there's a separate link at their firmware download page for the DIR-655 that says (in plain view, in a sensible spot): Click here for Firmware 1.21 WITHOUT SecureSpot 2.0
Well, I highly doubt that most customers know what "SecureSpot" is. So how are they supposed to know to download the non-annoying firmware update? Of course, you may say that this is the customer's problem: they should read up on all the features that are being installed in the firmware update, and be sure that this is really what they want, etc.
And, yes, in principle everyone should read every line of each and every EULA.
The fact is that any reasonable person would expect a firmware update to only fix bugs and security flaws. It would not be normal to expect entirely new features to be installed, and it is certainly abnormal for the new "feature" to actually include nagware that prompts you to pay for some new service.
The point here is that what they are doing is sleazy. The default configuration should have that redirect turned off. The link for a "without SecureSpot" firmware is nice, but the fact is that 99.9% of users will only notice that after they have already installed, and been annoyed by, the default update.
It's an annoying thing to do with a firmware update. And in that sense, it's a reason to not do business with them.
Perhaps, but you can have multiple PDF readers installed. And in terms of security, it's usually best to use the simplest application that will work.
So basically you could use FoxIt or Sumatra PDF to open most PDFs. And then for the rare one that uses some advanced stuff, you can fire up Acrobat. The fact is that most of the stuff that Acrobat supports that other PDF readers don't involves some kind of scripting. And really you shouldn't be running any scripts (even those that are, in principle, sandboxed) unless you have reason to trust them.
So a sensible strategy would seem to be that you open 99% of PDFs with a simpler reader, and only use Acrobat on the few that really need it, and only if the source of the PDF is trustworthy in your estimation.
(Yeah, I know... it's a bit of a pain to have multiple programs that do the same thing. In principle you "shouldn't have to" in the sense that your PDF reader should be secure. But in reality it seems like a reasonable precaution.)
In terms of navigation directories efficiently, I find that "cd -" is often forgotten (changes directory to your previous directory). I personally find it very useful, and couldn't live without it!
The developer blog mentions that the sheer number of options in the open source world actually makes this difficult to accomplish.
I call BS.
The diversity of options is only a problem if you try to specifically target a particular configuration. Doing so is dumb--In fact I can't think of any good reason to do so. (Hint: DRM is not a good reason.)
For instance, if you just have a link to a standard media file (e.g. mpeg) that the user can download, then you're done. The user can then use whatever browser they want, with whatever media player they want, and whatever operating system they want. The diversity of options is then the user's problem, not the distributors.
Invariably these "there are too many options to support" complaints arise because people have ulterior motives in wanting to target the OS/software/format more specifically (DRM, lockin, user tracking, advertising, promoting a particular OS/software/format, etc.).
(Note that I don't want the negative tone of my post to take anything away from the announcement. It is surely a good thing that they are working to make their shows available to us Linux users. That's great! But if they are truly going with a non-DRM solution, then why the heck don't they just skip the middleman and let everyone just download a simple file?)
Yes, many Open Source projects will require the funding of someone with a clear vision.
Then again, that's true of non-open source software. In fact, that's true of business in general. You always need some people with money, and the willingness to invest it. You need the people with money to have the vision that the project will be successful and useful.
Without the strong investment from those with deep pockets, can Open Source software progress at the rate needed to remain viable in the enterprise?
The companies that support and partner with open source projects do so because they expect a real return on investment. Google funds Firefox because they believe that this will ultimately help their bottom line (and it makes sense: competition in the browser space will lead to better products that make their web apps all the more attractive; they also get default search status in the browser).
This is just like in any other business: they will receive funding because the investors expect that this will benefit them (either directly through dividends/stock-value, or indirectly by otherwise strengthening their business, etc.).
So, I would say that most large-scale software projects are going to need some cash injections, just like any venture/business. The difference being that free/open-source projects can benefit both from corporate partnerships, and sources of "good will" funding (donations of money or code, free advertising/advocacy, etc.).
1. YouTube doesn't scan content and aggregate it. The users voluntarily upload the content to YouTube. This shows implicit consent for YouTube to redistribute. The terms of service make this consent explicit.
2. YouTube does actually share revenue with channels that get lots of views. It's called the YouTube Partner Program.
I think they'll be happy if it's invisible to the people who have bought the content and are playing by their rules.
Yeah but that's easy: the "rules" are by their nature restrictive. The DRM will indeed be invisible when you only buy songs via the approved website, only access them with the approved software, and only play them on the approved devices. Oh, and one of the rules is that you are not allowed to complain when the authentication servers are decommissioned and you have to re-buy your entire media collection.
The problem is that the distributor's rules are way too restrictive for the average person. A person wants to buy the content they like, buy the device they like, and then enjoy the content anytime and anyplace they like. In short, they hate the distributor's overly-restrictive rules.
To summarize: it's easy for a distributor to make DRM that is invisible when you play by their rules. It's impossible to make DRM that is invisible when the user plays by their own rules. Put otherwise: either the DRM will be annoyingly visible, or the rules will be annoyingly visible. Either way, the user is annoyed.
That sounds great... but I have trouble imagining a DRM system that could actually make the distinctions you mention (given that we haven't quite solved that whole artificial intelligence problem).
For instance, the average person (as you mention) is going to want to be able to copy a song to their spouse's computer over the LAN. But how exactly does the DRM recognize the difference between a copy to the spouse's computer, a copy to a friend's computer, a copy to your work computer, a copy to a coworker's computer, a copy to a stranger's computer, or a copy to a redistribution server?
The only way I can think is with encrypted content, and then by defining "permission zones" or somesuch, where various devices get authorized as part of a zone, with restrictions on how many devices can be registered in a zone at a time (so that you can't add your closest 30,000 P2P friends into your zone). But managing these zones isn't going to be invisible. You'll be adding new devices as they are purchased, removing old devices as they are sold/discarded (do you have to prove you've erased the previously authorized content?), flashing firmware to re-authorize devices (because keys will have been revoked), using a restricted set of software (that is able to understand the DRM), waiting for network connections to be available (because it's been too long since the last time the device phoned-home), and so on. The user will notice.
I don't think there is any scheme that is sufficiently permissive that users will never notice it, yet sufficiently restrictive to actually put a dent in the "really bad copying" (commercial redistribution, uploading to P2P networks,...). And TFA does nothing to actually address this issue: how does the software differentiate between good copies and bad copies.
