It's possible the gov't patented this so they could share the information with other people and not worry about some private company patenting the idea and then sueing everyone else for us it. Basically - patent to allow people to use it.
If that's the intent, and the patent system is working as intended, then the patenting is superfluous. Publication of all the details (without restriction) is sufficient to prevent anyone else from patenting the idea, because the publication acts as demonstrable prior art with which to challenge any subsequent patent application. (This is also why anyone who wants to patent something usually has to hold off on publication until the patent process is already underway--a publication can be used to show that the idea had already become "common knowledge" before the patent.)
Of course, in reality publishing details is usually not enough to prevent someone else from patenting, because it seems that no one (least of all patent examiners) does a thorough job of uncovering prior art. So I guess you could argue that they were being extra-careful and strategic, making sure to patent it to so that they could liberate the idea without any chance of someone else claiming it.
But if it really were their intention to liberate the idea for all to use, one would think that any press release mentioning the patent would also mention the associated "perpetual, worldwide, royalty-free license to use these patented methods" that the NSA is offering. I see no such offer.
So really what you think your doing in boycotting them is just being stupid and showing your lack of knowledge of the situation. Because Apple really isnt in the business to DRM crap...
It may not be Apple's "fault" that there is DRM. Maybe they don't want it to be there--maybe it's entirely the labels making that call, and Apple has no choice, if they want to sell those tracks.
But that doesn't mean that boycotting Apple is a mistake. The point of a boycott is to (1) draw attention to a cause; and (2) make it financially difficult for the corporation to continue their practice. If a boycott of Apple products were sufficiently large and successful, then the labels would take notice (since they would be selling fewer tracks, as compared to non-DRM music), and Apple would take notice. Apple may not have the final say about DRM, but they are in a bargaining position. If iTunes DRM was causing them to lose sales in other divisions, they might very well negotiate aggressively with the labels to offer more DRM-free tracks. The only way to make a corporation "care" is to make it about money. That's the point of a boycott.
Having said all that, I doubt the presently proposed boycott will have a large effect. Nor am I convinced that this is the best way to eliminate DRM. But I don't like arguments against the boycott that amount to "It's not Apple's fault--they don't want DRM either!" (which is irrelevant--they offer DRM'ed products, so they fall under the boycott); or "Why bother?" (which is bothersome--should people who care about an issue do nothing?); or even "Every company does it!" (fine--boycott them also).
lawsuits seem to be their main source of income these days
Only they know for sure, of course... but Ray's best estimate is that they are losing money overall: they may make money off settlements (or maybe just break even?), but the legal and investigative costs that go into all those dropped or default judgment cases destroys all that profit, and then some.
I don't think it's actually profitable for them. I think it was always intended as some kind of "campaign of fear" to make people too afraid to file-share... which seems not to have worked.
... the whole concept of touch screens for office use is fundementally flawed. It requires you to have your arms raised to operate the screen - which is an unnatural position and very tiring to do for long periods of time.
Agreed. But this "see-through" demo shows that an innovation can potentially fix what at first seems to be a major problem. (Of course I won't know if this "see-through" idea is really a valid solution until I have a chance to play with it myself...).
In the case of touchscreen technology, I agree that pressing on a vertical screen would get tiring real quick. And the sometimes-offered solution of having the touchscreen be horizontal (like Microsoft's "Surface") is also non-optimal (it would hurt your neck to sit like that all day).
Here's a potential solution I've been thinking about: a dual device where there is a conventional vertical monitor in front of you and a touchscreen horizontal on your desk. Both display the same image, and you can interact with the horizontal surface via multi-touch. But you can see the result on the more comfortable vertical display, too. Importantly, the system has some way (combining the multi-touch data with camera input, or somesuch) to determine the placement of your hands--so that the outline of your hands (semi-transparent) could be displayed on screen. (In this sense similar to this "see-through" demo, in the sense that it simulates the location of your finger, giving you the feedback you need to manipulate on-screen objects.)
I think this system could work, and might be useful for a variety of applications. The behavior of the touch-surface could of course vary by application (perhaps acting as an array of soft-buttons for some apps, but mirroring the main display for others). Having said all that, I still believe that the keyboard+mouse combo is here to stay--it is a very efficient way to interact with the computer. But I would love to have a multi-touch-screen-tablet thing off to the side that I could use for certain tasks (e.g. conventional tablet tasks like image editing, but also novel multi-touch tasks like sorting images or naturally re-arranging cells in a spreadsheet...).
(Okay, now go ahead and point out all the stupid problems with this idea that I haven't thought of yet...!)
This is an "intangible" that is too often forgotten. I have my computer set up the way I want it. Not just in terms of installed programs, but in terms of what applications are open and how they are arranged on screen (and how they are internally arranged: e.g. toolbars, options, documents). This arrangement conveys information to me in much the same way that a "disorganized" desk actually contains important information for the user (the spatial arrangement of papers and piles allows the user to access information rather efficiently).
This means that every shutdown or reboot forces me to reorganize (as if someone tosses all the papers off your desk). Yes, sleep/hibernate modes should retain this information, but a full reboot generally destroys it. KDE remembers a lot about what programs were open and where they were placed, but still some information about window sizes, options that were set, and documents that were open, gets lost.
What I would like (and now I'm dreaming off onto a tangent, I know) is to have an OS/GUI that was able to properly save the "state" of all open programs. Not just a memory dump, but an proper save of what applications were running, what options were set, how toolbars were aligned, etc. This would allow me to restore the proper state after reboots. It would also allow to close and open "task groups", where each "task group" would contain a variety of tools/apps/documents, all arranged on screen in a particular way. (E.g. I open the "website editing" task-group that I had open last week, and it opens all my tools and text editors, just the way I left them, launches a new Firefox window with tabs properly populated like they should be, etc...)
Bringing this back to the energy-savings issue, consider this user-complaint from TFA:
The Forrester report does acknowledge that end-users have very little patience for downtime.
Users shouldn't have to deal with downtime. However why can't our modern computers have some simple logic: like IT loads a policy onto all computers that if they are idle for >1 hr after 6pm, they automatically save their state and shutdown; and then automatically bootup/wake-up and restore state at 6:30 am (exact times would of course be tuned based on the particular business or even user). For 99% of users, they would never see their computer turned off, yet it wouldn't be running uselessly all night long. All we would need is a robust way to save the computer state. (For that matter, why don't companies currently do this using the sleep/hibernate modes?)
You are right that many successful ideas seem stupid at first. On the other hand, the vast majority of stupid-sounding ideas are, in fact, stupid and ultimately unsuccessful.
