a short timeframe, they are often pretty useless. Most don't scale well and rely on self-propagation to get any kind of performance. This doesn't happen overnight... unless linked to from Slashdot.
What is wanted, then, is increased turnover, shorter software lifecycles, less bundling, more fluidity. Instead of slowing the pirates down, speed the market up. It's all about relative performance, so the two have the same effect on how many will be interested in buying. What does this do to the bottom line, though? Well, now you eliminate the Dilemma altogether. By relying on high dexterity, rather than heavy armour, you don't need any extended cooperation from others, you are limited only by what the media readers are capable of. Nothing else.
What does this do to the division of labour? Well, those who would have been involved in PRing the DRM can now be selling the product. Those involved in researching, developing and implementing the DRM can now be making new products. You have the same number of people
The 2.5x case comes from two simple rules of thumb. Firstly, you need enough to be able to hold the whole of what is in RAM now, plus everything you want to swap in, plus enough to minimise fragmentation and cover overheads. Secondly, the more swap space you have, the more metadata you need to manage it AND the greater the latency to perform any kind of swap AND the more swapping you need to do to run all active processes. Too much virtual memory is a Bad Thing. Having 2.5 x RAM was considered a good compromise and it is one I use to this day.
Today, both rules of thumb still hold. The largest single object you can have is one that fills ALL of RAM after the kernel, and you absolutely must have sufficient swap space to be able to dump that object to disk. If you don't, then the kernel will either panic, kill the process or cause any other activity to behave unpredictably. It won't have the resources to behave correctly. Any number of these objects could, in theory, be swapped out - but remember that they don't run when on disk, only when in memory, so the more you have, the smaller the timeslice each will get - and the sum of those timeslices will go down, as you need to allow time for the swap to take place.
However, today isn't quite the same as yesterday. The difference in performance between hard drives and RAM has changed. There is better caching on the drive. The swap algorithms are smarter and there is more understanding of what metadata is useful and what really has no value. Process handling is also smarter, so processes aren't necessarily run in order - round-robin scheduling is used for some time-critical stuff on Linux, but most applications use a more relaxed system.
Also, programming has changed. There is greater re-use of tools and libraries - well, sometimes - and this means that the largest object you really have to handle at a time is much smaller than the size of RAM. A certain fraction of what's left will be used by shared libraries and shared resources.
Lastly, because hard drives are reasonably cheap and most PCs can handle several at the same time, you are far far better off getting a drive and dedicating it to swap. This is good for many reasons, not least because the drive won't have to move the read heads from data space to swap space and back. You eliminate a vast chunk of seek time, reduce the stress on the drive AND can experiment with different swap sizes without risking losing data.
I would therefore STRONGLY advise using the classic 2.5x and a different hard drive, but if you can't do this for some reason and want an updated formula, here is what I would suggest:
The meaningful RAM will be equal to the total RAM minus the space used by the kernel and vital, non-swappable resources/daemons. Multiply this by three for 7200RPM hard drives or by five for 15000RPM hard drives. Multiply by one and a quarter for basic swap schemes, or by one and an eighth for profiling/intelligent swap schemes. Add the size of the hard disk cache, if the cache uses a high water mark to control operations. Subtract the size of the hard drive cache (unless this takes the size below zero) if the behavior is controlled by a low water mark only. Add one megabyte per simultaneous user. Add one megabyte for each large -or- long-running application likely to be running simultaneously. Subtract the total size of all the shared libraries likely to be loaded in the case just considered.
This is a LOT more complex than 2.5x, so much so that I generally wouldn't bother using it except
"We would like an OS that has the following characteristics:
Linux does 2, most of 1 and 3 and some of 4 and 5. Plan 9 seems to be working on other aspects of 4 and 5. In theory, mandatory access controls should provide 6, as automagic functions would be tightly defined, but MAC is not well-understood on general-purposes operating systems to the point where the Fedora defaults of SELinux are almost useless.
I'll go through these cases, as the above list isn't entirely clear:
No OS in the world is capable of providin
A proposal I made many many years back was to run a league table, where so many points were scored for the release of an open source product, so many points were scored for updating someone else's open source product, so many for closed source products that supported or enhanced the usability of an open source product, etc. The more open source, the more points. Also, the more significant (in terms of power, flexibility, etc) the more points, and the more practical the more points.
