"Granted, the plugin compatibility issue is a problem that needs be be resolved before it drives more people away from Firefox."
that horse is going to leave the barn very soon if it hasn't already. I'm on FF 3.6 and have been told anything less than FF 4.x is unacceptable (due to web app requirements) so I *have* to move and I don't think I'm going to bother with FF5 -- it'll be replaced by FF6 (with no functioning plugins) before I can even get it installed. If I were on Windows I'd just go with IE or on OS X with Safari -- but I'm on linux and the other major choice is konqueror. Which, although more stable than it was a few years ago, still sends the CPU into an epileptic seizure when browsing.
Figuring out which broken browser to attempt to use is not something I'm looking forward to (and, no, I don't care for Opera, never have).
"It doesn't come pre-installed on any of the popular desktop operating systems (maybe it does come standard on some devices?)."
It comes pre-installed on the Wii, but is helpfully rebranded so most people aren't aware that it is Opera. I think it comes pre-installed on some phones, but I don't remember for sure.
I see what you're saying, but the charges the RIAA has pursued are not "possession without financial remuneration" they are for "redistributing without a license". I think the question is valid and I'm curious what the answer is.
Interesting, but reading the cdrfaq.org page it indicates that it is not actually inevitable at all and can be corrected in either software or the drive doing the extraction. What the jitter does explain is how to rips can differ (even though they *can* be identical). Another explanation for rips differing is errors during reading, and here I'm talking about a scratched disk. Scratches are an analog feature, as is the spinning of the drive, etc., and can result in minute variation in the digital data. Just wanting to point out that jitter isn't the only explanation.
He didn't assume "that the irradiance is absorbed linearly as a black body by the earth", you did. He pointed out very clearly that variations in solar irradiance simply don't vary as much as people seem to believe.
In fact, when you mention "atmospheric composition, oceanic current flow, heat from the core, drag from the moon and sun" all you manage is to explicitly mention other factors that have greater variation than solar irradiance does.
And yet you were modded informative? GP was insightful in pointing out people don't understand the variability in the sun's output is negligible compared to other factors, demonstrates that the temperature changes experienced *must* be caused by other variable factors butt is modded 'interesting'. Sheesh.
What I'm really griping about is you added nothing to the discussion. You say "People are also bad at understanding complex effects, as your post shows" but all you really demonstrate is that your reading comprehension is fairly limited. And then get modded informative. Sheesh
speaking as a parent, you anonymous coward, disgust me. Just because you can have children doesn't mean you should. Or are you advocating teenage pregnancy as well? Own up and take responsibility for your offspring.
That great bastion of knowledge known as wikibooks says "Federal registration is not mandatory to receive trademark protection, but a mark owner seeking to enforce trademarks in US federal court needs either a federal registration or a violation of the Lanham Act's specific sections on false advertising or unfair competition." [http://en.wikibooks.org/wiki/US_Trademark_Law]
Sounds like iCloud Communication will need to specify how Apple is committing false advertising or presenting unfair competition. They may be angling for a pay off, but it may just result in lawyer billing driving them out of business.
it is 100% trivial to pull out the email from an exchange server into a PST and how we do it when complying with similar requests. It is, in fact, necessary to retrieve them in some fashion before printing. That digital intermediate could then trivially have been distributed, limited bandwidth or not.
"If it were true that people like yourself really did feel Palin is a total idiot, the best solution is to ignore her."
That reads like you think you are correcting GP, when in fact there is nothing you say that contradicts what GP said. GP observed that people feel compelled to correct her erroneous statements, not just because they are wrong, but because they are said with such conviction. Of course its better to ignore her, it just isn't easy. Kind of like watching a train wreck, terrible but you can't look away.
I *really* wish I had mod points. Of course, I've already posted as well. Still...
And this is what really bugs me. I like my bank -- it is local, the tellers no me, they treat us well, we have a good mortgage from them, etc., but their online security is as sucky as all the others. People who say you can switch banks don't have a clue. First, I can't just move the mortgage account. And trying to re-finance it to pay it off and owe someone else even more money doesn't make any sense. Second, while it is important, online banking security is only one aspect. When the bad investment mortgage fiasco hit the fan I wasn't affected -- my bank wanted me as a customer and hadn't sold the mortgage to some shady speculator. Nor did they resell it to one of the mortgage collection firms that treats customers like criminals. A guy I know missed a mortgage payment because, even though the company which it had been sold to acknowledged receipt of payment they "lost it". What? Why is he liable for them losing it? After it was past due, they "found it" behind a filing cabinet. Yeah, right.