I think we all support the notion of researching new ideas, and trying out different approaches. If Microsoft wants to spend R&D dollars on these ideas, that's great. The annoying part is when companies issue hyperbole-filled press releases about how these ideas will "revolutionize the world" and so forth. In those cases, it is totally appropriate for others to point out how these ideas are not-novel/inefficient/stupid/etc.
If the new idea can't withstand some healthy skepticism and criticism, I doubt it's really of that much value.
I wouldn't say a black hole is "unobservable". It emits no light, but has a measurable gravitational field. Conversely consider something like light, which has no mass but can be measured by its electromagnetic interaction (e.g. using a camera).
Different subatomic particles interact in different ways. Four fundamental forces have been identified: electromagnetic, weak nuclear, strong nuclear, and gravitational. A particular particle may interact via 1 or more of these modes. Just because it is "invisible" with respect to a given force does not make it "unobservable": as long as it interacts via at least one force, it can be measured/observed using that force.
All the examples you've given are of things that are observable: black holes and dark matter and dark energy are all observable via the gravitational effects they produce. Just because they are not observable via light doesn't make them unobservable. (Strictly black holes do emit low-levels of measurable radiation (Hawking radiation), and could also be detected in this way.) The "strings" of string theory (if they exist) should in principle be measurable by studying the interactions of particles via the four forces (whether or not we will ever achieve the energy scales required to do so is a separate question). For that matter it is difficult to "see" air, but it is easy to observe/measure it in other ways.
You have falsely equated "interact strongly via the electromagnetic force" to "observable". It's a natural mistake for humans, since our visual sense is so well-developed. However just because it is invisible to our eyes does not make it an "unobservable object". A truly "unobservable object" would be one which doesn't interact via any force. Such an object isn't merely "unobservable", it is simply "nonexistent" by any physical definition (since it cannot interact with anything else in the universe).
The high price 2nd hand market helps people afford new games, by selling their old games.
Indeed. Moreover, if the games companies are convinced that the availability of 2nd-hand games reduces the purchasing of brand-new games, then they should offer to buy back games.
Seriously. If they just set a dollar value on the 2nd-hand game sales (in terms of how much it "takes away" from their other sales), and then offered to buy games at that price (well, a little lower, so they make a profit, of course), then they would make more money. In fact by giving gamers a "credit" on new titles (rather than straight cash), they would make even more money: by encouraging/forcing the buyers to spend their next gaming dollar with them. This keeps the 2nd-hand copies out of the 2nd-hand market, forcing other gamers to purchase brand new copies.
In other words: if there really is money to be made on buying/selling 2nd-hand games (money that would otherwise go to the game developers), then they can get that money by getting into that business. So why don't they? Perhaps it's not as big a deal as they pretend.
Anyone know good sources of legal free downloadable music?
The best "one stop shop" for legally downloadable music (everything Creative Commons licensed) is Jamendo. 14,000 albums covering just about every genre I can think of, all free to grab. The Jamendo site has a decent interface for tracking favorites, getting recommendations, and so forth. Overall the quality is quite high. Many bands accept donations, but they never nag you about it.
Another to look into is Magnatune. It's not free--but all the albums can be previewed on the site (flash player), and are licensed under Creative Commons. You can buy albums with a "pay what you want" system (there is a minimum amount, $5/album or something), and can thereafter share these downloads with friends. They recently introduced a "all you can eat" option, where with a monthly fee you can download as much music as you want (again, you set the price; I think the minimum is $10/month). All the music is DRM free, your choice of format, no hassles. Unlike Jamendo, Magnatune albums are selected and vetted; the overall quality of music is very high.
With these excellent sources of high-quality music (in terms of both musical aesthetics and encoding), in totally unencumbered formats, available free (or at very reasonable cost), I have a really hard time understanding why people still buy "big name" music. I guess marketing really works. In which case, we really need to spread the word about the availability of Creative Commons music.
Well, the IWF didn't really say "sorry, we were wrong to classify that image as indecent" or even "sorry, we were wrong to censor that website". What they actually said was "oops, we tried to censor but failed." As such, they are merely saying that they have reversed the censoring simply because of the publicity and the overall failure in suppressing the image (quite the opposite: more people were exposed to the image as a result of the ban).
This is hardly the admission of failure that many of us were looking for. They maintain that the image is indecent, and that they will continue to try to censor sites that host it (or images like it).
which raises the question as to whether procedures need to be put in place to prevent mistakes happening again.
They specifically state that they will continue to investigate sites (or at least UK sites) that host that image. Clearly they do not consider their previous actions to be "mistakes."
I hope for one thing- that the IWF now ensure they concentrate on what they're supposed to concentrate on, helping prevent child abuse
I won't hold my breath. Seems like they are backing away from a high-publicity situation. But there is nothing to suggest that they won't continue censoring to the limit that public outcry can support, including censoring numerous sites that do not contribute to child abuse. This is not a trend I like.
Well in this case the definition is a matter of university policy. In other words, the punitive actions are occurring because of a breach of the terms-of-use for the university network (which, presumably, every student and university employee agree to when they join). So, here is the university's policy. In particular, they state:
"Bulk e-mail" in this context means the transmission of an e-mail message within a short time frame to more than a small set of recipients who may not have elected voluntarily to receive the e-mail. "Short time frame" means an interval spanning as long as 2 days. "Small set of recipients" means the size of individual-recipient address lists (To, CC, BCC fields) typical of most e-mails in common use, ranging from 1, to a few, to as many as may be involved in a large committee or work group (~20-30). Use of mailing lists and listservs to which recipients may voluntarily opt in and opt out is encouraged, and this type of e-mail distribution is not included in the meaning of "bulk e-mail" in this document.
So it is considered unacceptable if the receipient did not voluntarily elect to receive it, and if the distribution size is excessively large (e.g. more than 30 people). The present case involves emailing 391 professors directly. The "proper" method is no doubt to send the message through a mailing list, in which case various people are voluntarily able to join/leave the list.
It seems pretty clear that she has broken the rules as set out by the university's policy. Now, whether an exception should be made because of the importance of her message is then up to the administration. But at least technically she has been warned (through the policy) that her actions are not acceptable.
A legal definition is not meaningful here, since the university is not taking her to court. They are suspending her (or just threatening to?) because she broke the network terms of use.
I've been using a simpler solution for a few years
I love Slashdot.
Only on Slashdot would it be "simpler" to code a custom script that automatically runs when starting a particular application, generates a new temporary profile, sets an environment variable to use that profile, and deletes the profile on exit; rather than sometimes click a menu item marked "private".
I'm not disagreeing that your solution is simpler, by the way. It is actually a great way to force a particular behavior in a robust way, and is simple to use once implemented. But it's only "simpler" for Slashdotters!