Answer: computers can't. Actually, given the confusion and disagreement around copyright law, evidently humans can't either.
Which I found especially hilarious/frustrating when I was required to upgrade the version of Norton on a bunch of lab computers. The upgrade wouldn't work, and told me I had to uninstall the previous version. Turns out uninstalling the previous version was unbelievably difficult. The auto-uninstall didn't work. The Norton removal tool didn't work. Finally I had to follow a series of manual step-by-step instructions about what files to delete and what registry keys to modify.
And after all this pain and suffering to remove Norton... I had to install a new version. (That I knew would be a pain to eventually uninstall or upgrade.)
Needless to say I now avoid Norton like the plague. Yet I would argue that Norton/Symantec is widespread not only because of default installs--but because they seem to do a good job marketing to the higher-ups. They win large-scale deployment contracts, where the software annoys end users and many admins, but looks good and secure on paper, I guess.
Also, Mason is pretty good about practicing what he preaches in that you can pirate his book on his own website
I dislike the use of the vague and slanted term "pirate" in place of the more exact "copyright infringement".
But the use in the summary is even worse. If he's freely offering the content, then those who download it are not pirating (even by the inaccurate, though generally-used, definition). Then are downloading it with permission.
(It's like someone giving out free food samples at a grocery store, and then saying "go ahead, steal another.")
it can be said that something that is only a little dangerous (relying on nuclear weapons for security, which has worked for almost 60 years)
As he points out in the article, "a little dangerous" isn't good enough. A 99% per year safety margin for nuclear deterrence would be small, but actually a 1% chance of nuclear war each year isn't an acceptable risk... and moreover is not a sustainable long-term strategy. As he points out, although nuclear weapons have not led to nuclear war in the last 50-60 years, that really only establishes that "the chance of nuclear war is somewhere between zero and 6% per year". Those are terrible odds.
His point is indeed that the risk of nuclear wars is small, and that the risks from nuclear power plants are VERY small (he claims 1000-times smaller). But his bigger point is that the risk from nuclear war is not small enough.
Having said all that, I'm certainly not convinced that the actual risk is 1%. Then again, his overall point is that we need better risk assessment: rather than just assuming something like "if it hasn't happened in the last 50 years, it will never happen", he advocates rigorously developing decision trees and assigning best-guess probabilities to all nodes. Further he advocates taking action that will continually reduce the threat.
His goals are laudable and his analysis isn't wrong. But it's obviously not easy to come up with really satisfactory estimates of the risks.
Highlighting a security problem means they have to put in the effort to fix what you brought to their attention, or threaten you to STFU.
Only because people have no clue about security.
When most people hear about a security vulnerability, they do indeed think that they have two options:
1. Fix it.
2. Bury all information about it.
The reality is that the third option is the one that is frequently the right one: Acknowledge it and move on. Security vulnerabilities are everywhere. It's better to be aware of them than not. And yes it's a good idea to fix them if doing so is not overly onerous. However it is not always necessary to fix them all.
For instance a store may not put magnetic tags on the chocolate bars they sell. Is the correct solution to tag everything? No, it's probably better to rely on people's (generally) good nature, the vigilance of employees, and then simply accept that a few chocolate bars will get stolen. It is cheaper (and less annoying for customers) to accept the losses. Or a movie theater can be tricked by having people exit with already-used tickets, and bring other friends in using them. Is the correct solution to require that everyone going to the movies show ID every time they enter the theater? No, it's better to simply accept the occasional teenager who "beats the system." Oftentimes the best "security" is just social norms. (Think of how much harm you could do, how much stuff you could steal, on a daily basis if you felt no remorse.)
Many geeks make this mistake, too (possibly because they are used to thinking about computer security, where applying a fix usually makes sense because the coding cost is fairly small compared to the damage that a exploit can cause).
I wish more people understood that security is a tradeoff, so that when someone points out a security hole, the people in charge can be honest and either say "that's not a sufficient concern to warrant fixing" or "that's a good point--we'll fix that now".
Computers of this size and form-factor are not totally unprecedented. Things like the Nokia N810 internet tablet are similar. (QWERTY keyboard, fits in your pocket, WiFi or bluetooth connectivity...). Also, many smartphones have all the features and functionality of this device (including having a physical keyboard, etc.) with the advantage of direct connectivity through the cell network.
The only thing this new device can offer is a somewhat lower price ($175 instead of >$400 for the N810). But I think this device will only appeal to a very small market (most people would prefer to spend a bit more for a more capable device, or get something with a bigger screen/keyboard).
Actually we don't even have a satisfactory definition of "life". Just look at the heated arguments about artificial intelligence or abortion to get a flavor for the lack of consensus on the issue.
There may be organizations of matter that are highly complex but not obviously sentient. Maybe species that are so long-lived and slow-moving that we overlook them as just another rock. Or maybe their composition will be so different (crystal? glass? gas?) that we will dismiss them. Or maybe they will be fairly similar to us (made of carbon, etc.) but we won't recognize their behavior as life-like because their customs are so alien.
Consider for a moment questions like "Is the Internet alive?" (It is a highly complex, interconnected system that exhibits emergent behavior. So is it alive?) "Is the galaxy alive?" (The extremely slow interactions between stars and dust clouds could encode information, forming some kind of creature/mind...) "Is a human alive?" (Why?)
And even if we discovered a bunch of bipedal humanoids made of carbon, there would still end up being many humans making arguments that they are not really alive--because they lack a "soul" or the divine touch of god or something like that.
I'm bothered by the fact that in most of these discussions about intelligent aliens, the question of "how do you recognize life" is taken as a given. As if it's obvious that "we'll know it when we see it". I question that assertion. For these kinds of debates to have any meaning, we need to decide what our criteria for "life" (and "intelligence" and "advanced") really are.
Actually it may be that we've learned our lesson. Even though nanotechnology is still a very nascent field, serious efforts are already underway to measure the health effects and safety concerns for these kinds of materials. For instance, in the United States, NIOSH (a branch of the CDC tasked with evaluating work-related risks) has an effort underway to quantify the effects of nano-materials on people (link). There are similar efforts worldwide for this "NanoEHS" issue (e.g. this).
Only time will tell, of course. But as someone working in the broad field of "nano", I can say that health, safety, and environmental impact are already a part of our research plans. There are considerable efforts to make sure we understand the impact of these materials before sending them to market. Also, since we are the ones working with these materials daily, we are certainly concerned with any possible toxicity.
