Prudence...
If you notice someone got into your house, I would think changing the locks would be among the first of your moves, too. For once, an article of security done (AFAIK) properly.
267 publicly visible posts • joined 30 May 2008
"Asymmetric bars" was an improper name. They are indeed symmetric along two axes--along a line that intersects both bars halfway and along a parallel line exactly halfway between them. Thus the term "uneven"--describing that each bar is a different height and thus not "even" to each other.
People become less flexible as they age--simple human biology. Younger ladies thus possess more natural flexibility and are better capable of bending themselves as needed for gymnastics events (which on the women's side usually place an emphasis on flexibility and agility--for the men, there's an added focus on strength).
Smaller dies reduce the manufacturing costs by increasing the economies of scale (more chips per silicon wafer). Furthermore, they increase reliability (to a point) by reducing the amount of material the current needs to pass through (reducing heat generated through resistance). This allows manufacturers to either put more into the chip at no additional cost or lower its TDP. This is why 45nm Core 2's are so popular. Intel took the first route--they have more cache than their 65nm counterparts yet they still have the same TDP. Also, apart from the high-demand CPUs, they are priced at only slight premiums to their 65nm counterparts, making them very tempting steps up.
So they jazzed up the looks a little bit. Does it play the same? I would imagine so. It simply may well be that the old posh look of Clue was in fact detrimental to its sales because today's people couldn't fit in (Who reads Agatha Christie anymore? We're more into watching CSI and the like--or whatever it watched across the pond.). If giving the game a face lift brings in more players, then bully to them.
As I recall, image spam used the same trick. All that happened was that OCR technology improved and became better able to pick out the juicy bits. Plus there's always the simple solution of a human agent visiting the site and typing the prices seen on screen into another program.
As the problem seems to stem from transactions performed abroad, where technology may not be available, the problem becomes making a form of authentication that is BOTH secure AND portable--such that it can be used just about ANYWHERE. Signatures are an obvious 'no'--they've been known to be forged. And it's become clear Chip-And-PIN can't apply--such devices can't be counted on to exist abroad. So, what can you try?
The problem is that this frequently-executed gesture is used in myriad places besides the checkout line, and as long as a person possesses a copy of your handiwork, this person can train his/her hand to replicate it. Combine this with a cloned copy of your card (or even the card itself--with a handy reference signature)...and it's open season. As I recall, this had been the procedure en vogue in the past and is still used in the greater crime of full identity theft.
There's also the matter of handicaps. Not everyone has a steady hand at the checkout line. If someone possesses a nervous tic in the writing arm--or worse, a palsy--one's signature cannot be counted on to be consistent.
First, in order to establish a remote connection, there must be a means to access the Internet. If the location you're in happens to lack the means (no landlines, no WiFi, etc.), then you're SOL. But you may still need that data at that moment.
And as for standardising security, I give you one very important question: Who's going to PAY for all this (expletive)? I don't care if it's a matter of life or death, but we can't put in what we don't have. Where's the BUDGET for it?
In order for the computer to be able to read the graphics memory, it must be MAPPED into an address range. Now, in the 32-bit world, there are only 4 billion possible addresses you can use, but both the system RAM and the graphics RAM are getting so big they're competing for limited addressing space. Two CrossFired 4870X2's require the mapping of 4GB of memory. As it turns out, 32-bit addressing only has 4GB available. If you're going to be using one of these things (and definitely if you plan to CrossFire), you must use a 64-bit OS, whose address space isn't expected to be completely occupied anytime soon.
Not necessarily. As long as there is SOME competition, the situation is far from unbearable. In fact, some may welcome it since it helps narrow options while still keeping the competitive spirit going. The performance graphics market has been a two-horse race for years (only recently has Intel declared intentions to enter it) with barely a complaint.
These young men, being in their mid teens, may indeed be smart enough to know how to elude their parents. They may have gotten their copy of GTA4 through lesser-known means, such as flashing a fake ID, enlisting the aid of a legal-age friend or relative (say, a friend in college who would want the game anyway), or stealing it.
So remember, before blaming the parents, ask yourself, "What if the kid is smarter than the parents?" Because if this is true, then all bets are off when it comes to control.