1. What do you mean about thumbnailing the pictures? Isn't there only one picture?
The image "080229lefthandup.jpg" is displayed on the page using scaling (your "img" tag has a "width=225 pixels" to make the picture smaller). This is considered unprofessional because you are forcing the user's browser to load a big picture when a small image would do (and would load faster). Some browsers also won't scale it nicely. So the "professional" solution is to resize the image (in Photoshop or GIMP or whatever) to the correct size and use that image, unscaled. You can always link to the full-size version if you want people to have access to it. That having been said, I don't think it's a big deal on your site: as you said it's only a single image, and you're only slightly rescaling it. (This becomes a much bigger issue when people use gigantic pictures scaled down to thumbnails, because the page load takes forever.)
2. What should I do instead of using font tags?
Modern practice is to define styles in a "cascading style sheet". You either redefine how the standard tags will display, or you create new styles with custom names. Then you always consistently use the appropriate style tags. Changes in font (and font size, etc.) can then be made just in the style file. That having been said, you run your blog off of blogspot; I'm not sure if it allows you to define custom CSS or not.
3. What is semantic markup?
This is the same complaint as #2. Basically if you're defining a site "properly" you define classes of text, like "emphasis" or "quoting" or "blogentry". Then you flag your text (using markup like "div class=blogentry" or whatever) and it gets formatted properly. This is better because you avoid hard-coding the changes for each and every time you want to change a look. For example, once all the quotes have been labeled with "quoting" you can easily change the look/feel of the entire site just by changing the definition of the style for "quoting". This also means that the various text regions have been flagged semantically (all the quotes are labeled as such). As much as possible you should use the established html tags (p for paragraph, ul for a list, etc.), since their meaning is, in principle, already defined.
4. I thought the old fashioned fonts were more reflective of my 19th century personality, but maybe I'll experiment with something else.
Your font is fine... but sans-serif fonts are almost always easier to read on computer screens.
Again, I think the complaints are somewhat unfair, because you are using the blogspot engine, which is restrictive. Many of the ugly markup in the page source are the fault of blogspot.
I do confess that I find some parts of the page inelegant (the large number of links at the bottom of the page seem disorganized and not properly aligned; I would prefer clearer demarcation between comments on the post pages (bounding boxes, or horizontal lines, or color contrast or something)...). But overall your blog is such as great source of insightful commentary and information that any such complaint seems trite by comparison.
Remember that if you ever want suggestions or help the Slashdot community is here for you. (And plenty of us are willing to help without even the nominal fees that Hurricane78 mentioned.)
something that Apple politically couldn't say: Mac antivirus software primarily protects against Windows viruses
Considering that Apple runs ads that directly state that "PCs" get viruses whereas "Macs" do not, I don't see why they would mind saying roughly the same thing in a tech note.* Seems to me that they have already taken a pretty visible stance on that political issue.
That having been said, I suspect you are right: once this whole issue blew up, it was safer to completely distance themselves from the original tech note, rather than try and explain why they had originally issued it.
[*] Conceivably the tech note was written by some lower-level employee who didn't want to say something controversial. So instead he/she left it vague and just suggested that "antivirus is a good idea" and so on.
I really don't understand #3 at all. If it's free, and in the public domain as other commenters have pointed out, what's the problem?
The issue with #3 is fair competition. For example, they might release a video press release to the major network, and then a day later post a lower-quality version on their website for download. In this case, the major networks were receiving preferential treatment (both because they were given earlier access and access to a higher-quality version).
The point is that in addition to openness, there should be fairness: the content should be made available to everyone (citizens, news outlets, corporations, etc.) at the same time, with the same quality, with the same "ease", and under the same terms. This means that companies are able to compete fairly in their dissemination/commentary/reporting/whatever (instead of one company having an advantage of some sort). This also means that citizens have the same opportunities for analysis/commentary/mashup that corporations do.
You're correct that security doesn't need to be perfect to be useful--as long as it helps, that's something.
However, a major problem occurs when people over-estimate the quality of a security measure. Two immediate consequences are (1) security overall may decrease as people mistakenly rely on an ineffective tool; and (2) people are falsely accused. In the extreme case of over-confidence in technology, a person can not only be falsely accused but also falsely detained, charged or even convicted. It is dangerous for security personnel or law enforcement to be over-estimating the tools at their disposal.
In this case I can easily imagine people being detained because they failed the "terrorist test"--without any other evidence existing to incriminate them. I would argue that using bad security tools, or over-estimating the effectiveness of tools, is worse than useless: it is actively dangerous for the well-being of innocents.
don't polygraphs also rely (in part) on body temperature, heart rate and respiration?
Polygraphs measure those things, but don't do much with the data. The main purpose of a polygraph is not to detect lies, but to intimidate the person being questioned. The idea is to trick the person into thinking that the polygraph is infallible and can determine when they are lying. This gives the interrogator another way to pressure the person into talking. (The person may incorrectly believe that the interrogator "already knows" or may reveal secrets because they feel that they no longer have any control--they don't feel culpable since they can't hide secrets from the machine.) Of course admitting that this is the purpose of a polygraph would undermine the tactic.
I'm guessing this new technology will be much the same: it won't actually work by measuring anything useful; but it may have a psychological effect that makes people easier to interrogate. This might be (marginally) useful for uncovering the occasional teenager smuggling pot, but I doubt it will do anything useful when it comes to terrorism. This quote is hilarious:
Developers say the combination of these technologies can detect a person's reaction to certain stimuli by reading body temperature, heart rate and respiration -- signals a terrorist unwittingly emits before he plans to commit an attack
For this to be true--for them to actually have calibrated their machine in a rigorous way, so that it can detect "terrorist intentions" with any kind of certainty--they would need to have tested it with a statistically-significant number of terrorists. Somehow I doubt their R&D facility has a few hundred terrorists in lockup (willing to lie and not lie on demand). I'm guessing their actual sample size was closer to zero. In other words they are just guessing that someone with "terrorist intentions" will exhibit similar physiological responses to someone who is nervous for other reasons.
Yet another worthless security measure being sold to worthless security organizations.
I think the take-home message from most of those discussions is that you need to make preparations. Just like with everything else in your life, you should ideally keep things organized enough so that your survivors can deal with it, both in terms of wrapping up your estate, and keeping the things that matter. So this means keeping a list of passwords and encryption keys somewhere (e.g. in a safe deposit box), and even instructions about what to do with various accounts. Your data should also be organized so that your family can make copies of things like photos and find nostalgic things that you've written.
Another point to consider is the things that you don't want your survivors seeing. If you have any secrets you want to take to your grave, be sure to encrypt them. And for the sake of your children's sanity, hide your porn! (Or label it so they can avoid it!)
He's never had to do one day of REAL work. He's only ever worked at a university and he now works for the FSF.