Mistakes may still be made (e.g. a product released ends up having an unforeseen interaction with some other material/drug/etc.), but presently it seems that agencies are being appropriately proactive in terms of assessing risk before commercialization is even a serious consideration.
According to the article, buckypaper "conducts electricity like copper or silicon." So it's either a conductor or an insulator. The article smells like roses or shit.
Actually, carbon nanotubes can be either metallic or semiconducting, depending on the type. (Different "types" have a different arrangement of the graphene hexagons with respect to the tube axis: zigzag, armchair, or chiral.) So it is in fact correct to say that carbon nanotubes are either conductors or semiconductors.
Buckypaper is made of nanotubes, so it will be conducting or semiconducting depending on its composition. Most nanotube production techniques create a mixture of tube types, so most samples of buckypaper will be a mixture of metallic and semiconducting components. The final electrical properties will then of course depend on the relative inclusion of the various types. (As well as other things, like alignment of the tubes, and interactions or bridging between tubes.) This is a virtue of buckypaper, in fact, since (in principle) we can tune the electrical properties as required for a particular application (while maintaining nearly the same mechanical performance).
(I agree that the article is poorly worded. The sentence is technically correct, but that's probably an accident.)
I certainly take exception to the conclusion that "flat universe = infinite mass"
By "flat" I mean space has no curvature and is thus effectively Euclidean. This means (as far as I understand it), that space continues on forever in each direction. It doesn't wrap back upon itself and reconnect (which is what would happen for a curved geometry like spherical space). So "flat universe" = "infinite space." Now, if the matter density of the universe is roughly homogeneous on large scales (which it is within our Hubble volume), then every region of space has some mass. So "flat universe" + "homogeneous matter" = "infinite mass".
Our best measurements have not detected any curvature to the universe at large scales. It looks flat. The current consensus in cosmology, as far as I can tell, is that the universe is thus probably infinite in volume and mass. (Of course, it could be a spherical geometry with a very, very small curvature... in which case it would not be infinite, but just very, very large.) Quoting from an expert in the field, Max Tegmark:
A generic prediction of inflation is an infinite ergodic universe, which contains Hubble volumes realizing all initial conditions
In this paper he's discussing the implications in terms of "copies" of Hubble volumes, but the basic conclusion of "infinite universe" appears robust:
..the simplest and most popular cosmological model today predicts that [there are copies of our galaxy] about 10^(10^29) meters from here. This does not even assume speculative modern physics, merely that space is infinite and rather uniformly filled with matter as indicated by recent astronomical observations. Your alter ego is simply a prediction of the so-called concordance model of cosmology,
which agrees with all current observational evidence and is used as the basis for most calculations and simulations presented at cosmology conferences. In contrast, alternatives such as a fractal universe, a closed universe and a multiply connected universe have been seriously challenged by observations.
(Emphasis added.)
And:
Although the implications may seem crazy and counter-intuitive, this spatially infinite cosmological model is in fact the simplest and most popular one on the market today. It is part of the cosmological concordance model, which agrees with all current observational evidence and is used as the basis for most calculations and simulations presented at cosmology conferences.
I believe you are correct that (absent dark energy) a flat universe should expand forever but with a continually decreasing rate. Of course dark energy modifies this, causing a negative pressure that accelerates the expansion at later times. Modern measurements indicate that the mass density of the universe is very close to the critical value, and that space is flat, but that dark energy is accelerating the expansion.
The cosmological models relate not to the total mass of the universe, but to the density. If the density of the universe is at the critical value, then it is flat and expands forever (absent dark energy). If the density is above the critical value, the universe has positive curvature, is closed and will collapse. If the density is below the critical value then space has negative curvature, is open, and expands forever. The universe can have infinite volume and infinite mass while still having finite density.
An infinite amount of mass would require an infinite amount of energy during the Big Bang. That's just absurd.
Counter-intuitive, yes. But that's what the data suggest. (Strictly, the data simply don't rule out an infinite universe. A very, very large universe with curvature too small to measure is also consistent with observation. So I suppose on aesthetic grounds you could prefer one over the other. But the data so far do not rule out either.)
If the universe is smaller than its Schwarzschild radius, it should collapse into a singularity. It hasn't, so it apparently isn't.
As mentioned here, the concept of a Schwarzschild radius is one limiting case of Einstein's equations of general relativity. It is a useful concept with various rules-of-thumb, but one must be careful in applying it to all situations. In particular, the approximation breaks down, and a full treatment using the equations of general relativity is instead necessary, for "extreme" situations (like inside a black hole, during the big bang, when applied to the entire universe, etc.).
More specifically (this site seems to explain it somewhat), the "Schwarzschild black hole" is just one solution to the equations of general relativity--it is a limiting case for nominally static matter (that is also non-rotating, spherically symmetric). Other solutions are required in other cases (e.g. the Kerr solution for rotating black holes). The Schwarzschild solution doesn't apply to dynamic systems (e.g. rapidly expanding matter). In particular the big bang and subsequent expansion of the universe represent a different solution to the equations of GR. This solution provides for a roughly flat space but massive expansion (hence highly curved spacetime, as one would expect for such high mass-density). Our best understanding suggests that inflation occurred (where space was expanding faster than the speed of light, although light/energy/matter/information was still constrained by c).
In my previous post I was just pointing out that the expected size for the Schwarzschild radius is very large. However that is based on a naive application of the usual rules-of-thumb. The big bang, if you will, is extreme enough that it requires a more careful treatment. Moreover, our best data right now suggests that the universe is roughly flat and infinite (and thus with infinite or at least extremely large mass), meaning that there is probably no meaningful way to apply the "Schwarzschild radius" concept to it.
Disclaimer: I'm not a cosmologist. Hopefully I didn't make a mistake.
Uptime contests are fun and all. But I'm always suspicious of systems with very long uptimes: they probably haven't had a kernel update since that last reboot... meaning that they are a highly vulnerable box.
My Ubuntu machine has uptimes that are about 30-90 days, which is entirely based on new kernel releases. I've never had an unintended reboot (e.g. from a freeze or crash).
(Yes, there are methods of updating the kernel without rebooting... but most people with massive uptimes seem to achieve it not by using these tricks but rather by not touching the box.)