So while the "security" of online banking is highly questionable options are limited and realistic options are even more limited. I'm not particularly worried about the security of my home systems, but that isn't the only point on the communication loop. And something like OTPs can be done quite reliably for significantly increased security. Why don't they? Although I don't know this for a fact, I'd hazard because implementing actual security has a cost and is not a regulatory requirement. The cost isn't going to go away -- so it will only improve when it becomes a regulatory requirement. Which would, IMO, require a major meltdown of personal banking to the extent that people emptied their accounts and stuffed the money into mattresses because it was safer.
This court case isn't about establishing the banks liability regardless of GGPs wishful thinking, it is about establishing their *lack* of liability. That any pretense at security counts as the real thing. That the industry standard of leaving the doors wide open is just peachy. So any loss experienced by a consumer is, well, an act of god or the consumer's fault.
What you describe sounds like two channel communications.
"The first one, the response is either on a paper sheet you have (which you can then move to a computer file or whatever if you want to spend some time typing it in) or is sent to your cellphone along with the amount (so that no transactions can sneak in without it being shown in the same text)."
So, did you get that paper sheet from your web browser? No? Then the method by which you obtained the paper sheet is the second channel.
"The other one, the response is something generated on an external device - looks like a little calculator - after entering the challenge."
I'm fairly certain that the external device didn't come through your web browser. The information contained therein is necessary for the authentication and as that information was not delivered over your network it was delivered via a second channel.
US banks like to do stupid, meaningless gimmicks, such as having you select an icon that will be served on the page so you know it wasn't someone else (apparently they still don't understand the meaning of man-in-the-middle). That icon is being served over the same channel as the rest of the web page. But lets go to a different scheme: using SMS to obtain a pin. That sounds good in terms of providing a second channel (using the cellular network for SMS rather than your home broadband connection), but it only really works if you have to go to the bank to provide the phone number. If the bank allows you to configure what phone number to send an SMS to online it doesn't provide meaningful protection because it can be co-opted.
To clarify: Mycin was not a general purpose diagnosis engine, but built for a specific purpose: produce probabilistic recommendations for anti-bacterial prescriptions for bacterial infections. It accomplished this not through some miracle AI, but (to quote wikipedia) a "simple inference engine" and pre-fed rules. There was no analysis given of the accuracy of the test in which it reportedly outperformed the human experts.
An immediate question I have has to do with skewing of results (in other words, how was it determined what the correct answer was). Another question has to with how Mycin's answers were compared to humans: it gave a laundry list with carefully calculated probabilities and meticulously tracked rationale. It did not provide a single diagnosis (which is what a doctor does when he prescribes for you) -- were the human doctors marked down for failing to provide exact probabilities? For tracking exactly what observations led to the diagnosis?
Another thing I noticed was it arrived at diagnosis after providing yes/no answers to a "long list" of questions. A "long list", eh? How long? How long would a patient need to mark off yes/no answers to get a diagnosis? How well is a patient going to score on his submission? And, if he mis-answers questions resulting in a misdiagnosis is it the patient's fault? Might something more than just an expert system detect impatience and frustration with answering a long string of yes/no questions?
Another issue is that in the real world you don't know before hand whether or not something is a bacterial or viral infection. How useful is an anti-bacterial medication prescriber in the real world?
I'm not knocking expert systems, but Mycin does not seem to be a particularly convincing one. Its underlying model was shown to be flawed and developer claims to be unfounded (according to that bastion of knowledge, wikipedia). An expert system like Mycin is incapable of making judgement calls, it just reports probabilities based on assumed valid inputs.