The idea was to show who actually was doing work, versus who was merely bragging about it. The idea was also to make major Open Source figures (some of whom have never actually released a product but have contributed to many) just as prominent in the table as major corporations with Really Big Bucks to throw around. It would also show those who are working on making Open Source a key player in the computing world, even if their products are not themselves Open Source.
(Oracle would score points for having put their corporate database on Linux, for example, but it would not be as much as Computer Associates for putting their corporate database - Ingres - not only on Linux but opening up the source as well. Postgresql would score more yet, as it is not only Open Source but regularly maintained.)
Newcomers are at no disadvantage, because whatever REAL added value will show up notwithstanding the newness or the lack of awareness. If there's no added value, then there is nothing there to watch. It's merely a rebadge. If there's added value and this value is constantly added to (which is what a new company should be doing), then it will be a very obvious rapid-riser through the charts.
Proprietary vendors who are wary of opening their high-value major product lines will obviously not score as well, nor should they, but they will be represented as a function of what they have contributed - directly in terms of products, and indirectly in terms of improved usability.
Then, journalists MIGHT have a clue as to what is interesting and what is not. They might also have a clue as to what is significant, what is likely to become "big news" and what is worth the effort of covering.
As it stands, they neither know nor care. Whoever pays the piper calls the tune. Mind you, what I'm aiming for, with this idea, is not to introduce honesty, but rather to give people a copy of the songbook. Let people see for themselves if the tune is any good or is even what it's claimed to be. It would seem to me that an informed userbase will take care of the honesty issue by itself.
Tiger is another good hash function - faster than Whirlpool, smaller for those embedded cases where even the bytes matter, and I believe it is not known to have any attacks against it. Tiger also appears in mhash, not certain if it's in the kernel but it should be.
I don't see that it is really of any consequence whether anyone has actually demonstrated an attack on SHA - the point of security is to not wait until AFTER the house has been plundered to upgrade. SHA is FIPS-180, but if there is even a theory on how it might become broken, I would urge people to use something stronger. Security that is only certain to be good against skript kiddies is really not very useful security.
So, don't imagine that this is going to be an easy one. Open Source projects by IBM might be easier to get past the Great And All-Seeing (but definitely not all-knowing) Pointy Haired Bosses - at least some will be familiar with the phrase "nobody ever got sacked for buying IBM" and may even still believe that. Some of the Apache projects might be workable, provided you use the line of reasoning that since Apache is listed on IBM's website as a project they are working on, it is covered by the "nobody ever got sacked for buying IBM". You don't have to tell them that IBM is only one member of a large consortium, and it might be better not to.
Some projects were connected to IBM and other major corporations but are now independent. Postfix is an example of that. I believe evlog (Enterprise-grade event logging for Linux) is also such a project. Speaking of evlog, I would DEFINITELY suggest using it in any commercial or Government setting. It's not that good and Linux has plenty of other security, but "Enterprise-grade logging" is mandatory in many cases and this provides it. It ticks the right box, even if it doesn't do a whole lot more for you. It's a pure CYA and nothing more.
ISO 9000 (or later) compliance is probably the toughest requirement, as it stipulates documenting the process and activities, where the level of documentation depends on how critical the project is, and Open Source projects have neither that type of documentation or any real concept of criticalness as components are freely reusable. Your best bet is to work through vendors that are themselves ISO 900x certified AND supply either the Open Source OR the links to those projects, then argue that by documenting the use of a project that comes from an ISO 900x certified source, you inherit the certification indirectly. Some bosses will buy that easier than others and depending on the structure of the organization, you may have flexibility on who you present the case to. If so, shop.
It's probably no worse than the extreme conditions it endures in space or during re-entry, but it sure looks nasty.
Nobody but a wilfull participant in intellectual fraud would - today - contend that he discovered the laws of motion (he plagarised them off Descartes) or calculus (Gottfried Leibniz produced the modern version and published before Newton, Archimedes was the original discoverer although it is very unlikely either Newton or Leibniz was aware of the earlier work).
Nobody but a fool would deny that Newton's pressure to obliterate the wave theory of light set optical science back many centuries, and that his hostility towards others caused serious enough rifts that collaboration between Europe and Britain, or between British scientists, was seriously inhibited for a long time after his death, retarding science in general.