What I want to know, is how did the universe expand beyond its own swartzchild radius?
It didn't.
To compute the Schwarzschild radius of the universe, we need to know its mass. Recent measurements suggest that the universe is flat, and so may have infinite mass. However at a minimum we can count up the mass within the observable universe. The observable stars in the universe have a mass of ~2*10^52, but they are overwhelmed by dark matter, which brings the total mass within our observation volume to ~4*10^53 kg. So the Schwarzschild radius for the universe is:
r = (2*G*m)/(c^2) = 2*(6.7E-11 m^3kg^-1s^-2)*(4E53 kg)/(3E8 m/s)^2 = 6E26 m = 60 billion light-years.
Since the observable universe is ~46 billion light-years in radius, this means that the Schwarzschild radius of the universe is bigger than what we consider to be "the universe." In other words, we are well within the Schwarzschild radius, leading some people to describe the universe itself as a massive black hole that we are actually inside of.
The universe probably has a mass larger than what we can observe, making the radius even larger than the above estimate. If the universe truly has infinite mass, then the radius is infinite. In other words, the universe may not have a Schwarzschild radius at all.
The question is phrased in a sort of black/white manner: either you fight tooth-and-nail to maintain maximum privacy, or you give up and sign up for every crazy privacy-eroding service.
The obvious answer is "all things in moderation." I consider myself privacy-conscious. I don't run Windows. I do use Facebook and Gmail. However I use them with privacy in mind. So my Facebook profile has very little information, has privacy options set quite high, and I only accept friend invites from people that I reasonably trust. (So many people seem to get sucked into the "I need my friend count to be higher" game--which invariable means accepting invites from strangers.)
My strategy works, more or less. There are times when friends reveal information about me online I would rather they didn't (e.g. tagging me in photos on Facebook). But you can't completely prevent these kinds of things. In the same way that friends can give out your phone number or gossip about you in real-life, there will be some privacy loss online. The goal should be to keep things private without it becoming a burden to do so.
It sounds like you're taking the privacy thing to far--to the point that it's harder for you to socialize and enjoy life. So loosen your rules a little bit. Remember that every company (the power company, the cable company, your bank, etc.) has tons of privacy-eroding data on you. Online companies will also get some privacy-eroding data. But as long as you keep it within reasonable bounds, then it won't cause a problem.
Remember, privacy isn't really something that has to be maintained for its own sake. Privacy is a means for you to enjoy your life free from bother, and to prevent people harming/taking advantage of you. Calibrate accordingly.
A small loss of privacy is okay if it achieves the greater objective of making you happy.
a TV tuner that can be plugged into a PC -- will let Net viewers of the likes of Hulu.com and ABC.com skip commercials in the nascent medium of online video.
What? Why do I need a TiVo TV Tuner box to watch online videos? Stripping commercials from online streaming video sounds like a software task. And saving the streaming video so that you can jump past the commercials doesn't require any special TiVo magic (whether hardware, subscription, or software). Will we see software and utilities that allow you to skip ads on online video? Probably. But what does this have to do with a TV-Tuner card for your PC?
The article also asks some nonsensical questions:
Does the arrival of the TiVo PC set-top box represent the final convergence of television and Internet video?
No. TV-Tuner cards and online video have existed for awhile. I don't see how a TiVo box changes anything. Yes, it might make "TV on your computer" more accessible to the masses... but that isn't a "final convergence" of anything, really. Sure, the lines are blurring between TV and Internet. And TiVo is part of that inevitable change. But this box isn't a revolution.
What will be the impact of TiVo's new device on the online video economy?
None. It's a TV-Tuner card, isn't it? (People watch Hulu because they don't want to pay for the equivalent cable channels.)
Will TiVo be remembered as the company that helped slaughter the advertising golden goose that has enriched the broadcasting industry for the last 50 years?
Doubtful. TiVo hasn't demolished TV ads yet. Strangely, PVRs in general haven't either. And AdBlock hasn't demolished web ads. These are all part of the arms race which keep ads sufficiently non-annoying that a sizeable fraction of the population doesn't bother avoiding. There will always be people who avoid them. But most people don't bother.
Add to this the fact that part of TiVo's strategy is to deliver ads to customers somehow... I hardly think that this new box changes much for the ad industry.
FYI -- there is some sensational press out there that makes it sound like we're planning to break/have already broken the 2nd law of thermodynamics. This is, of course, absurd -- but I think it's imperative we set the record straight before everyone starts jumping all over us.
The context.... a colleague and I received funding to study non-equilibrium switching invoking a concept called 'Brownian Ratchets' that has been well studied in nonequilibrium statistical physics over the years. The potential benefactor of this study is the chip industry, in a very broad way, as it is worried about rapidly increasing thermal budgets (chips are becoming very hot). We're simply trying to examine the physics of Brownian ratchets in a device context. A popular model for heat dissipation in binary switching (proposed by Victor Zhirnov and co-workers) looks at a two well one barrier geometry, with a gate controlling the barrier and a drain controlling the overall directionality. Each such raising and lowering of a barrier at the end dissipates energy irreversibly (during the reset step where one erases information), leading to a kTln2 dissipation per operation (kT is the thermal energy). And this analysis is usually done by assuming that you wait after you raise or lower a barrier and then let the electrons move and reach equilibrium with the surroundings. The analysis is thus based on equilibrium Boltzmann statistics -- since the electron was at equilibrium before a computation and reaches equilibrium after. What is not clear is what happens during the non-equilibrium transition phase, or if you switch before the equilibrium is reached. The aim of the study is not to attempt to deviate from cherished physical principles, but on the contrary to see what these cherished principles posit for such a situation. A ratchet is known to be able to rectify non-equilibrium noise to produce directed motion by transducing spatial asymmetries in the system (this is well recognized in nonequilibrium statistical mechanics and has been mulled over for years). The physics is well studied, but the context is perhaps new... we are interested in seeing if rectifying such non-equilibrium noise (as a ratchet does) can perhaps shave off some of the power dissipation limit associated with a drain bias in the regular example.
This is, of course, still at a toy model -- we need to worry about how to deal with compatibility of input and output, for example. Simply put, we don't know if this will bear fruit for the big picture of low-power device operation, but it's worth investigating.
That's about it... but then, cooling laptops as hot as the sun through the power of thinking or by breaking the 2nd law sounds fancier... doesn't it?
(Emphasis added.)