Besides which, liability is a *very* important issue. As a potential patient I don't want to be held liable for a misdiagnosis because I didn't properly understand a question. As a programmer I don't want to be held liable for a misdiagnosis because the patient did not have the most prevalent cause producing the observed symptoms. As a medical institution I don't want to be liable for a misdiagnosis due to poor programming, inaccurate models, limited models, incomplete or inaccurate patient input to a computer system. As an insurer I don't want to underwrite a system where every misdiagnosis is laid at my foot because no one else will take it.
I'm not sure you get the size of the numbers involved. Unsurprising, but you might want to put things into perspective. Its convenient to write a large number in a notation like 2^64 (being the number of tokens sold before expecting a dupe according to GP), but another to even remotely consider how large it is.
1. Realistically, dupes don't matter. Really, they don't. Arguing about a potential collision is about as useful as saying someone could randomly concoct the sequence. No reason to ever record seeds
2. If a hash mechanism were employed to check for dupes you would *want* it to have fewer bits. Otherwise you aren't really hashing, you are just substituting. But even if the numbers are too big to figure out how freaking utterly huge they are allow me to point out that the company would by *very* happy to sell enough to be chucking some for being "dupes" even if the odds of being a dupe were actually only 1 in 2^32
There currently is no reason I can see to buy a MacBook -- if you want that size then for basically the same price you get a MacBook Pro, except that it has better hardware (thunderbolt and firewire ports, for example). With the introduction of a 13" MacBook Pro that is priced like a MacBook... I'm expecting something to happen to the MacBook line.
The fact that someone is willing to exchange their bit coins for real money doesn't mean you will be able to turn around and exchange those bit coins for the same (or nearly the same) quantity of real money. There is no meaningful currency exchange for bit coins.
Yes and no. It depends on the malware and what vulnerabilities it is exercising. In *principle* you get a warning from Win7 about executing a file downloaded from the Internet. In practice, not necessarily. For example, one of the Java "apps" I reversed did a simple "fetch and execute". Most users aren't going to be warned about a java app embedded in the page, nor are they going to be warned about it fetching a remote file and executing it. The fetched file wasn't even given an executable extension, it was directly loaded/executed by function from Java.
Not that that same can't be achieved on OSX, but I haven't seen any examples of it.
a true low level format hasn't been possible on drives for a long time, and that is a good thing. Most people think "writing to every sector" is a low level format, but that isn't what it used to mean. A low level format actually dealt with how data was encoded on the drive and was peculiar to the drive. This is easier to describe with MFM/RLL drives [http://en.wikipedia.org/wiki/Run-length_limited] than later models.
Modern drives use more complex encodings that play into better performance/storage efficiency/what-have-you. The ability to ruin a drive by an improper low level format may have been cool, but it also didn't serve a useful purpose for the user. On a modern drive the internal encoding mechanism is not exposed through the interface, presenting instead a black box that allows data to be stored and retrieved through a set of standard commands. Bad block remapping being completely internalized is an advantage.
Seriously, who misses needing to record the bad blocks of a drive so they could be mapped pre-emptively when reformatting the drive? The labels used to have an area for recording the bad block list so it would stay with the drive. I'm much happier now for the drive to handle this, and for those that care you can access the statistics on it via SMART. Sure, you don't know *which* blocks are bad but you know how many. While it might be nice to identify a run of bad blocks versus a scattering of them in practice what matters most is the how many and how fast the count is rising. I have a hard drive that has 16 bad blocks remapped since I first checked. The count hasn't increased in years so it was probably a manufacturing defect.
If you need to wipe all data from a drive any recent one supports a secure wipe command that, once initiated, will run to completion regardless of power interruptions or being switched to different hardware and has the capability of wiping areas not accessible through the interface. Sure, you can't apply a fresh low level format, but you can make very sure the drive gets wiped.
In short, low level formats aren't relevant with today's drives. Today's drives also are faster, larger and more robust with better error handling and reporting than before. They also have new capabilities, such as secure wipe, that simply weren't possible on 30-year old hard drives.
not just that, but the sophos article glosses over the fact that you still get
1. an operating system warning about executing a file downloaded from the internet (complete with reference to where it was downloaded from). They mention it in the text, but omit it in their "slideshow" showing the steps to getting infected.