Does that make Newton any less of a great scientist? No. Does it devalue the Principa? No. So why, if this holds true of Newton can it not hold true of Yau? Why, if we can accept that Newton had serious character flaws that led to intellectual fraud, forgery, deceit and political manipulation for his own ends, can we not accept that an equally brilliant but equally troubled intellect would suffer the same flaws?
Furthermore, since a genius HAS exceptional intelligence, should we not hold them to a higher standard of conduct, rather than a lesser one? Particularly in modern times, where most such flaws are easily treatable?
(Incidently, since I am drawing parallels with Newton, does this mean I believe Newton should have his knighthood revoked? Actually, yes. He has no business holding credit where that credit is not due but was acquired through malpractice.)
As for accessibility - that is of no consequence to the matter at hand, which is a matter of who discovered what. Clarifying is not the same as discovering. Clarifying is a continuous process that starts from the moment of discovery and will never stop. Original discovery only happens once - otherwise it ceases to be original - and the issue at hand is the originality of the discovery, not whether it is readable.
Sure, "human nature" is inevitable, being humans. However, that begs the question of what is human nature. Most philosophers over time seem to agree that character flaws and delusions are largely optional. (This does not include chemical imbalances, which are optional today only insofar as treatments exist.) Professor Nash, for example, was fortunate. He suffers from schizophrenia but when he was diagnosed, treatments did exist. Not very good ones, sure, but they worked. Einstein was not so fortunate, predating the ability to diagnose - never mind treat - his disorders. Newton (paranoid schizophrenic, possibly bipolar as well) is a mixed bag - his temper tantrums and disorderly conduct are generally forgiven as the consequences of his condition(s), but neither the conditions he had nor any other conditions known to exist can be used to excuse him of the other charges. Paranoia may make you suspicious of others, but it doesn't make you rob them.
Yau - well, he hasn't the excuse of Einstein or Newton, even as far as such excuses went. If he has a personality disorder, he is electing - entirely of his own free will - not to treat it. Such defects are no longer an excuse, they have become a choice (at least, if you have money) and so cannot be used to justify behaviour. Nobody of Yau's stature can claim ignorance or lack of resources. The alternative - that there is no defect to treat and his actions are concious and deliberate - hardly exonerates him.
I would consider Yau's attitude, if the New Yorker piece is accurate, to be academic fraud, plagarism and the wilfull falsifying of results - any of which are severe enough in academia to warrant the nullification of previous awards, even if these took place afterwards. There have, in fact, been cases where doctorates have been revoked by the awarding University in England as a result of later scholarly abuses. They are certainly sufficiently serious that any professional mathematical society to which Yau belongs should investigate matters for possible disciplinary action should they be true.
(Sure, you can't do much. The military can court-martial, the Government can haul you off to Gitmo, but the mathematician's union is a little more limited. They could probably ban him from conferences they specifically held, and they could probably lean on journals to be more careful in refereeing his work, but that's about it. Well, actually, given his ego, they could probably take out an ad in a major Chinese newspaper, satirizing him. That could probably do him more damage than any official action.)
Personally, I think the Fields medal should have been awarded to Perelman specifically BECAUSE he refused it. They can't make him accept it - but that's what Swiss bank safety deposit boxes are for. On the other hand, they need to make it clear - to him and to everyone else - that mathematics is about truth, and truth has nothing to do with who accepts what. If a proof is correct, then it is correct and that is the end of the matter. Neither politics nor personalities have any say in it.
Furthermore, yes, it does turn him somewhat into a figurehead. And which would YOU prefer to be the role-model for all future mathematicians - the egomaniac or the gentleman? I'd argue that the sciences (and I include maths as a science) need to emphasise honesty, integrity and quality. Most here are computer programmers, or at least familiar with programming, so it would perhaps make sense to liken this to code. Would you rather a program work right (even if it's hard to understand), or be broken and/or stolen (even if it's made easy)? (I'll let you pick which OS' I am referring to, and which one I believe to be inherently superior.)
Perelman's proofs might be "high magic" in the coder's sense of being so hard very few (to none) can understand it, but I fail to see why that should be a problem. If anything, it is proof of the quality of his intellect and instinct. Those who reject that which they cannot understand are superstitious peasents. (Ok, that's a bit of a troll, but it's also true. You cannot learn that which you already understand, so it is only by not understanding that you are capable of learning. Thus, only the intelligent admit ignorance and only the ignorant claim certainty.)