However the computer system itself works, the paper printout is the key to the final result. Be it on a full or partial sheet of paper, the printout should contain the following:
1) The machine-readable portion of the ballot, encoded cleartext in a public-domain format such as PDF417 or DataMatrix. This machine-readable portion should ONLY contain the codes for the votes cast on the ballot. This would allow for a separate machine to read the code if necessary and allow a person to verify the machine part against the following human-readable part.
2) The human-readable portion of the ballot, printed in clear legible text literate people with reasonable vision can easily read. For each line, there is both the code and the name(s)/decision(s) of the ballot/proposition in question.
The biggest problem Linux faces in the end-user world is that there are tons of compelling reasons to stick with Windows (the giant product library), but the most important one is not a product but a service. For Linux, the only way you'll get an average end-user to accept your product is to make it "turnkey" simple: turnkey as in you turn it on and it goes with no intervention on the user's part--and no matter what the user's machine presents to the OS. But the Linux world's base philosophy of customizability ultimately clashes with this simplicity--many kinds of customization leads to information overload for most people, causing them to balk.
You got 1 and 2 the other way around. Active radars transmit pulses and listen for returns. Passive radars simply listen for ambient radiations. The military uses passive radars, sonars, etc. because they don't give away your position.
As for #4, ins't nuclear power still more a NIMBY or even NIABY issue than wind turbines?
For the China scenario, this doesn't work. All that's needed is an assurance that the exit node is outside China. Once that happens, it's outside their jurisdiction. Or are you saying that various Chinese sleepers (located outside China) are really going to pose as Tor exit nodes, sniff the data, and be able to pass it along back to the homeland without breaking some kind of data protection law in the country in which they're currently stationed (if not charged with out-and-out espionage)?
I imagine most of the Work watts would be put into the thruster assembly, transferring the energy into the "reaction mass" as a newtonian force and "pushing" it out, providing a resultant desired reaction of forward acceleration.
IOW, those watts wouldn't become heat but rather kinetic energy.
According to what I've read, the i860 suffered because it tried to put too much of the pipelining and scheduling work on the programmer and compiler (similar to the Itanium). It proved to be too difficult to program efficiently.
Whereas with the Fusion, they're integrating two well-established chips: an x64-based CPU with a modern ATI GPU, both of which have a long real-world history.
I was just thinking of that angle. Of course, the built-in GPU chipset needs to be pretty decent to start with. I can tell you now that if it's pretty basic (say like an nVidia 8400GS), it's not really worth the benefit compared to a modern multicore CPU. OTOH, if it's closer to, say, the 8800GT (which is capable of doing games at a pretty solid clip) then you can do some serious stuff with it, with or without the added boost of a new GPU.
Not earth-shattering to say the least, but right in line with electronic noses to detect scents and perhaps dangerous substances, now there is an electronic tongue to apply the same principles to the sense of taste. Although its initial applications are commercial (to authenticate wines), it's not too hard to see its application extended to perhaps taste for toxic or contaminated drinks and so on.
Being a financial analyst, he had to be able to determine the credit-worthiness of customers. That means having access to individual records. This is basically an inside job--always the toughest thefts to control. Somewhere along the line, SOMEONE has to have access to the data. And at some point, according to statistics, THAT someone is going to be a double agent.
Just for the record, the kiloyard is an actually-used-in-real-life measurement. US Navy submarines use this to measure distances while underwater (because their other units are imperial and mile is probably too long for measurement of relatively-slow-moving vessels like submerged subs).
So you're basically saying that, even at the most basic level, the sheer scope of such projects prevents an exhaustive look at a program's security. And I already know the necessary robustness of most programs prevents the KISS principle, too.
It just seems frustrating that you keep hearing about these exploits, especially those old-school buffer overflow exploits. We've been had more than twice, yet they keep on coming.
Thing is, wouldn't you think they people would code their software to such a point that security difficulties never show up as an issue? If not, why has "getting it right the first time" become impossible? What about platforms where a fixed non-updatable installation is necessary?
Believe me. Virtual keyboards are the pits. I have to put up with them on my Palm T|X. The big problem with virtual keyboards is that most of us are used to typing on the horizontal and reading against the vertical and receiving tactile feedback to acknowledge our actions and correct those near-misses. With virtual keyboards, you get no tactile feedback and (unless you use a projected keyboard) have to type on the same plane as the screen--very slow and uncomfortable.