2. an osx installer gui which means it can be canceled
What this is *not* is a hidden and silent install like what is going on with Windows.
you are seriously confused, and don't know how an Amiga worked. The Workbench was a full blown GUI, not a mostly-there gui. Having a shell window without a desktop is *not* Workbench. The A1000 (and later the A3000) differed in that they had some of the information on a kickstart floppy rather than in ROM which *did* require a floppy to boot (I think the A3000 was able to read its kickstart from a hard drive, but to be honest I don't remember).
But, hey, I'm sure I can't convince you. One thing the Amiga always had going for it were users who were blind to anything they didn't want to believe. But if you care to read you might try http://en.wikipedia.org/wiki/AmigaOS#Kickstart The wikipedia entry says "full windowing environment" which is a bit of an exaggeration (it allowed close, move, resize and front back window operations and redrew window contents). Then there's http://www.amigahistory.co.uk/earlystart.html which shows the initial screen. Sure, you have to hold down both mouse buttons on boot to access it or you get the "insert disk" message -- but you already knew that, right?
Kickstart as a "BIOS-equivalent"? I don't know of any BIOS that support windowing, where you have terminal windows and common commands available to you, including the ability to launch new terminal windows. On the other hand I don't remember configuring drives or boot order from KickStart. On an Amiga, KickStart may not have been a full system, but it was not like a BIOS either.
Ah, you've bought the company line. Very good for you. I can respect that you "don't support wealth re-distribution" -- but that is the underlying premise of an economy. Perhaps what you meant to say was that you "support the tendency of unregulated wealth or power to concentrate." Which is rather like saying you don't support elevators because they work against gravity.
But the concentration is tendency only, not an absolute rule. And from time to time there are corrections. The ruling class in medieval europe forced (not necessarily consciously) Mediterranean agriculture on regions (now France, Germany and so on) where it was ineffective and inefficient, extremely intensive in man power for exceedingly low return. Such low returns that some historians discount the contemporary records for crop returns, refusing to accept the evidence. But a middle class still managed to emerge, not respected by the ruling class and despised by the impoverished masses, but still able to siphon off wealth and effect some of this redistribution you say you don't support.
The medieval lords have been replaced and now we have a small fraction of the population controlling the vast majority of the wealth. US production levels don't require near 100% employment to maintain, but the mercantile class has convinced the population at large that "work-for-pay" is not just one method of economic regulation, but the only right and proper method -- at least for the masses.
The balance will readjust at some point, but I do hope it is able to do so more gradually and peacefully than, say, the French Revolution. And there were impoverished people in revolutionary France who supported the authority, wealth and power of the aristocracy. It never fails that you can find people who support what objectively is counter to their better interests.
I don't think Apple would do it, but I sure wouldn't say never. At the least, developers will need a computer to actually develop on and Apple wants them to use one of/their/ computers to do so and pretty much by definition a development system can't be locked down against unapproved software. So I doubt they will entirely do away with "open" systems. On the other hand, they might very well market a "home computer" that was effectively locked down.
GP assumed that the cap mechanism would not count bytes because that is apparently supposed to be difficult or something. I'm not sure.
The traffic shaper we use where I work would be a bit of overkill for the job, but it could be used to enforce a capping mechanism. The hardware is, shocking I know, primarily sold to ISPs. And counting the number of bytes passing on the wire is trivial -- when low latency hardware can make shaping decisions based on source and destination IP address, ports used and the protocol employed (e.g., http regardless of the port it is running on) -- little things like tracking the number of bytes transferred are insignificant.
But even if I was going on the cheap and not using a real product with a real company behind it (important for PHBs to point the finger at in case of problems) then bandwidth capping can be accomplished at a byte-count level by processing flow data -- which any decent router can provide.
Seriously, an ISP is going to have some sort of traffic shaping device and it can count packets without increasing system load. Where a discrepancy is more likely to occur is between binary and decimal reckoning. Because I'd be surprised if an ISP didn't use 1,000,000,000 bytes to one gigabyte when specifying capping levels, but file systems still tend to report using powers of 1024 -- so when you download a 4.2 GB ISO it is likely to count as 4.5 GB of the stated quota.