Yau has claimed that he does not understand the proof. So where does the problem lie - with pto proof or Yau? Well, obviously Yau. If the problem was the proof, then Yau could establish where the error was that resulted in the proof being nonsense. The inability to establish such a proof does not mean that Perelman's work is perfect, only that it is beyond Yau to make any claims about it whatsoever. Were I to write a flawless program in raw assembly, would flaws magically apear if someone who could not read assembly state that it was incomprehensible to them? That would be stupid.
This entire dispute cuts to the heart of ALL theoretical and practical sciences and SHOULD be examined in depth by all official bodies with any degree of say in the matter. Perelman should NOT be permitted to walk away and play victim. If he is a victim of academic fraud, then academia has a responsib
However, it should not be hard to roll a distribution that supports the devices out there. This might include commercial add-ons for Linux (the commercial version of CUPS adds a LOT of printers to the database, for example), but a hardened, newbie-tolerent, desktop-friendly Linux distribution could be done today.
It would take a bit of time to assemble, a lot of disk space and bandwidth I don't have to distribute it, and some sort of investment to be able to get the utilities out there. If these were all done tomorrow, I could have something ready for newbies within a week that really was closer to "plug and play" than "plug and pray".
As things stand, I don't see enough interest in such a product, or any real chance of the investment needed to fund the compilation. Everything's there, everything's done, but the only way to get past the "some assembly required" stage is for someone to do the assembly. That's not cheap, but it isn't difficult. What's lacking is the green stuff in the wallet, the grey stuff in the heads of programers has already been applied.
I'm not advocating a gigantic crime ring to bust everyone's IP into the open - although reverse-engineering is supposed to be legal in many places in part to prevent illegal leveraging of technological monopolies in other markets. It has also got a track record of working.
Waiting for corporations to supply, even when paid, would very likely be a waste of time. There are many closed-source drivers out there which are rarely (if ever) updated, for commercial products those vendors sell. By not updating, those vendors are voluntarily choosing to put doing nothing over potential earnings. In the case of ARINC drivers, the hardware can easily cost a few thousand dollars for what is little more than an RS-232 device with an aviation rating. Why should they be any more active if you paid them? In fact, how much would you need to pay them before they'd notice? A few hundred thousand, to be economically viable?
For a couple of hundred thousand dollars, you could have a full-time staff dedicated to nothing but writing Linux drivers in one of a number of IP "safe havens". You could then guarantee the drivers would work with the latest kernel, work on all applicable platforms, be bug-free and be open-source. And, personally, I'd rather have a dedicated team of Linux gurus writing such code than have a couple of part-time hacks who really know nothing about Linux but are being paid by a corporate warlord to churn out something that is good enough for the company to keep the cheque.
If the Linux community could pool together enough cash to buy up-to-date usable drivers of any quality, then they could pool together enough cash to pay someone to write vastly superior drivers for a vastly wider range of Linux platforms including the most recent.
I'm not sure hitting New Jersey would be good, though. Could you divert it so it hits Texas, or New Orleans?
Well, the International Space Station is currently incapable of performing science, is due to be retired before it is due to be complete, and the length of time it has been in space is roughly the same as the expected mean time between significant meteorite strikes on an object of that size in that orbit. Oh, and bits are being held on with duct tape. A scrap metal merchant might offer $10 for it.
My father worked in military communications just prior to the Turkish invasion of Cyprus. Based on the few comments he has made on the subject - and it's harder to get information from him than from a Microsoft manual - the British used one-time pads on tape. Both tapes would be generated in parallel, one would then be delivered by courier, and each tape would have enough garbage at the front to guarantee the tapes were synchronised. (They could not be aligned simply on the Nth word on the tape, the tape machines themselves had to be configured to work at identical speeds, as there was no clock to synchronise with and it was intended to work on a lossy network.) Once the tapes were used, they were suppsed to be destroyed, though it wasn't completely clear to me if that always happened.
This was only one of several cipher machines they used, but my impression was that the bulk of encrypted information went via one-time pads despite the problems with getting the tapes to the other person and despite the enormous overheads of retuning everything in the system.
It is arguable that if you had a one-time pad system (which is provably totally secure so long as the key is secure and used no more than once) where each bit/byte of the message was XORed with a totally random bit/byte from start to finish and where (since it's a one-time pad) the total input stream for the message was less than or equal to the length of the random pad it was XORed with, that you could refer to this as "encryption by means of XOR". It is also entirely reasonable, particularly if the phrase is an accepted shorthand amongst a particular group, for this to be shortened to "XOR encryption", although that is not strictly correct. Shorthand rarely is.