I think one of the reasons screen sizes aren't too big is power consumption. Driving a larger screen inevitably requires more power to alter the LCD as well as more power on the graphics chip to render the higher resolution--especially since LCDs are not a persistent-display technology. This may change in future with further development of "e-ink" persistent displays (which are also being developed to be deformable--maybe not foldable but would you take rollable?).
Actually, for it to be properly classified as a byte, there has to be 2^8 (256) possible combinations. I don't think they've made the technology that precise as yet. The closest term is "nibble" (4 bits--16 possibilities).
As for why use DVDs, I'm pretty sure this is just an experiment--a way to see if it can be done using well-available technology. Adopting the technology in blue-laser discs will probably be the next step--the smaller pits will require higher levels of precision in making those beveled pits.
Well, rehabilitation doesn't seem to work, either. They're either already solidly against the system so that they'll just "bide their time" and go right back to business as soon as it's over or they're either so dense or so empty that anything you try to teach them either bounces off (the former) or passes right through (the latter).
And could the proxy server be able to do all this while acting just like a real-life Internet user without breaking something (by using the wrong user agent, for example)...or being detected by the malware authors and being tricked with a false negative or--worse--something that compromises or bypasses the proxy server itself?
So you're against immunization? That's a proactive approach, too. It's also the only approach that's effective most of the time. The reactive approach is like trying to shut the door after the horse already got out. The screening (proxy) approach won't protect against a zero-day vulnerability since the proxy can't distinguish it. So if loading the page is too late, but you have to load the page to determine its safety (think of it like a steel-encased bomb with a trapped lid and the only way to get to the innards of the bomb to defuse it is to lift the lid...), how do you go about it?
So how else do you take a proactive approach to stopping web exploits, particularly zero-day vulnerabilities, short of shunning the Web? Since the exploiters are being shrewd enough to only serve the exploit once per IP and only to requests from genuine web browsers not protected by proxies, it's like a bunch of highway robbers geared to only ambush lone travelers. So how else can you spring the ambush except by acting like a lone traveler?
What AVG seems to believe is that it needs to look through the search results proactively, before the web browser has even a sliver of a chance to get it into memory...because by then it could be too late. The proxy approach, for example, wouldn't work if the zero-day stuff happens to come before detectable stuff. The critical stuff would've been let through by the time AVG realizes there's a problem. And blacklists don't work anymore because of the increase of drive-by downloads that are infiltrating perfectly legitimate sites--they're becoming like AV signatures.
Essentially, AVG is saying the user clicking the actual link is equivalent to opening Pandora's Box--too late to do anything about it.
We could be facing a serious and hard-to-solve conflict of interests. Both sides have valid points (AVG's technique skews the statistics, but it's also probably one of the first techniques that prevents opening Pandora's Box).
The intense power of a NightSun is intended to light up a large area--it's a flashlight on steroids. Since the helicopter is up in the air, having that large spotlight makes for an effective night canvass. Down at ground level, it's supposed to be no worse than a car's headlights or a street lamp. And since it's going *downward*, drivers usually are at little risk of blindness because they're shaded by their vehicle roofs.
There are two potential snags with a two-century disc.
1. Readers for the disc may not be available in two hundred years, making them impossible to read (for a more recent example, try to read a 3.5" floppy disc in an age when floppy drives are few and far between).
2. The data contained within the discs may be of a format that eventually gets lost to time and will (when the need arises in the far future) look only like so much gobbledygook.
Not necessarily. The non-ASCII DNS lookups would more than likely simply match up to their extensive ASCII counterparts for the time being. As noted, a comprehensive plan of action concerning non-ASCII domains in general will probably take a more gradual approach--perhaps long enough to formulate a proper solution.
I think some people are stretching things a bit here. Sure, a corporation isn't pursuing Vista, and perhaps that is a good thing if everything is running fine with no compelling need to upgrade. Seeing as how the parts of Intel that would be dealing with modeling and so on would probably be on heavier-duty workstations using more professional cards and driver sets (and perhaps even different operating systems), the old saying applies:
"If it ain't broke, don't fix it."