Slashdot Mirror
"Granted, the plugin compatibility issue is a problem that needs be be resolved before it drives more people away from Firefox."
that horse is going to leave the barn very soon if it hasn't already. I'm on FF 3.6 and have been told anything less than FF 4.x is unacceptable (due to web app requirements) so I *have* to move and I don't think I'm going to bother with FF5 -- it'll be replaced by FF6 (with no functioning plugins) before I can even get it installed. If I were on Windows I'd just go with IE or on OS X with Safari -- but I'm on linux and the other major choice is konqueror. Which, although more stable than it was a few years ago, still sends the CPU into an epileptic seizure when browsing.
Figuring out which broken browser to attempt to use is not something I'm looking forward to (and, no, I don't care for Opera, never have).
"It doesn't come pre-installed on any of the popular desktop operating systems (maybe it does come standard on some devices?)."
It comes pre-installed on the Wii, but is helpfully rebranded so most people aren't aware that it is Opera. I think it comes pre-installed on some phones, but I don't remember for sure.
I see what you're saying, but the charges the RIAA has pursued are not "possession without financial remuneration" they are for "redistributing without a license". I think the question is valid and I'm curious what the answer is.
Interesting, but reading the cdrfaq.org page it indicates that it is not actually inevitable at all and can be corrected in either software or the drive doing the extraction. What the jitter does explain is how to rips can differ (even though they *can* be identical). Another explanation for rips differing is errors during reading, and here I'm talking about a scratched disk. Scratches are an analog feature, as is the spinning of the drive, etc., and can result in minute variation in the digital data. Just wanting to point out that jitter isn't the only explanation.
He didn't assume "that the irradiance is absorbed linearly as a black body by the earth", you did. He pointed out very clearly that variations in solar irradiance simply don't vary as much as people seem to believe.
In fact, when you mention "atmospheric composition, oceanic current flow, heat from the core, drag from the moon and sun" all you manage is to explicitly mention other factors that have greater variation than solar irradiance does.
And yet you were modded informative? GP was insightful in pointing out people don't understand the variability in the sun's output is negligible compared to other factors, demonstrates that the temperature changes experienced *must* be caused by other variable factors butt is modded 'interesting'. Sheesh.
What I'm really griping about is you added nothing to the discussion. You say "People are also bad at understanding complex effects, as your post shows" but all you really demonstrate is that your reading comprehension is fairly limited. And then get modded informative. Sheesh
speaking as a parent, you anonymous coward, disgust me. Just because you can have children doesn't mean you should. Or are you advocating teenage pregnancy as well? Own up and take responsibility for your offspring.
That great bastion of knowledge known as wikibooks says "Federal registration is not mandatory to receive trademark protection, but a mark owner seeking to enforce trademarks in US federal court needs either a federal registration or a violation of the Lanham Act's specific sections on false advertising or unfair competition." [http://en.wikibooks.org/wiki/US_Trademark_Law]
Sounds like iCloud Communication will need to specify how Apple is committing false advertising or presenting unfair competition. They may be angling for a pay off, but it may just result in lawyer billing driving them out of business.
it is 100% trivial to pull out the email from an exchange server into a PST and how we do it when complying with similar requests. It is, in fact, necessary to retrieve them in some fashion before printing. That digital intermediate could then trivially have been distributed, limited bandwidth or not.
"If it were true that people like yourself really did feel Palin is a total idiot, the best solution is to ignore her."
That reads like you think you are correcting GP, when in fact there is nothing you say that contradicts what GP said. GP observed that people feel compelled to correct her erroneous statements, not just because they are wrong, but because they are said with such conviction. Of course its better to ignore her, it just isn't easy. Kind of like watching a train wreck, terrible but you can't look away.
I *really* wish I had mod points. Of course, I've already posted as well. Still...
And this is what really bugs me. I like my bank -- it is local, the tellers no me, they treat us well, we have a good mortgage from them, etc., but their online security is as sucky as all the others. People who say you can switch banks don't have a clue. First, I can't just move the mortgage account. And trying to re-finance it to pay it off and owe someone else even more money doesn't make any sense. Second, while it is important, online banking security is only one aspect. When the bad investment mortgage fiasco hit the fan I wasn't affected -- my bank wanted me as a customer and hadn't sold the mortgage to some shady speculator. Nor did they resell it to one of the mortgage collection firms that treats customers like criminals. A guy I know missed a mortgage payment because, even though the company which it had been sold to acknowledged receipt of payment they "lost it". What? Why is he liable for them losing it? After it was past due, they "found it" behind a filing cabinet. Yeah, right.