So this seems like still very early work (just an idea, really)... and it appears that the intention is to build new kinds of switches (e.g. transistors) that exploit the fact that switching is inherently non-equilibrium, and extract some of the energy that is dissipated during these switching events.
Yeah the article is unclear. Here's my best shot at clarification:
A "Brownian ratchet" is a thought-device about extracting energy from the random Brownian motion of a hot gas. Similar to Maxwell's Demon, it can't work in a system at equilibrium. Without a temperature gradient, there is no way to extract useful work. The ratchet will be undergoing random motion equal in magnitude to the energy we hope to extract, so we can't actually extract anything.
However, if we're not at equilibrium, the rules are different. These researchers are talking about "non-equilibrium Brownian ratchets", which you could also call a "Brownian motor". In an non-equilibrium situation, you will have a gradient of heat or chemical potential that could, in principle, be converted into useful work.
So my guess is the researchers are trying to do something like:
1. Build devices that exploit the temperature gradient that exists in the device. So a bunch of nano-sized ratchets that convert the heat gradient on the outside of the chip (relative to the cool air) to recharge a capacitor or something.
2. Build switching elements (e.g. transistors) that directly store the excess switching energy in some way. That is, build switching elements that both do computational switching, but immediately utilize the resulting temperature gradient of the dissipated heat.
In either case, all they are suggesting is to take advantage of the heat gradients that inherently occur when you have imperfect switching elements dissipating heat. It's not really that novel, conceptually... although if they actually have a specific way to do this in mind, then that could be quite interesting.
Slashdot Mirror
there's a separate link at their firmware download page for the DIR-655 that says (in plain view, in a sensible spot): Click here for Firmware 1.21 WITHOUT SecureSpot 2.0
Well, I highly doubt that most customers know what "SecureSpot" is. So how are they supposed to know to download the non-annoying firmware update? Of course, you may say that this is the customer's problem: they should read up on all the features that are being installed in the firmware update, and be sure that this is really what they want, etc.
And, yes, in principle everyone should read every line of each and every EULA.
The fact is that any reasonable person would expect a firmware update to only fix bugs and security flaws. It would not be normal to expect entirely new features to be installed, and it is certainly abnormal for the new "feature" to actually include nagware that prompts you to pay for some new service.
The point here is that what they are doing is sleazy. The default configuration should have that redirect turned off. The link for a "without SecureSpot" firmware is nice, but the fact is that 99.9% of users will only notice that after they have already installed, and been annoyed by, the default update.
It's an annoying thing to do with a firmware update. And in that sense, it's a reason to not do business with them.
Perhaps, but you can have multiple PDF readers installed. And in terms of security, it's usually best to use the simplest application that will work.
So basically you could use FoxIt or Sumatra PDF to open most PDFs. And then for the rare one that uses some advanced stuff, you can fire up Acrobat. The fact is that most of the stuff that Acrobat supports that other PDF readers don't involves some kind of scripting. And really you shouldn't be running any scripts (even those that are, in principle, sandboxed) unless you have reason to trust them.
So a sensible strategy would seem to be that you open 99% of PDFs with a simpler reader, and only use Acrobat on the few that really need it, and only if the source of the PDF is trustworthy in your estimation.
(Yeah, I know... it's a bit of a pain to have multiple programs that do the same thing. In principle you "shouldn't have to" in the sense that your PDF reader should be secure. But in reality it seems like a reasonable precaution.)
Another option for PDF reading on Windows is Sumatra PDF (if you prefer open-source).
In terms of navigation directories efficiently, I find that "cd -" is often forgotten (changes directory to your previous directory). I personally find it very useful, and couldn't live without it!
The developer blog mentions that the sheer number of options in the open source world actually makes this difficult to accomplish.
I call BS.
The diversity of options is only a problem if you try to specifically target a particular configuration. Doing so is dumb--In fact I can't think of any good reason to do so. (Hint: DRM is not a good reason.)
For instance, if you just have a link to a standard media file (e.g. mpeg) that the user can download, then you're done. The user can then use whatever browser they want, with whatever media player they want, and whatever operating system they want. The diversity of options is then the user's problem, not the distributors.
Invariably these "there are too many options to support" complaints arise because people have ulterior motives in wanting to target the OS/software/format more specifically (DRM, lockin, user tracking, advertising, promoting a particular OS/software/format, etc.).
(Note that I don't want the negative tone of my post to take anything away from the announcement. It is surely a good thing that they are working to make their shows available to us Linux users. That's great! But if they are truly going with a non-DRM solution, then why the heck don't they just skip the middleman and let everyone just download a simple file?)
Without the strong investment from those with deep pockets, can Open Source software progress at the rate needed to remain viable in the enterprise?
The companies that support and partner with open source projects do so because they expect a real return on investment. Google funds Firefox because they believe that this will ultimately help their bottom line (and it makes sense: competition in the browser space will lead to better products that make their web apps all the more attractive; they also get default search status in the browser).
This is just like in any other business: they will receive funding because the investors expect that this will benefit them (either directly through dividends/stock-value, or indirectly by otherwise strengthening their business, etc.).
So, I would say that most large-scale software projects are going to need some cash injections, just like any venture/business. The difference being that free/open-source projects can benefit both from corporate partnerships, and sources of "good will" funding (donations of money or code, free advertising/advocacy, etc.).
1. YouTube doesn't scan content and aggregate it. The users voluntarily upload the content to YouTube. This shows implicit consent for YouTube to redistribute. The terms of service make this consent explicit.
2. YouTube does actually share revenue with channels that get lots of views. It's called the YouTube Partner Program.
I think they'll be happy if it's invisible to the people who have bought the content and are playing by their rules.
Yeah but that's easy: the "rules" are by their nature restrictive. The DRM will indeed be invisible when you only buy songs via the approved website, only access them with the approved software, and only play them on the approved devices. Oh, and one of the rules is that you are not allowed to complain when the authentication servers are decommissioned and you have to re-buy your entire media collection.
The problem is that the distributor's rules are way too restrictive for the average person. A person wants to buy the content they like, buy the device they like, and then enjoy the content anytime and anyplace they like. In short, they hate the distributor's overly-restrictive rules.
To summarize: it's easy for a distributor to make DRM that is invisible when you play by their rules. It's impossible to make DRM that is invisible when the user plays by their own rules. Put otherwise: either the DRM will be annoyingly visible, or the rules will be annoyingly visible. Either way, the user is annoyed.