Newcomers to encryption may misunderstand such shorthand, or may misunderstand the OTP methodology, or may have examined very common encryption algorithms on the Internet and developed the term themselves on the basis of what they see. Such newcomers deserve commendation for being observent enough to get to that point in the first place, then forwarded to the information needed to complete their understanding. DON'T tell me you've never misused terminology in a subject you were new to, and DON'T tell me you wouldn't profit in such cases by getting forwarded to the necessary information.
So we've experts using shortcuts and confused newcomers, are there other categories? Well, yes, there are the idiots. There are a lot of these, as can be seen from the very large number of methods marked as fatally flawed on the block crypto lounge and hash function lounge. And those only list the better algorithms. The number of unlisted (even if open source) algorithms is an order or two magnitude greater, but nobody bothers to look at them beyond the most trivial of analysis because it's so obvious that the approach is utterly flawed from beginning to end.
Unfortunately, a lot of idiots get to produce professional crypto products (the Quartz public-key system is one example) so it's not even possible to go by appearances.
Since Microsoft aren't required to use any specific license for any specific product, and since a product can be "open source" and under multiple licenses that include non-open ones (eg: the Windows version of the Qt toolkit, for the longest time) and "non-free" ones (virtually all other dual-licensing cases), the argument that Microsoft would be limited in some way obviously doesn't apply.
Furthermore, since if the license HAD been deemed acceptable by the OSI, Microsoft could have claimed a massive PR coup over competitors who claim it is hostile to opening up code in any kind of meaningful way (as per EU lawsuit), it follows that Microsoft either did not want to claim such openness (which would make no sense, given how busy they are in claiming that they are perfectly open enough) OR they feared a bad PR reaction in the event of a rejection OR the license is deliberately intended NOT to be OSI-compliant but is intended to confer that impression to those corporations, countries, continents, etc, that have demanded greater openness in the software industry.
Those Slashdotters living in Europe whose European MPs are sympathetic to Open Source (or hostile to Microsoft - works in either case) may wish to notify their EuroMPs of Microsoft's request that the OSI not look closely at their terms and conditions. At worst, the EuroMP won't do anything, so it can't hurt. At best, it may be very useful ammunition the EU can use against Microsoft in Microsoft's tedious appeals - it's going to be harder for Microsoft to claim compliance with the EU court ruling on providing technical information for analysis if they won't even provide the license for the technical information to be analysed and prohibit the analysis even when the license is provided anyway.
(Hey, the EU does a lot of crappy things, makes strange drug-induced choices, and is rife with power-plays and nefarious dealings. The least we can do is exploit all that so that one of the few sensible rulings is actually upheld.)
Well, what did you expect, once they developed calculus?
...he'd shown that the Fields Medal award system had no holes and was therefore topologically equivalent to a sphere, as per the Poincare Hypothesis that he had proved earlier. However, as he had proved that since cows are spherical, it follows that the medal must also be topologically the same as a cow, since the transform works both ways. A cow can weight 850 lbs. Clearly this is utterly impractical to carry, never mind wear as a medal. Besides which, it would obviously exceed the maximum weight for carry-on luggage.
(Often, scientists like to cheat a little and define small, well-behaved regions in isolation. An example would be Hooke's Law with springs, which is a crude approximation that is only valid over a narrow range but works "well enough" within that. Another would be the tustle between Quantum Mechanics - only valid for the microscopic, Relativity - only valid for the macroscopic or near light-speed, and Newton's Laws - the only useful way of modelling human-scale phenomena. It's only by understanding what happens at the boundaries, when two or more models produce the right answer for provably the wrong reasons or where no simple model works at all, that you get the really interesting discoveries, the real unification of ideas.)
Ok, your examples are obviously sufficient to prove that what an object orbits is irrelevent. We now have to decide what is important.
I suggested on K5 that what it forms from is what's important. A planetary accretion disk will have a composition dependent on the planet, a stellar accretion disk will have a composition dependent on the stellar nursary. A body that forms at a distance D from a planet out of a planetary accretion disk should contain a far narrower range of elements than a body that forms from a stellar accretion disk at the same distance, and the total span of elements should also be far narrower. This means that bodies that form in planetary accretion disks should be far more uniform than in a solar system with a comparable number of bodies within it. Also, because you have different sources for the material, because the spacing determines the relative abundances, and because the gravitational fields are going to be very different, the composition of a body B should be unique to a specific orbit and an accretion disk of specific type. No other possible combination of starting conditions should be able to produce the same overall composition.