So while the "security" of online banking is highly questionable options are limited and realistic options are even more limited. I'm not particularly worried about the security of my home systems, but that isn't the only point on the communication loop. And something like OTPs can be done quite reliably for significantly increased security. Why don't they? Although I don't know this for a fact, I'd hazard because implementing actual security has a cost and is not a regulatory requirement. The cost isn't going to go away -- so it will only improve when it becomes a regulatory requirement. Which would, IMO, require a major meltdown of personal banking to the extent that people emptied their accounts and stuffed the money into mattresses because it was safer.
This court case isn't about establishing the banks liability regardless of GGPs wishful thinking, it is about establishing their *lack* of liability. That any pretense at security counts as the real thing. That the industry standard of leaving the doors wide open is just peachy. So any loss experienced by a consumer is, well, an act of god or the consumer's fault.
What you describe sounds like two channel communications.
"The first one, the response is either on a paper sheet you have (which you can then move to a computer file or whatever if you want to spend some time typing it in) or is sent to your cellphone along with the amount (so that no transactions can sneak in without it being shown in the same text)."
So, did you get that paper sheet from your web browser? No? Then the method by which you obtained the paper sheet is the second channel.
"The other one, the response is something generated on an external device - looks like a little calculator - after entering the challenge."
I'm fairly certain that the external device didn't come through your web browser. The information contained therein is necessary for the authentication and as that information was not delivered over your network it was delivered via a second channel.
US banks like to do stupid, meaningless gimmicks, such as having you select an icon that will be served on the page so you know it wasn't someone else (apparently they still don't understand the meaning of man-in-the-middle). That icon is being served over the same channel as the rest of the web page. But lets go to a different scheme: using SMS to obtain a pin. That sounds good in terms of providing a second channel (using the cellular network for SMS rather than your home broadband connection), but it only really works if you have to go to the bank to provide the phone number. If the bank allows you to configure what phone number to send an SMS to online it doesn't provide meaningful protection because it can be co-opted.
To clarify: Mycin was not a general purpose diagnosis engine, but built for a specific purpose: produce probabilistic recommendations for anti-bacterial prescriptions for bacterial infections. It accomplished this not through some miracle AI, but (to quote wikipedia) a "simple inference engine" and pre-fed rules. There was no analysis given of the accuracy of the test in which it reportedly outperformed the human experts.
An immediate question I have has to do with skewing of results (in other words, how was it determined what the correct answer was). Another question has to with how Mycin's answers were compared to humans: it gave a laundry list with carefully calculated probabilities and meticulously tracked rationale. It did not provide a single diagnosis (which is what a doctor does when he prescribes for you) -- were the human doctors marked down for failing to provide exact probabilities? For tracking exactly what observations led to the diagnosis?
Another thing I noticed was it arrived at diagnosis after providing yes/no answers to a "long list" of questions. A "long list", eh? How long? How long would a patient need to mark off yes/no answers to get a diagnosis? How well is a patient going to score on his submission? And, if he mis-answers questions resulting in a misdiagnosis is it the patient's fault? Might something more than just an expert system detect impatience and frustration with answering a long string of yes/no questions?
Another issue is that in the real world you don't know before hand whether or not something is a bacterial or viral infection. How useful is an anti-bacterial medication prescriber in the real world?
I'm not knocking expert systems, but Mycin does not seem to be a particularly convincing one. Its underlying model was shown to be flawed and developer claims to be unfounded (according to that bastion of knowledge, wikipedia). An expert system like Mycin is incapable of making judgement calls, it just reports probabilities based on assumed valid inputs.
Besides which, liability is a *very* important issue. As a potential patient I don't want to be held liable for a misdiagnosis because I didn't properly understand a question. As a programmer I don't want to be held liable for a misdiagnosis because the patient did not have the most prevalent cause producing the observed symptoms. As a medical institution I don't want to be liable for a misdiagnosis due to poor programming, inaccurate models, limited models, incomplete or inaccurate patient input to a computer system. As an insurer I don't want to underwrite a system where every misdiagnosis is laid at my foot because no one else will take it.