That sounds great... but I have trouble imagining a DRM system that could actually make the distinctions you mention (given that we haven't quite solved that whole artificial intelligence problem).
For instance, the average person (as you mention) is going to want to be able to copy a song to their spouse's computer over the LAN. But how exactly does the DRM recognize the difference between a copy to the spouse's computer, a copy to a friend's computer, a copy to your work computer, a copy to a coworker's computer, a copy to a stranger's computer, or a copy to a redistribution server?
The only way I can think is with encrypted content, and then by defining "permission zones" or somesuch, where various devices get authorized as part of a zone, with restrictions on how many devices can be registered in a zone at a time (so that you can't add your closest 30,000 P2P friends into your zone). But managing these zones isn't going to be invisible. You'll be adding new devices as they are purchased, removing old devices as they are sold/discarded (do you have to prove you've erased the previously authorized content?), flashing firmware to re-authorize devices (because keys will have been revoked), using a restricted set of software (that is able to understand the DRM), waiting for network connections to be available (because it's been too long since the last time the device phoned-home), and so on. The user will notice.
I don't think there is any scheme that is sufficiently permissive that users will never notice it, yet sufficiently restrictive to actually put a dent in the "really bad copying" (commercial redistribution, uploading to P2P networks, ...). And TFA does nothing to actually address this issue: how does the software differentiate between good copies and bad copies.
Answer: computers can't. Actually, given the confusion and disagreement around copyright law, evidently humans can't either.
Norton is ... ALMOST IMPOSSIBLE TO REMOVE.
Which I found especially hilarious/frustrating when I was required to upgrade the version of Norton on a bunch of lab computers. The upgrade wouldn't work, and told me I had to uninstall the previous version. Turns out uninstalling the previous version was unbelievably difficult. The auto-uninstall didn't work. The Norton removal tool didn't work. Finally I had to follow a series of manual step-by-step instructions about what files to delete and what registry keys to modify.
And after all this pain and suffering to remove Norton... I had to install a new version. (That I knew would be a pain to eventually uninstall or upgrade.)
Needless to say I now avoid Norton like the plague. Yet I would argue that Norton/Symantec is widespread not only because of default installs--but because they seem to do a good job marketing to the higher-ups. They win large-scale deployment contracts, where the software annoys end users and many admins, but looks good and secure on paper, I guess.
Also, Mason is pretty good about practicing what he preaches in that you can pirate his book on his own website
I dislike the use of the vague and slanted term "pirate" in place of the more exact "copyright infringement".
But the use in the summary is even worse. If he's freely offering the content, then those who download it are not pirating (even by the inaccurate, though generally-used, definition). Then are downloading it with permission.
(It's like someone giving out free food samples at a grocery store, and then saying "go ahead, steal another.")
it can be said that something that is only a little dangerous (relying on nuclear weapons for security, which has worked for almost 60 years)
As he points out in the article, "a little dangerous" isn't good enough. A 99% per year safety margin for nuclear deterrence would be small, but actually a 1% chance of nuclear war each year isn't an acceptable risk... and moreover is not a sustainable long-term strategy. As he points out, although nuclear weapons have not led to nuclear war in the last 50-60 years, that really only establishes that "the chance of nuclear war is somewhere between zero and 6% per year". Those are terrible odds.
His point is indeed that the risk of nuclear wars is small, and that the risks from nuclear power plants are VERY small (he claims 1000-times smaller). But his bigger point is that the risk from nuclear war is not small enough.
Having said all that, I'm certainly not convinced that the actual risk is 1%. Then again, his overall point is that we need better risk assessment: rather than just assuming something like "if it hasn't happened in the last 50 years, it will never happen", he advocates rigorously developing decision trees and assigning best-guess probabilities to all nodes. Further he advocates taking action that will continually reduce the threat.
His goals are laudable and his analysis isn't wrong. But it's obviously not easy to come up with really satisfactory estimates of the risks.
Highlighting a security problem means they have to put in the effort to fix what you brought to their attention, or threaten you to STFU.
Only because people have no clue about security.
When most people hear about a security vulnerability, they do indeed think that they have two options:
1. Fix it.
2. Bury all information about it.
The reality is that the third option is the one that is frequently the right one: Acknowledge it and move on. Security vulnerabilities are everywhere. It's better to be aware of them than not. And yes it's a good idea to fix them if doing so is not overly onerous. However it is not always necessary to fix them all.
For instance a store may not put magnetic tags on the chocolate bars they sell. Is the correct solution to tag everything? No, it's probably better to rely on people's (generally) good nature, the vigilance of employees, and then simply accept that a few chocolate bars will get stolen. It is cheaper (and less annoying for customers) to accept the losses. Or a movie theater can be tricked by having people exit with already-used tickets, and bring other friends in using them. Is the correct solution to require that everyone going to the movies show ID every time they enter the theater? No, it's better to simply accept the occasional teenager who "beats the system." Oftentimes the best "security" is just social norms. (Think of how much harm you could do, how much stuff you could steal, on a daily basis if you felt no remorse.)
Many geeks make this mistake, too (possibly because they are used to thinking about computer security, where applying a fix usually makes sense because the coding cost is fairly small compared to the damage that a exploit can cause).
I wish more people understood that security is a tradeoff, so that when someone points out a security hole, the people in charge can be honest and either say "that's not a sufficient concern to warrant fixing" or "that's a good point--we'll fix that now".
Computers of this size and form-factor are not totally unprecedented. Things like the Nokia N810 internet tablet are similar. (QWERTY keyboard, fits in your pocket, WiFi or bluetooth connectivity...). Also, many smartphones have all the features and functionality of this device (including having a physical keyboard, etc.) with the advantage of direct connectivity through the cell network.
The only thing this new device can offer is a somewhat lower price ($175 instead of >$400 for the N810). But I think this device will only appeal to a very small market (most people would prefer to spend a bit more for a more capable device, or get something with a bigger screen/keyboard).
Actually we don't even have a satisfactory definition of "life". Just look at the heated arguments about artificial intelligence or abortion to get a flavor for the lack of consensus on the issue.
There may be organizations of matter that are highly complex but not obviously sentient. Maybe species that are so long-lived and slow-moving that we overlook them as just another rock. Or maybe their composition will be so different (crystal? glass? gas?) that we will dismiss them. Or maybe they will be fairly similar to us (made of carbon, etc.) but we won't recognize their behavior as life-like because their customs are so alien.