Umm.... so working at a university is not "REAL work"? That's rather offensive to those of us who work in academia and/or science, and totally unfair.
Working at a university is work. Contrary to the "ivory tower" insults, the work done is relevant and helps drive society forward. Think of how many people are trained through universities. Think of how many technological advances resulted from university research. Think of how many fundamental scientific discoveries were made at universities. Then think about how much of the technology, progress, and economic activity around you is based on those discoveries.
Universities and the academics they employ play a vital role in society, and in the economy. The people working at these institutes work very hard (well, most of them anyways). So, in what way is the work they do not "REAL work"?
Sorry. I wasn't especially specific in my previous post because that wasn't my point. My point was that alternatives exist.
To get more specific, I wasn't suggesting the scenario you are attacking. I was thinking more like this: A programmer has an idea. He creates a demo/beta/first-version, and releases it for free (with source). This generates enthusiasm. People say they would like to see a full version (or second version, etc.). The programmer starts work on said version, and says that he thinks X would be a fair price (where X will fully cover his development time; obviously X will be a large number like $100,000 or a million, or whatever). He then solicits payments. Depending on the kind of software, this may mean a few big companies paying big sums of money (because they want the software). If it's a more consumer-focused software, then it may be more like thousands of individuals all putting in $20 or so. When the target sum is reached, the programmer releases the software (with source) to everyone who participated. Obviously everyone can give copies to friends; but the programmer doesn't mind, since his target sum was already reached.
The above is basically patronage-supported software. People who care would pay an amount they think is fair. The whole thing could no doubt be streamlined, with convenient intermediaries helping with "getting the word out" and accumulating the money (and refunding if the product isn't delivered) and all that. If the programmer's request is too large, he doesn't get his money and no one gets the software (so a free-market back-and-forth will set suitable prices).
Would this proposal be better than what we have now? I don't know. In some ways it's better (everything can be open-source; piracy becomes irrelevant; etc.), in some ways it's worse (change is hard; original developers can't maintain singular control over the future of a project; people don't like to pay ahead of time...). My point isn't really to invent a perfect system; merely to point out that alternatives exist.
You seem to be pretty knowledgeable about free and open-source software... so I'm a little surprised by some of the things in your post.
Specifically, you say:
Stallman [should] stop begrudging others the right to make their own products and sell them
Stallman has been very clear over the years that he has no issue with people monetizing software, making money off of programming, or even selling software. He merely emphasizes that anyone who obtains software must have access to code.
You seem to think that consulting is the only way to make money in an all-OSS software ecology. I don't think that's the case. In addition to programmers being paid by the hour to code, it's not hard to imagine situations where well-organized "payment requests" are created. Someone codes v1 of a product (or releases a beta), and then requests funds to deliver the completed version. Once the requested money has been sent in (by interested buyers), the full version (with source code) is delivered. (The buyer could be other companies or many individual consumers.)
Would that be different from current software business methods? Yes. But I don't think it's impossible (the main reason it doesn't exist more routinely today is because everyone finds it simpler to just do the same thing as everyone else), and companies could continue to make profits from selling innovating new software. I'm not trying to specifically advocate that this would be better; merely pointing out that Stallman's "software should be free" is not in conflict with people making money. (You may not like the details of alternate money-making models, but that doesn't mean they are not viable.)
I just don't think it's fair to say that Stallman is against selling software, or that consulting is the only way to make money off OSS.
The real problem is psychology is not very scientific.... The whole science is wish washy and based on subjective judgment as opposed to a first order science that basis it's classification scheme on measurable objective facts.
Well it's worth differentiating between "clinical psychology" which seeks to treat mental illness, and "scientific psychology" (for lack of a better term) which merely seeks to understand the brain/mind/etc.
Clinical psychology necessarily makes judgments about what is normal and abnormal, what is detrimental and not detrimental. There is an implied value assessment because the presumption is that people want to be stable and happy (just as in medicine there is a presumption that people want to be healthy and stay alive). We can all agree that psychology in this mode can often be subjective, is influenced by our cultural values, and indeed can be imprecise and "wishy washy" at times.
The scientific pursuit of understanding thoughts and emotions ("scientific psychology" or whatever) has bigger error bars than, say, physics, since the subject matter is inherently varied and complex. However there are many very rigorous branches of psychology (neuroscience, evolutionary psychology) which are doing science "the right way", with testable hypotheses, real experiments, mathematical models, and non-arbitrary classification (well, as much as any science).
It's unfortunate that there is no commonly-accepted terminology (at least none that I'm aware of) to differentiate between the two... because there are stark differences. On the one hand we have researchers who are abstractly studying how the brain and mind work; on the other we have people trying to use that research to heal mental illness. Obviously the two groups can help each other (and overlap frequently), but I would be careful about saying that psychology isn't scientific, merely because the application of that knowledge-base to fixing real-world problems is murky and subjective.
Physics is quite objective and amoral; but the application of physics to the real-world (e.g. mechanical engineering to build a bridge) will of course include some moral component (e.g. in terms of deciding whether money or safety is more important).
What does the halting problem have to do with this?
The halting problem might be relevant if you absolutely needed a provably (in the mathematical sense) perfect analysis of the algorithm that a binary implements. But any practical implementation of mawlare detection wouldn't need to be perfect: it would just have to disassemble the binary and pull out algorithmic motifs, and check these against a database (where the motif "loop over values and store to single file" is okay but "loop over all files on disk and insert code into each one" is not okay).
Bringing up the halting problem is basically saying "there are theoretical reasons why you cannot build a perfect algorithm classifier... so why bother trying?" We don't need perfect. We just need useful.
This is rather offtopic, but I'd like to say it anyways...
It used to be that the Slashdot frontpage was riddled with dupes. We complained every time that the editors were idiots, that they should have a system in place to recognize dupes. Many people even said that they emailed regarding the dupes while they were in the mysterious future, and yet the still hit the front page. Thus was born the Slashdot meme about every story being eventually duped.
But, I have to say that over the last while, the number of dupes is way, way down. The firehose and tagging seem to have alot to do with it--dupes are flagged earlier in the process, giving the editors the feedback they need.
Dupes are not entirely eliminated, but the frequency is down. So I'd like to say: thanks to the Slashdot staff for fixing the issue that we complained about. We are a whiny bunch, and it's too easy for us to complain but then forget to appreciate the things that are fixed (or have always been good). So, again, good work on the dupe reduction.
I'm not so sure about that. I mean, the military currently uses a whole bunch of stealth technology against their enemies: everything from simple paint color and camouflage, to radar-reflective stealth paint or ultra-quiet engines for submarines. None of these are perfect, but all are useful.