Where is the expert who can be held liable?
This is probably the most important comment so far. I wish I had mod points.
I'm not sure you get the size of the numbers involved. Unsurprising, but you might want to put things into perspective. Its convenient to write a large number in a notation like 2^64 (being the number of tokens sold before expecting a dupe according to GP), but another to even remotely consider how large it is.
1. Realistically, dupes don't matter. Really, they don't. Arguing about a potential collision is about as useful as saying someone could randomly concoct the sequence. No reason to ever record seeds
2. If a hash mechanism were employed to check for dupes you would *want* it to have fewer bits. Otherwise you aren't really hashing, you are just substituting. But even if the numbers are too big to figure out how freaking utterly huge they are allow me to point out that the company would by *very* happy to sell enough to be chucking some for being "dupes" even if the odds of being a dupe were actually only 1 in 2^32
There currently is no reason I can see to buy a MacBook -- if you want that size then for basically the same price you get a MacBook Pro, except that it has better hardware (thunderbolt and firewire ports, for example). With the introduction of a 13" MacBook Pro that is priced like a MacBook... I'm expecting something to happen to the MacBook line.
so you are unaware of how easy it was to break Tor's anonymity? http://en.wikipedia.org/wiki/Tor_(anonymity_network)#Weaknesses
really? you can exchange bit coins for money?
no? I didn't think so...
The fact that someone is willing to exchange their bit coins for real money doesn't mean you will be able to turn around and exchange those bit coins for the same (or nearly the same) quantity of real money. There is no meaningful currency exchange for bit coins.
Yes and no. It depends on the malware and what vulnerabilities it is exercising. In *principle* you get a warning from Win7 about executing a file downloaded from the Internet. In practice, not necessarily. For example, one of the Java "apps" I reversed did a simple "fetch and execute". Most users aren't going to be warned about a java app embedded in the page, nor are they going to be warned about it fetching a remote file and executing it. The fetched file wasn't even given an executable extension, it was directly loaded/executed by function from Java.
Not that that same can't be achieved on OSX, but I haven't seen any examples of it.
a true low level format hasn't been possible on drives for a long time, and that is a good thing. Most people think "writing to every sector" is a low level format, but that isn't what it used to mean. A low level format actually dealt with how data was encoded on the drive and was peculiar to the drive. This is easier to describe with MFM/RLL drives [http://en.wikipedia.org/wiki/Run-length_limited] than later models.
Modern drives use more complex encodings that play into better performance/storage efficiency/what-have-you. The ability to ruin a drive by an improper low level format may have been cool, but it also didn't serve a useful purpose for the user. On a modern drive the internal encoding mechanism is not exposed through the interface, presenting instead a black box that allows data to be stored and retrieved through a set of standard commands. Bad block remapping being completely internalized is an advantage.
Seriously, who misses needing to record the bad blocks of a drive so they could be mapped pre-emptively when reformatting the drive? The labels used to have an area for recording the bad block list so it would stay with the drive. I'm much happier now for the drive to handle this, and for those that care you can access the statistics on it via SMART. Sure, you don't know *which* blocks are bad but you know how many. While it might be nice to identify a run of bad blocks versus a scattering of them in practice what matters most is the how many and how fast the count is rising. I have a hard drive that has 16 bad blocks remapped since I first checked. The count hasn't increased in years so it was probably a manufacturing defect.
If you need to wipe all data from a drive any recent one supports a secure wipe command that, once initiated, will run to completion regardless of power interruptions or being switched to different hardware and has the capability of wiping areas not accessible through the interface. Sure, you can't apply a fresh low level format, but you can make very sure the drive gets wiped.
In short, low level formats aren't relevant with today's drives. Today's drives also are faster, larger and more robust with better error handling and reporting than before. They also have new capabilities, such as secure wipe, that simply weren't possible on 30-year old hard drives.
not just that, but the sophos article glosses over the fact that you still get
1. an operating system warning about executing a file downloaded from the internet (complete with reference to where it was downloaded from). They mention it in the text, but omit it in their "slideshow" showing the steps to getting infected.