Consider for a moment questions like "Is the Internet alive?" (It is a highly complex, interconnected system that exhibits emergent behavior. So is it alive?) "Is the galaxy alive?" (The extremely slow interactions between stars and dust clouds could encode information, forming some kind of creature/mind...) "Is a human alive?" (Why?)
And even if we discovered a bunch of bipedal humanoids made of carbon, there would still end up being many humans making arguments that they are not really alive--because they lack a "soul" or the divine touch of god or something like that.
I'm bothered by the fact that in most of these discussions about intelligent aliens, the question of "how do you recognize life" is taken as a given. As if it's obvious that "we'll know it when we see it". I question that assertion. For these kinds of debates to have any meaning, we need to decide what our criteria for "life" (and "intelligence" and "advanced") really are.
Actually it may be that we've learned our lesson. Even though nanotechnology is still a very nascent field, serious efforts are already underway to measure the health effects and safety concerns for these kinds of materials. For instance, in the United States, NIOSH (a branch of the CDC tasked with evaluating work-related risks) has an effort underway to quantify the effects of nano-materials on people (link). There are similar efforts worldwide for this "NanoEHS" issue (e.g. this).
Only time will tell, of course. But as someone working in the broad field of "nano", I can say that health, safety, and environmental impact are already a part of our research plans. There are considerable efforts to make sure we understand the impact of these materials before sending them to market. Also, since we are the ones working with these materials daily, we are certainly concerned with any possible toxicity.
Mistakes may still be made (e.g. a product released ends up having an unforeseen interaction with some other material/drug/etc.), but presently it seems that agencies are being appropriately proactive in terms of assessing risk before commercialization is even a serious consideration.
According to the article, buckypaper "conducts electricity like copper or silicon." So it's either a conductor or an insulator. The article smells like roses or shit.
Actually, carbon nanotubes can be either metallic or semiconducting, depending on the type. (Different "types" have a different arrangement of the graphene hexagons with respect to the tube axis: zigzag, armchair, or chiral.) So it is in fact correct to say that carbon nanotubes are either conductors or semiconductors.
Buckypaper is made of nanotubes, so it will be conducting or semiconducting depending on its composition. Most nanotube production techniques create a mixture of tube types, so most samples of buckypaper will be a mixture of metallic and semiconducting components. The final electrical properties will then of course depend on the relative inclusion of the various types. (As well as other things, like alignment of the tubes, and interactions or bridging between tubes.) This is a virtue of buckypaper, in fact, since (in principle) we can tune the electrical properties as required for a particular application (while maintaining nearly the same mechanical performance).
(I agree that the article is poorly worded. The sentence is technically correct, but that's probably an accident.)
I certainly take exception to the conclusion that "flat universe = infinite mass"
By "flat" I mean space has no curvature and is thus effectively Euclidean. This means (as far as I understand it), that space continues on forever in each direction. It doesn't wrap back upon itself and reconnect (which is what would happen for a curved geometry like spherical space). So "flat universe" = "infinite space." Now, if the matter density of the universe is roughly homogeneous on large scales (which it is within our Hubble volume), then every region of space has some mass. So "flat universe" + "homogeneous matter" = "infinite mass".
Our best measurements have not detected any curvature to the universe at large scales. It looks flat. The current consensus in cosmology, as far as I can tell, is that the universe is thus probably infinite in volume and mass. (Of course, it could be a spherical geometry with a very, very small curvature... in which case it would not be infinite, but just very, very large.) Quoting from an expert in the field, Max Tegmark:
In this paper he's discussing the implications in terms of "copies" of Hubble volumes, but the basic conclusion of "infinite universe" appears robust:
(Emphasis added.)
And:
I believe you are correct that (absent dark energy) a flat universe should expand forever but with a continually decreasing rate. Of course dark energy modifies this, causing a negative pressure that accelerates the expansion at later times. Modern measurements indicate that the mass density of the universe is very close to the critical value, and that space is flat, but that dark energy is accelerating the expansion.
The cosmological models relate not to the total mass of the universe, but to the density. If the density of the universe is at the critical value, then it is flat and expands forever (absent dark energy). If the density is above the critical value, the universe has positive curvature, is closed and will collapse. If the density is below the critical value then space has negative curvature, is open, and expands forever. The universe can have infinite volume and infinite mass while still having finite density.
An infinite amount of mass would require an infinite amount of energy during the Big Bang. That's just absurd.
Counter-intuitive, yes. But that's what the data suggest. (Strictly, the data simply don't rule out an infinite universe. A very, very large universe with curvature too small to measure is also consistent with observation. So I suppose on aesthetic grounds you could prefer one over the other. But the data so far do not rule out either.)
If the universe is smaller than its Schwarzschild radius, it should collapse into a singularity. It hasn't, so it apparently isn't.
As mentioned here, the concept of a Schwarzschild radius is one limiting case of Einstein's equations of general relativity. It is a useful concept with various rules-of-thumb, but one must be careful in applying it to all situations. In particular, the approximation breaks down, and a full treatment using the equations of general relativity is instead necessary, for "extreme" situations (like inside a black hole, during the big bang, when applied to the entire universe, etc.).
More specifically (this site seems to explain it somewhat), the "Schwarzschild black hole" is just one solution to the equations of general relativity--it is a limiting case for nominally static matter (that is also non-rotating, spherically symmetric). Other solutions are required in other cases (e.g. the Kerr solution for rotating black holes). The Schwarzschild solution doesn't apply to dynamic systems (e.g. rapidly expanding matter). In particular the big bang and subsequent expansion of the universe represent a different solution to the equations of GR. This solution provides for a roughly flat space but massive expansion (hence highly curved spacetime, as one would expect for such high mass-density). Our best understanding suggests that inflation occurred (where space was expanding faster than the speed of light, although light/energy/matter/information was still constrained by c).
In my previous post I was just pointing out that the expected size for the Schwarzschild radius is very large. However that is based on a naive application of the usual rules-of-thumb. The big bang, if you will, is extreme enough that it requires a more careful treatment. Moreover, our best data right now suggests that the universe is roughly flat and infinite (and thus with infinite or at least extremely large mass), meaning that there is probably no meaningful way to apply the "Schwarzschild radius" concept to it.