You may not be able to make yourself 100% invisible to an enemy that has good tech, but as long as you can give yourself an advantage in hiding, it's worth using. The "advantage" could be increased survival (enemy hit accuracy is reduced), better range (you can get closer before being detected), or maybe just the cost to the enemy for them to launch all the overhead imaging and use all magnetic field sensing equipment you just mentioned.
If cloaking became viable, it would definitely be used by the military against other high-tech enemies. In battle, every advantage counts.
Slashdot Mirror
It's possible the gov't patented this so they could share the information with other people and not worry about some private company patenting the idea and then sueing everyone else for us it. Basically - patent to allow people to use it.
If that's the intent, and the patent system is working as intended, then the patenting is superfluous. Publication of all the details (without restriction) is sufficient to prevent anyone else from patenting the idea, because the publication acts as demonstrable prior art with which to challenge any subsequent patent application. (This is also why anyone who wants to patent something usually has to hold off on publication until the patent process is already underway--a publication can be used to show that the idea had already become "common knowledge" before the patent.)
Of course, in reality publishing details is usually not enough to prevent someone else from patenting, because it seems that no one (least of all patent examiners) does a thorough job of uncovering prior art. So I guess you could argue that they were being extra-careful and strategic, making sure to patent it to so that they could liberate the idea without any chance of someone else claiming it.
But if it really were their intention to liberate the idea for all to use, one would think that any press release mentioning the patent would also mention the associated "perpetual, worldwide, royalty-free license to use these patented methods" that the NSA is offering. I see no such offer.
So really what you think your doing in boycotting them is just being stupid and showing your lack of knowledge of the situation. Because Apple really isnt in the business to DRM crap...
It may not be Apple's "fault" that there is DRM. Maybe they don't want it to be there--maybe it's entirely the labels making that call, and Apple has no choice, if they want to sell those tracks.
But that doesn't mean that boycotting Apple is a mistake. The point of a boycott is to (1) draw attention to a cause; and (2) make it financially difficult for the corporation to continue their practice. If a boycott of Apple products were sufficiently large and successful, then the labels would take notice (since they would be selling fewer tracks, as compared to non-DRM music), and Apple would take notice. Apple may not have the final say about DRM, but they are in a bargaining position. If iTunes DRM was causing them to lose sales in other divisions, they might very well negotiate aggressively with the labels to offer more DRM-free tracks. The only way to make a corporation "care" is to make it about money. That's the point of a boycott.
Having said all that, I doubt the presently proposed boycott will have a large effect. Nor am I convinced that this is the best way to eliminate DRM. But I don't like arguments against the boycott that amount to "It's not Apple's fault--they don't want DRM either!" (which is irrelevant--they offer DRM'ed products, so they fall under the boycott); or "Why bother?" (which is bothersome--should people who care about an issue do nothing?); or even "Every company does it!" (fine--boycott them also).
lawsuits seem to be their main source of income these days
Only they know for sure, of course... but Ray's best estimate is that they are losing money overall: they may make money off settlements (or maybe just break even?), but the legal and investigative costs that go into all those dropped or default judgment cases destroys all that profit, and then some.
I don't think it's actually profitable for them. I think it was always intended as some kind of "campaign of fear" to make people too afraid to file-share... which seems not to have worked.
... the whole concept of touch screens for office use is fundementally flawed. It requires you to have your arms raised to operate the screen - which is an unnatural position and very tiring to do for long periods of time.
Agreed. But this "see-through" demo shows that an innovation can potentially fix what at first seems to be a major problem. (Of course I won't know if this "see-through" idea is really a valid solution until I have a chance to play with it myself...).
In the case of touchscreen technology, I agree that pressing on a vertical screen would get tiring real quick. And the sometimes-offered solution of having the touchscreen be horizontal (like Microsoft's "Surface") is also non-optimal (it would hurt your neck to sit like that all day).
Here's a potential solution I've been thinking about: a dual device where there is a conventional vertical monitor in front of you and a touchscreen horizontal on your desk. Both display the same image, and you can interact with the horizontal surface via multi-touch. But you can see the result on the more comfortable vertical display, too. Importantly, the system has some way (combining the multi-touch data with camera input, or somesuch) to determine the placement of your hands--so that the outline of your hands (semi-transparent) could be displayed on screen. (In this sense similar to this "see-through" demo, in the sense that it simulates the location of your finger, giving you the feedback you need to manipulate on-screen objects.)
I think this system could work, and might be useful for a variety of applications. The behavior of the touch-surface could of course vary by application (perhaps acting as an array of soft-buttons for some apps, but mirroring the main display for others). Having said all that, I still believe that the keyboard+mouse combo is here to stay--it is a very efficient way to interact with the computer. But I would love to have a multi-touch-screen-tablet thing off to the side that I could use for certain tasks (e.g. conventional tablet tasks like image editing, but also novel multi-touch tasks like sorting images or naturally re-arranging cells in a spreadsheet...).
(Okay, now go ahead and point out all the stupid problems with this idea that I haven't thought of yet...!)
This is an "intangible" that is too often forgotten. I have my computer set up the way I want it. Not just in terms of installed programs, but in terms of what applications are open and how they are arranged on screen (and how they are internally arranged: e.g. toolbars, options, documents). This arrangement conveys information to me in much the same way that a "disorganized" desk actually contains important information for the user (the spatial arrangement of papers and piles allows the user to access information rather efficiently).
This means that every shutdown or reboot forces me to reorganize (as if someone tosses all the papers off your desk). Yes, sleep/hibernate modes should retain this information, but a full reboot generally destroys it. KDE remembers a lot about what programs were open and where they were placed, but still some information about window sizes, options that were set, and documents that were open, gets lost.
What I would like (and now I'm dreaming off onto a tangent, I know) is to have an OS/GUI that was able to properly save the "state" of all open programs. Not just a memory dump, but an proper save of what applications were running, what options were set, how toolbars were aligned, etc. This would allow me to restore the proper state after reboots. It would also allow to close and open "task groups", where each "task group" would contain a variety of tools/apps/documents, all arranged on screen in a particular way. (E.g. I open the "website editing" task-group that I had open last week, and it opens all my tools and text editors, just the way I left them, launches a new Firefox window with tabs properly populated like they should be, etc...)
Bringing this back to the energy-savings issue, consider this user-complaint from TFA:
Users shouldn't have to deal with downtime. However why can't our modern computers have some simple logic: like IT loads a policy onto all computers that if they are idle for >1 hr after 6pm, they automatically save their state and shutdown; and then automatically bootup/wake-up and restore state at 6:30 am (exact times would of course be tuned based on the particular business or even user). For 99% of users, they would never see their computer turned off, yet it wouldn't be running uselessly all night long. All we would need is a robust way to save the computer state. (For that matter, why don't companies currently do this using the sleep/hibernate modes?)