2. an osx installer gui which means it can be canceled
What this is *not* is a hidden and silent install like what is going on with Windows.
you are seriously confused, and don't know how an Amiga worked. The Workbench was a full blown GUI, not a mostly-there gui. Having a shell window without a desktop is *not* Workbench. The A1000 (and later the A3000) differed in that they had some of the information on a kickstart floppy rather than in ROM which *did* require a floppy to boot (I think the A3000 was able to read its kickstart from a hard drive, but to be honest I don't remember).
But, hey, I'm sure I can't convince you. One thing the Amiga always had going for it were users who were blind to anything they didn't want to believe. But if you care to read you might try http://en.wikipedia.org/wiki/AmigaOS#Kickstart The wikipedia entry says "full windowing environment" which is a bit of an exaggeration (it allowed close, move, resize and front back window operations and redrew window contents). Then there's http://www.amigahistory.co.uk/earlystart.html which shows the initial screen. Sure, you have to hold down both mouse buttons on boot to access it or you get the "insert disk" message -- but you already knew that, right?
Kickstart as a "BIOS-equivalent"? I don't know of any BIOS that support windowing, where you have terminal windows and common commands available to you, including the ability to launch new terminal windows. On the other hand I don't remember configuring drives or boot order from KickStart. On an Amiga, KickStart may not have been a full system, but it was not like a BIOS either.
Ah, you've bought the company line. Very good for you. I can respect that you "don't support wealth re-distribution" -- but that is the underlying premise of an economy. Perhaps what you meant to say was that you "support the tendency of unregulated wealth or power to concentrate." Which is rather like saying you don't support elevators because they work against gravity.
But the concentration is tendency only, not an absolute rule. And from time to time there are corrections. The ruling class in medieval europe forced (not necessarily consciously) Mediterranean agriculture on regions (now France, Germany and so on) where it was ineffective and inefficient, extremely intensive in man power for exceedingly low return. Such low returns that some historians discount the contemporary records for crop returns, refusing to accept the evidence. But a middle class still managed to emerge, not respected by the ruling class and despised by the impoverished masses, but still able to siphon off wealth and effect some of this redistribution you say you don't support.
The medieval lords have been replaced and now we have a small fraction of the population controlling the vast majority of the wealth. US production levels don't require near 100% employment to maintain, but the mercantile class has convinced the population at large that "work-for-pay" is not just one method of economic regulation, but the only right and proper method -- at least for the masses.
The balance will readjust at some point, but I do hope it is able to do so more gradually and peacefully than, say, the French Revolution. And there were impoverished people in revolutionary France who supported the authority, wealth and power of the aristocracy. It never fails that you can find people who support what objectively is counter to their better interests.
I don't think Apple would do it, but I sure wouldn't say never. At the least, developers will need a computer to actually develop on and Apple wants them to use one of /their/ computers to do so and pretty much by definition a development system can't be locked down against unapproved software. So I doubt they will entirely do away with "open" systems. On the other hand, they might very well market a "home computer" that was effectively locked down.
GP assumed that the cap mechanism would not count bytes because that is apparently supposed to be difficult or something. I'm not sure.
The traffic shaper we use where I work would be a bit of overkill for the job, but it could be used to enforce a capping mechanism. The hardware is, shocking I know, primarily sold to ISPs. And counting the number of bytes passing on the wire is trivial -- when low latency hardware can make shaping decisions based on source and destination IP address, ports used and the protocol employed (e.g., http regardless of the port it is running on) -- little things like tracking the number of bytes transferred are insignificant.
But even if I was going on the cheap and not using a real product with a real company behind it (important for PHBs to point the finger at in case of problems) then bandwidth capping can be accomplished at a byte-count level by processing flow data -- which any decent router can provide.
Seriously, an ISP is going to have some sort of traffic shaping device and it can count packets without increasing system load. Where a discrepancy is more likely to occur is between binary and decimal reckoning. Because I'd be surprised if an ISP didn't use 1,000,000,000 bytes to one gigabyte when specifying capping levels, but file systems still tend to report using powers of 1024 -- so when you download a 4.2 GB ISO it is likely to count as 4.5 GB of the stated quota.