Disclaimer: I'm not a cosmologist. Hopefully I didn't make a mistake.
Uptime contests are fun and all. But I'm always suspicious of systems with very long uptimes: they probably haven't had a kernel update since that last reboot... meaning that they are a highly vulnerable box.
My Ubuntu machine has uptimes that are about 30-90 days, which is entirely based on new kernel releases. I've never had an unintended reboot (e.g. from a freeze or crash).
(Yes, there are methods of updating the kernel without rebooting... but most people with massive uptimes seem to achieve it not by using these tricks but rather by not touching the box.)
What I want to know, is how did the universe expand beyond its own swartzchild radius?
It didn't.
To compute the Schwarzschild radius of the universe, we need to know its mass. Recent measurements suggest that the universe is flat, and so may have infinite mass. However at a minimum we can count up the mass within the observable universe. The observable stars in the universe have a mass of ~2*10^52, but they are overwhelmed by dark matter, which brings the total mass within our observation volume to ~4*10^53 kg. So the Schwarzschild radius for the universe is:
r = (2*G*m)/(c^2) = 2*(6.7E-11 m^3kg^-1s^-2)*(4E53 kg)/(3E8 m/s)^2 = 6E26 m = 60 billion light-years.
Since the observable universe is ~46 billion light-years in radius, this means that the Schwarzschild radius of the universe is bigger than what we consider to be "the universe." In other words, we are well within the Schwarzschild radius, leading some people to describe the universe itself as a massive black hole that we are actually inside of.
The universe probably has a mass larger than what we can observe, making the radius even larger than the above estimate. If the universe truly has infinite mass, then the radius is infinite. In other words, the universe may not have a Schwarzschild radius at all.
This is also a decent description.
The question is phrased in a sort of black/white manner: either you fight tooth-and-nail to maintain maximum privacy, or you give up and sign up for every crazy privacy-eroding service.
The obvious answer is "all things in moderation." I consider myself privacy-conscious. I don't run Windows. I do use Facebook and Gmail. However I use them with privacy in mind. So my Facebook profile has very little information, has privacy options set quite high, and I only accept friend invites from people that I reasonably trust. (So many people seem to get sucked into the "I need my friend count to be higher" game--which invariable means accepting invites from strangers.)
My strategy works, more or less. There are times when friends reveal information about me online I would rather they didn't (e.g. tagging me in photos on Facebook). But you can't completely prevent these kinds of things. In the same way that friends can give out your phone number or gossip about you in real-life, there will be some privacy loss online. The goal should be to keep things private without it becoming a burden to do so.
It sounds like you're taking the privacy thing to far--to the point that it's harder for you to socialize and enjoy life. So loosen your rules a little bit. Remember that every company (the power company, the cable company, your bank, etc.) has tons of privacy-eroding data on you. Online companies will also get some privacy-eroding data. But as long as you keep it within reasonable bounds, then it won't cause a problem.
Remember, privacy isn't really something that has to be maintained for its own sake. Privacy is a means for you to enjoy your life free from bother, and to prevent people harming/taking advantage of you. Calibrate accordingly.
A small loss of privacy is okay if it achieves the greater objective of making you happy.
a TV tuner that can be plugged into a PC -- will let Net viewers of the likes of Hulu.com and ABC.com skip commercials in the nascent medium of online video.
What? Why do I need a TiVo TV Tuner box to watch online videos? Stripping commercials from online streaming video sounds like a software task. And saving the streaming video so that you can jump past the commercials doesn't require any special TiVo magic (whether hardware, subscription, or software). Will we see software and utilities that allow you to skip ads on online video? Probably. But what does this have to do with a TV-Tuner card for your PC?
The article also asks some nonsensical questions:
Does the arrival of the TiVo PC set-top box represent the final convergence of television and Internet video?
No. TV-Tuner cards and online video have existed for awhile. I don't see how a TiVo box changes anything. Yes, it might make "TV on your computer" more accessible to the masses... but that isn't a "final convergence" of anything, really. Sure, the lines are blurring between TV and Internet. And TiVo is part of that inevitable change. But this box isn't a revolution.
What will be the impact of TiVo's new device on the online video economy?
None. It's a TV-Tuner card, isn't it? (People watch Hulu because they don't want to pay for the equivalent cable channels.)
Will TiVo be remembered as the company that helped slaughter the advertising golden goose that has enriched the broadcasting industry for the last 50 years?
Doubtful. TiVo hasn't demolished TV ads yet. Strangely, PVRs in general haven't either. And AdBlock hasn't demolished web ads. These are all part of the arms race which keep ads sufficiently non-annoying that a sizeable fraction of the population doesn't bother avoiding. There will always be people who avoid them. But most people don't bother.
Add to this the fact that part of TiVo's strategy is to deliver ads to customers somehow... I hardly think that this new box changes much for the ad industry.
On his webpage, Ghosh has this to say:
(Emphasis added.)
So this seems like still very early work (just an idea, really)... and it appears that the intention is to build new kinds of switches (e.g. transistors) that exploit the fact that switching is inherently non-equilibrium, and extract some of the energy that is dissipated during these switching events.
Yeah the article is unclear. Here's my best shot at clarification:
A "Brownian ratchet" is a thought-device about extracting energy from the random Brownian motion of a hot gas. Similar to Maxwell's Demon, it can't work in a system at equilibrium. Without a temperature gradient, there is no way to extract useful work. The ratchet will be undergoing random motion equal in magnitude to the energy we hope to extract, so we can't actually extract anything.
However, if we're not at equilibrium, the rules are different. These researchers are talking about "non-equilibrium Brownian ratchets", which you could also call a "Brownian motor". In an non-equilibrium situation, you will have a gradient of heat or chemical potential that could, in principle, be converted into useful work.
So my guess is the researchers are trying to do something like:
1. Build devices that exploit the temperature gradient that exists in the device. So a bunch of nano-sized ratchets that convert the heat gradient on the outside of the chip (relative to the cool air) to recharge a capacitor or something.
2. Build switching elements (e.g. transistors) that directly store the excess switching energy in some way. That is, build switching elements that both do computational switching, but immediately utilize the resulting temperature gradient of the dissipated heat.
In either case, all they are suggesting is to take advantage of the heat gradients that inherently occur when you have imperfect switching elements dissipating heat. It's not really that novel, conceptually... although if they actually have a specific way to do this in mind, then that could be quite interesting.