You are right that many successful ideas seem stupid at first. On the other hand, the vast majority of stupid-sounding ideas are, in fact, stupid and ultimately unsuccessful.
I think we all support the notion of researching new ideas, and trying out different approaches. If Microsoft wants to spend R&D dollars on these ideas, that's great. The annoying part is when companies issue hyperbole-filled press releases about how these ideas will "revolutionize the world" and so forth. In those cases, it is totally appropriate for others to point out how these ideas are not-novel/inefficient/stupid/etc.
If the new idea can't withstand some healthy skepticism and criticism, I doubt it's really of that much value.
That's the nature of an unobservable object.
I wouldn't say a black hole is "unobservable". It emits no light, but has a measurable gravitational field. Conversely consider something like light, which has no mass but can be measured by its electromagnetic interaction (e.g. using a camera).
Different subatomic particles interact in different ways. Four fundamental forces have been identified: electromagnetic, weak nuclear, strong nuclear, and gravitational. A particular particle may interact via 1 or more of these modes. Just because it is "invisible" with respect to a given force does not make it "unobservable": as long as it interacts via at least one force, it can be measured/observed using that force.
All the examples you've given are of things that are observable: black holes and dark matter and dark energy are all observable via the gravitational effects they produce. Just because they are not observable via light doesn't make them unobservable. (Strictly black holes do emit low-levels of measurable radiation (Hawking radiation), and could also be detected in this way.) The "strings" of string theory (if they exist) should in principle be measurable by studying the interactions of particles via the four forces (whether or not we will ever achieve the energy scales required to do so is a separate question). For that matter it is difficult to "see" air, but it is easy to observe/measure it in other ways.
You have falsely equated "interact strongly via the electromagnetic force" to "observable". It's a natural mistake for humans, since our visual sense is so well-developed. However just because it is invisible to our eyes does not make it an "unobservable object". A truly "unobservable object" would be one which doesn't interact via any force. Such an object isn't merely "unobservable", it is simply "nonexistent" by any physical definition (since it cannot interact with anything else in the universe).
The high price 2nd hand market helps people afford new games, by selling their old games.
Indeed. Moreover, if the games companies are convinced that the availability of 2nd-hand games reduces the purchasing of brand-new games, then they should offer to buy back games.
Seriously. If they just set a dollar value on the 2nd-hand game sales (in terms of how much it "takes away" from their other sales), and then offered to buy games at that price (well, a little lower, so they make a profit, of course), then they would make more money. In fact by giving gamers a "credit" on new titles (rather than straight cash), they would make even more money: by encouraging/forcing the buyers to spend their next gaming dollar with them. This keeps the 2nd-hand copies out of the 2nd-hand market, forcing other gamers to purchase brand new copies.
In other words: if there really is money to be made on buying/selling 2nd-hand games (money that would otherwise go to the game developers), then they can get that money by getting into that business. So why don't they? Perhaps it's not as big a deal as they pretend.
Anyone know good sources of legal free downloadable music?
The best "one stop shop" for legally downloadable music (everything Creative Commons licensed) is Jamendo. 14,000 albums covering just about every genre I can think of, all free to grab. The Jamendo site has a decent interface for tracking favorites, getting recommendations, and so forth. Overall the quality is quite high. Many bands accept donations, but they never nag you about it.
Another to look into is Magnatune. It's not free--but all the albums can be previewed on the site (flash player), and are licensed under Creative Commons. You can buy albums with a "pay what you want" system (there is a minimum amount, $5/album or something), and can thereafter share these downloads with friends. They recently introduced a "all you can eat" option, where with a monthly fee you can download as much music as you want (again, you set the price; I think the minimum is $10/month). All the music is DRM free, your choice of format, no hassles. Unlike Jamendo, Magnatune albums are selected and vetted; the overall quality of music is very high.
With these excellent sources of high-quality music (in terms of both musical aesthetics and encoding), in totally unencumbered formats, available free (or at very reasonable cost), I have a really hard time understanding why people still buy "big name" music. I guess marketing really works. In which case, we really need to spread the word about the availability of Creative Commons music.
the IWF has accepted they're wrong
Well, the IWF didn't really say "sorry, we were wrong to classify that image as indecent" or even "sorry, we were wrong to censor that website". What they actually said was "oops, we tried to censor but failed." As such, they are merely saying that they have reversed the censoring simply because of the publicity and the overall failure in suppressing the image (quite the opposite: more people were exposed to the image as a result of the ban).
This is hardly the admission of failure that many of us were looking for. They maintain that the image is indecent, and that they will continue to try to censor sites that host it (or images like it).
which raises the question as to whether procedures need to be put in place to prevent mistakes happening again.
They specifically state that they will continue to investigate sites (or at least UK sites) that host that image. Clearly they do not consider their previous actions to be "mistakes."
I hope for one thing- that the IWF now ensure they concentrate on what they're supposed to concentrate on, helping prevent child abuse
I won't hold my breath. Seems like they are backing away from a high-publicity situation. But there is nothing to suggest that they won't continue censoring to the limit that public outcry can support, including censoring numerous sites that do not contribute to child abuse. This is not a trend I like.
So it is considered unacceptable if the receipient did not voluntarily elect to receive it, and if the distribution size is excessively large (e.g. more than 30 people). The present case involves emailing 391 professors directly. The "proper" method is no doubt to send the message through a mailing list, in which case various people are voluntarily able to join/leave the list.
It seems pretty clear that she has broken the rules as set out by the university's policy. Now, whether an exception should be made because of the importance of her message is then up to the administration. But at least technically she has been warned (through the policy) that her actions are not acceptable.
A legal definition is not meaningful here, since the university is not taking her to court. They are suspending her (or just threatening to?) because she broke the network terms of use.
I've been using a simpler solution for a few years
I love Slashdot.
Only on Slashdot would it be "simpler" to code a custom script that automatically runs when starting a particular application, generates a new temporary profile, sets an environment variable to use that profile, and deletes the profile on exit; rather than sometimes click a menu item marked "private".
I'm not disagreeing that your solution is simpler, by the way. It is actually a great way to force a particular behavior in a robust way, and is simple to use once implemented. But it's only "simpler" for Slashdotters!
1. What do you mean about thumbnailing the pictures? Isn't there only one picture?
The image "080229lefthandup.jpg" is displayed on the page using scaling (your "img" tag has a "width=225 pixels" to make the picture smaller). This is considered unprofessional because you are forcing the user's browser to load a big picture when a small image would do (and would load faster). Some browsers also won't scale it nicely. So the "professional" solution is to resize the image (in Photoshop or GIMP or whatever) to the correct size and use that image, unscaled. You can always link to the full-size version if you want people to have access to it. That having been said, I don't think it's a big deal on your site: as you said it's only a single image, and you're only slightly rescaling it. (This becomes a much bigger issue when people use gigantic pictures scaled down to thumbnails, because the page load takes forever.)