To me, that says that the composition is a key factor. It can pinpoint exactly what sort of body the object formed around and at what distance. Ok, what other key factors are there?
The internal structure (eg: the number of cores) identifies the WAY it formed. It's not totally unique, as bodies will form and unform in an early system, but certain generalities exist. A body with no core or one core is probably an object that has accumulated mass over time. An object with multiple cores was, at one time, multiple objects of comparable size that have adhesed together but are still distinct. The structure also limits the timeframe over which it formed. An object with no core probably formed quite quickly - no strata and no migration. An object with one or more cores probably formed over a much longer time.
So let's go back to your example of a body orbiting three stars in the way described - and such systems are probably quite common, as stars tend to form in twos or threes, so it's a very useful test to apply. Well, here we have a problem. My definition won't work, as the scenario doesn't tell me what the body is made of, where it formed, how it formed, what its structure is, etc. As such, my definition is clearly of no value for pure observation. It's totally the wrong perspective. (In the same way, since e=mc^2, and since specific heat is measured as joules per mole per kelvin, it would be entirely correct to state your weight in degrees fahrenheit. One can be directly converted to the other, so neither is "wrong". Which is useful depends on what the weight is needed for.)
Your definition is good for observational purposes, in that it would allow an astronomer 100 light-years away to tell if an object is a planet or not. 100 l
I am firmly of the belief that, in a subject like Astronomy, isolated definitions which have no broader context and no reason that can be backed by some combination of pure maths and/or pure physics are fine as personal labels but useless as categories in a science.
In science, you produce a hypothesis that can be tested in a manner that is repeatable by anyone. A test for gravitational rounding is problematic, as it is a continuum and not an on/off switch. A test of orbiting a star is just as problematic, as it immediately prohibits planets that reach escape velocity from remaining planets. The object hasn't changed, only the relative velocity. If such a planet is recaptured by another star, without slowing down, then it becomes a planet again. This time, all that has changed has been the force exerted by another object entirely.
When the definition of A depends on the characteristics of B, you've a lousy definition. A cat remains a cat, even if a dog walks down a street backwards.
The IAU's definition, then, does NOT meet any quantifiable standard whatsoever, it is merely political window-dressing. The choice of Pluton, on the grounds that it wasn't in a spell-checker, is further evidence that the IAU themselves never took the definition seriously. Do you think the IAU would consider it acceptable for an astronomer to claim something was peer-reviewed, if Grammatik or some other grammer tool didn't complain? Of course not! If it's not a refereed paper in a credible journal, then they would use the astronomer's guts for fishing bait. Then why accept any less from the IAU themselves?
It is a pity that there has been so little adverse reaction to the decision, and that those of us who are opposed to the new definition have no voice in the issue. It is not acceptable, IMHO, for a major international organization to behave in a manner that is clearly harmful to the credibility of astronomy in particular but also the scientific community in general. I doubt they'll be ashamed of themselves - I doubt they'll even care until they run into so many contradictions that the definition is totally untennable, and then they'll whine that they could not possibly have forseen it happening, and how were they to know?
Whimpering? Oh! I thought that was a sound effect from the hard drive. After digitizing and amplifying it, though....
That's one reason I do it - another, related, reason is that invariably someone at work will want to do something outrageously strange but not know what options exist. A quick search through the installed documentation files will usually tell me exactly what's needed, what the commandline options are, etc, ad nausium. Alternatively, and this has happened a few times, someone who knows I'm a computer geek will tell me about their latest pet project which invariably involves something they have no idea how to do but they do know that it can be done by computer. I can flat-out guarantee it'll either require something I installed in the last 48 hours -or- something I regularly update the Freshmeat record for.
My greater fear is that this will lead to both the filching of investors and the feeding stereotypes amongst the less tolerent of Britain - a far more likely outcome than the elimination of energy costs, methinks.
A far more interesting behaviour is when dolphins deliberately chase fish up onto river banks to die. The dolphins then use the river banks as a gigantic seafood buffet, picking what fish they want before rolling/sliding back into the water. So they certainly do make very good use of the beach...