2. What should I do instead of using font tags?
Modern practice is to define styles in a "cascading style sheet". You either redefine how the standard tags will display, or you create new styles with custom names. Then you always consistently use the appropriate style tags. Changes in font (and font size, etc.) can then be made just in the style file. That having been said, you run your blog off of blogspot; I'm not sure if it allows you to define custom CSS or not.
3. What is semantic markup?
This is the same complaint as #2. Basically if you're defining a site "properly" you define classes of text, like "emphasis" or "quoting" or "blogentry". Then you flag your text (using markup like "div class=blogentry" or whatever) and it gets formatted properly. This is better because you avoid hard-coding the changes for each and every time you want to change a look. For example, once all the quotes have been labeled with "quoting" you can easily change the look/feel of the entire site just by changing the definition of the style for "quoting". This also means that the various text regions have been flagged semantically (all the quotes are labeled as such). As much as possible you should use the established html tags (p for paragraph, ul for a list, etc.), since their meaning is, in principle, already defined.
4. I thought the old fashioned fonts were more reflective of my 19th century personality, but maybe I'll experiment with something else.
Your font is fine... but sans-serif fonts are almost always easier to read on computer screens.
Again, I think the complaints are somewhat unfair, because you are using the blogspot engine, which is restrictive. Many of the ugly markup in the page source are the fault of blogspot.
I do confess that I find some parts of the page inelegant (the large number of links at the bottom of the page seem disorganized and not properly aligned; I would prefer clearer demarcation between comments on the post pages (bounding boxes, or horizontal lines, or color contrast or something)...). But overall your blog is such as great source of insightful commentary and information that any such complaint seems trite by comparison.
Remember that if you ever want suggestions or help the Slashdot community is here for you. (And plenty of us are willing to help without even the nominal fees that Hurricane78 mentioned.)
something that Apple politically couldn't say: Mac antivirus software primarily protects against Windows viruses
Considering that Apple runs ads that directly state that "PCs" get viruses whereas "Macs" do not, I don't see why they would mind saying roughly the same thing in a tech note.* Seems to me that they have already taken a pretty visible stance on that political issue.
That having been said, I suspect you are right: once this whole issue blew up, it was safer to completely distance themselves from the original tech note, rather than try and explain why they had originally issued it.
[*] Conceivably the tech note was written by some lower-level employee who didn't want to say something controversial. So instead he/she left it vague and just suggested that "antivirus is a good idea" and so on.
I really don't understand #3 at all. If it's free, and in the public domain as other commenters have pointed out, what's the problem?
The issue with #3 is fair competition. For example, they might release a video press release to the major network, and then a day later post a lower-quality version on their website for download. In this case, the major networks were receiving preferential treatment (both because they were given earlier access and access to a higher-quality version).
The point is that in addition to openness, there should be fairness: the content should be made available to everyone (citizens, news outlets, corporations, etc.) at the same time, with the same quality, with the same "ease", and under the same terms. This means that companies are able to compete fairly in their dissemination/commentary/reporting/whatever (instead of one company having an advantage of some sort). This also means that citizens have the same opportunities for analysis/commentary/mashup that corporations do.
You're correct that security doesn't need to be perfect to be useful--as long as it helps, that's something.
However, a major problem occurs when people over-estimate the quality of a security measure. Two immediate consequences are (1) security overall may decrease as people mistakenly rely on an ineffective tool; and (2) people are falsely accused. In the extreme case of over-confidence in technology, a person can not only be falsely accused but also falsely detained, charged or even convicted. It is dangerous for security personnel or law enforcement to be over-estimating the tools at their disposal.
In this case I can easily imagine people being detained because they failed the "terrorist test"--without any other evidence existing to incriminate them. I would argue that using bad security tools, or over-estimating the effectiveness of tools, is worse than useless: it is actively dangerous for the well-being of innocents.
don't polygraphs also rely (in part) on body temperature, heart rate and respiration?
Polygraphs measure those things, but don't do much with the data. The main purpose of a polygraph is not to detect lies, but to intimidate the person being questioned. The idea is to trick the person into thinking that the polygraph is infallible and can determine when they are lying. This gives the interrogator another way to pressure the person into talking. (The person may incorrectly believe that the interrogator "already knows" or may reveal secrets because they feel that they no longer have any control--they don't feel culpable since they can't hide secrets from the machine.) Of course admitting that this is the purpose of a polygraph would undermine the tactic.
I'm guessing this new technology will be much the same: it won't actually work by measuring anything useful; but it may have a psychological effect that makes people easier to interrogate. This might be (marginally) useful for uncovering the occasional teenager smuggling pot, but I doubt it will do anything useful when it comes to terrorism. This quote is hilarious:
Developers say the combination of these technologies can detect a person's reaction to certain stimuli by reading body temperature, heart rate and respiration -- signals a terrorist unwittingly emits before he plans to commit an attack
For this to be true--for them to actually have calibrated their machine in a rigorous way, so that it can detect "terrorist intentions" with any kind of certainty--they would need to have tested it with a statistically-significant number of terrorists. Somehow I doubt their R&D facility has a few hundred terrorists in lockup (willing to lie and not lie on demand). I'm guessing their actual sample size was closer to zero. In other words they are just guessing that someone with "terrorist intentions" will exhibit similar physiological responses to someone who is nervous for other reasons.
Yet another worthless security measure being sold to worthless security organizations.
Incidentally, related questions have been the topic of previous "Ask Slashdot" stories:
What Does Your Dead Man's Switch Do
Your Digital Inheritance
What Happens To Your Data When You Die
I think the take-home message from most of those discussions is that you need to make preparations. Just like with everything else in your life, you should ideally keep things organized enough so that your survivors can deal with it, both in terms of wrapping up your estate, and keeping the things that matter. So this means keeping a list of passwords and encryption keys somewhere (e.g. in a safe deposit box), and even instructions about what to do with various accounts. Your data should also be organized so that your family can make copies of things like photos and find nostalgic things that you've written.
Another point to consider is the things that you don't want your survivors seeing. If you have any secrets you want to take to your grave, be sure to encrypt them. And for the sake of your children's sanity, hide your porn! (Or label it so they can avoid it!)
He's never had to do one day of REAL work. He's only ever worked at a university and he now works for the FSF.
Umm.... so working at a university is not "REAL work"? That's rather offensive to those of us who work in academia and/or science, and totally unfair.
Working at a university is work. Contrary to the "ivory tower" insults, the work done is relevant and helps drive society forward. Think of how many people are trained through universities. Think of how many technological advances resulted from university research. Think of how many fundamental scientific discoveries were made at universities. Then think about how much of the technology, progress, and economic activity around you is based on those discoveries.
Universities and the academics they employ play a vital role in society, and in the economy. The people working at these institutes work very hard (well, most of them anyways). So, in what way is the work they do not "REAL work"?
Sorry. I wasn't especially specific in my previous post because that wasn't my point. My point was that alternatives exist.
To get more specific, I wasn't suggesting the scenario you are attacking. I was thinking more like this: A programmer has an idea. He creates a demo/beta/first-version, and releases it for free (with source). This generates enthusiasm. People say they would like to see a full version (or second version, etc.). The programmer starts work on said version, and says that he thinks X would be a fair price (where X will fully cover his development time; obviously X will be a large number like $100,000 or a million, or whatever). He then solicits payments. Depending on the kind of software, this may mean a few big companies paying big sums of money (because they want the software). If it's a more consumer-focused software, then it may be more like thousands of individuals all putting in $20 or so. When the target sum is reached, the programmer releases the software (with source) to everyone who participated. Obviously everyone can give copies to friends; but the programmer doesn't mind, since his target sum was already reached.
The above is basically patronage-supported software. People who care would pay an amount they think is fair. The whole thing could no doubt be streamlined, with convenient intermediaries helping with "getting the word out" and accumulating the money (and refunding if the product isn't delivered) and all that. If the programmer's request is too large, he doesn't get his money and no one gets the software (so a free-market back-and-forth will set suitable prices).
Would this proposal be better than what we have now? I don't know. In some ways it's better (everything can be open-source; piracy becomes irrelevant; etc.), in some ways it's worse (change is hard; original developers can't maintain singular control over the future of a project; people don't like to pay ahead of time...). My point isn't really to invent a perfect system; merely to point out that alternatives exist.
Specifically, you say:
Stallman [should] stop begrudging others the right to make their own products and sell them
Stallman has been very clear over the years that he has no issue with people monetizing software, making money off of programming, or even selling software. He merely emphasizes that anyone who obtains software must have access to code.
You seem to think that consulting is the only way to make money in an all-OSS software ecology. I don't think that's the case. In addition to programmers being paid by the hour to code, it's not hard to imagine situations where well-organized "payment requests" are created. Someone codes v1 of a product (or releases a beta), and then requests funds to deliver the completed version. Once the requested money has been sent in (by interested buyers), the full version (with source code) is delivered. (The buyer could be other companies or many individual consumers.)
Would that be different from current software business methods? Yes. But I don't think it's impossible (the main reason it doesn't exist more routinely today is because everyone finds it simpler to just do the same thing as everyone else), and companies could continue to make profits from selling innovating new software. I'm not trying to specifically advocate that this would be better; merely pointing out that Stallman's "software should be free" is not in conflict with people making money. (You may not like the details of alternate money-making models, but that doesn't mean they are not viable.)
I just don't think it's fair to say that Stallman is against selling software, or that consulting is the only way to make money off OSS.
The real problem is psychology is not very scientific. ... The whole science is wish washy and based on subjective judgment as opposed to a first order science that basis it's classification scheme on measurable objective facts.
Well it's worth differentiating between "clinical psychology" which seeks to treat mental illness, and "scientific psychology" (for lack of a better term) which merely seeks to understand the brain/mind/etc.
Clinical psychology necessarily makes judgments about what is normal and abnormal, what is detrimental and not detrimental. There is an implied value assessment because the presumption is that people want to be stable and happy (just as in medicine there is a presumption that people want to be healthy and stay alive). We can all agree that psychology in this mode can often be subjective, is influenced by our cultural values, and indeed can be imprecise and "wishy washy" at times.
The scientific pursuit of understanding thoughts and emotions ("scientific psychology" or whatever) has bigger error bars than, say, physics, since the subject matter is inherently varied and complex. However there are many very rigorous branches of psychology (neuroscience, evolutionary psychology) which are doing science "the right way", with testable hypotheses, real experiments, mathematical models, and non-arbitrary classification (well, as much as any science).
It's unfortunate that there is no commonly-accepted terminology (at least none that I'm aware of) to differentiate between the two... because there are stark differences. On the one hand we have researchers who are abstractly studying how the brain and mind work; on the other we have people trying to use that research to heal mental illness. Obviously the two groups can help each other (and overlap frequently), but I would be careful about saying that psychology isn't scientific, merely because the application of that knowledge-base to fixing real-world problems is murky and subjective.
Physics is quite objective and amoral; but the application of physics to the real-world (e.g. mechanical engineering to build a bridge) will of course include some moral component (e.g. in terms of deciding whether money or safety is more important).
What does the halting problem have to do with this?
The halting problem might be relevant if you absolutely needed a provably (in the mathematical sense) perfect analysis of the algorithm that a binary implements. But any practical implementation of mawlare detection wouldn't need to be perfect: it would just have to disassemble the binary and pull out algorithmic motifs, and check these against a database (where the motif "loop over values and store to single file" is okay but "loop over all files on disk and insert code into each one" is not okay).
Bringing up the halting problem is basically saying "there are theoretical reasons why you cannot build a perfect algorithm classifier... so why bother trying?" We don't need perfect. We just need useful.
This is rather offtopic, but I'd like to say it anyways...
It used to be that the Slashdot frontpage was riddled with dupes. We complained every time that the editors were idiots, that they should have a system in place to recognize dupes. Many people even said that they emailed regarding the dupes while they were in the mysterious future, and yet the still hit the front page. Thus was born the Slashdot meme about every story being eventually duped.
But, I have to say that over the last while, the number of dupes is way, way down. The firehose and tagging seem to have alot to do with it--dupes are flagged earlier in the process, giving the editors the feedback they need.
Dupes are not entirely eliminated, but the frequency is down. So I'd like to say: thanks to the Slashdot staff for fixing the issue that we complained about. We are a whiny bunch, and it's too easy for us to complain but then forget to appreciate the things that are fixed (or have always been good). So, again, good work on the dupe reduction.
I'm not so sure about that. I mean, the military currently uses a whole bunch of stealth technology against their enemies: everything from simple paint color and camouflage, to radar-reflective stealth paint or ultra-quiet engines for submarines. None of these are perfect, but all are useful.
You may not be able to make yourself 100% invisible to an enemy that has good tech, but as long as you can give yourself an advantage in hiding, it's worth using. The "advantage" could be increased survival (enemy hit accuracy is reduced), better range (you can get closer before being detected), or maybe just the cost to the enemy for them to launch all the overhead imaging and use all magnetic field sensing equipment you just mentioned.
If cloaking became viable, it would definitely be used by the military against other high-tech enemies. In battle, every advantage counts.