Funny you should mention that. I spent a good chunk of this morning getting my MythTV setup back up and running due to continuing bizarre side effects of my last Seagate drive crash... and shivering in horror at the realization that the only drive I had lying around to clone the screeching, dying drive onto was also a Seagate of the same basic model (but PATA instead of SATA) and vintage.... I swear if hard drives get much more unreliable, I may go back to punch cards and VHS. At least VHS tapes didn't suddenly stop working, though you might have to splice them once in a while if the VCR got hungry.
If you do it right, they'd be pretty useless anyway. Ensure that no two missiles use the same code, order the codes randomly in the document, and don't release the secondary document that provides the lookup table for associating a particular missile with code number 79 on page 5428.:-)
The more common term is "tyranny of the majority", but that sounds like what is being advocated. Indeed, the sheep majority who are all too eager to give up all their rights at the first hint by a charlatan that doing so would help "protect" them is precisely why we have a Constitution and a Bill of Rights.
Granting retroactive immunity to a corporation is generally unconstitutional unless the federal government accompanies that by setting up a fund to compensate victims of the crime, effectively shifting the liability from the company to the government. Since the FISA bill did not do that, it will almost certainly be held unconstitutional. Unfortunately, it will take years for the courts to get around to undoing all the damage, and by the time they do, those same charlatans will have passed even more draconian laws that we'll have to spend even more years to wipe from the books. Any way you look at it, our nation became deeply screwed several years ago, and the only way to get control back in the hands of someone with even a modicum of concern for the well-being of the American people is to throw every single member of Congress out on his/her ass.
It's not a tough call at all. There's no grey area here. A bill is either good or evil. Period. Allowing telecom immunity is tantamount to saying that a guy who raped and murdered a child but spends every weekend volunteering at the homeless shelter and helping underprivileged kids is a great choice for a babysitter because he knows how to watch kids.... A bad rider on a good bill makes it a bad bill. One bad apple spoils the barrel and all that.
More to the point, not only is Obama a hypocrite, everyone who did not vote against this bill voted AGAINST the will of the American people---against the voters who elected them---and voted against the U.S. Constitution. Thus, they are twice hypocrites to the oath they swore before Congress:
"I do solemnly swear (or affirm) that I will support and defend the Constitution of the United States against all enemies, foreign and domestic; that I will bear true faith and allegiance to the same; that I take this obligation freely, without any mental reservation or purpose of evasion; and that I will well and faithfully discharge the duties of the office on which I am about to enter. So help me God.".
Can someone explain how any bill that retroactively grants permission for companies to conspire with illegal actions by the federal government to spy on its citizens and subvert the fourth amendment can possibly be interpreted in any way other than as a direct attack on the U.S. Constitution? Seriously? Anyone?
Everyone who voted in favor of the FISA legislation is also, IMHO, a traitor against the United States and is guilty of treason:
"whoever, owing allegiance to the United States, levies war against them or adheres to their enemies, giving them aid and comfort within the United States or elsewhere, is guilty of treason and shall suffer death, or shall be imprisoned not less than five years and fined under this title but not less than $10,000; and shall be incapable of holding any office under the United States."
Their actions are directly aiding and abetting terrorists by reducing the freedoms that those terrorists despise, thus effectively winning the terrorists' war from within our own government without the bad guys having to lift a finger. The whole lot of those Senators and Representatives should have their citizenship revoked and be ejected from this country for their disloyalty to the Constitution and to the American people.
Do your part. Vote to impeach Congress. Whoever the incumbent is, regardless of your party affiliation, vote for the other candidate. We have to send a message to our government that the public will not roll over and allow our rights to be trampled upon. We must do it NOW before it is too late. And elect an independent for President. But please, not Ron Paul....
I gathered from the comments that the faster random read of SSD caused more transactions to be performed per second, and that shortened the battery life as much as anything else.
In general, what eats battery power is writing and erasing flash. If you don't have enough RAM and end up paging to flash, that's going to cost lots of battery life (and SSD lifespan as well). There's also a wide range of power management among flash controllers from those that do little or no power management at all to those that only power up an individual flash part when needed. There's also the problem of computer filesystems being horrible in terms of minimizing writes to the flash. When you have to rewrite an entire 128kB flash block for a 4kB cluster write, you can see why this is inefficient.
More significantly, this means that even small improvements to write caching in the OS can make a huge difference in battery life. I would not be at all surprised if somebody turned around and did the same benchmark on a different OS and finds that the same SSD performs better than the hard drive instead of worse. Indeed, AnandTech did just that on Mac OS X and got very, very different results that showed the SSD providing a significant improvement in battery life.
This is, of course, comparing to drives at the 1.8" size, however. Those same tests with 2.5" drives showed the SSD being slightly worse than the latest hard drives, though still favorable compared to drives from a couple of years ago, and only on the order of 5% worse than the latest drives on average---nowhere near the difference seen in the Tom's Hardware test, and pretty clearly proving wrong what Tom's hardware said about 1.8" SSDs versus 1.8" drives ("As a result, the flash based SSD will lose the power consumption battle against 1.8" mechanical hard drives.").
Thus, the question of SSD power consumption becomes mostly one of how much of the wildly different results is due to better write caching, better hot files clustering, etc. in Mac OS X, and how much of it is due to differences in the workload between the two benchmarks used. Discuss.
Agreed. In a truly free market, most submarkets would quickly devolve into monopolies that would then abuse their monopoly power to ensure that no newcomers could enter the market either by flooding the market with goods at a loss until the newcomer went bankrupt or by using extra money to exhaust crucial resources from the newcomers' suppliers, ensuring that they could not obtain enough of those resources to meet demands. This, of course, assumes that there are still laws preventing what would be the obvious tools of a truly free market---knocking off their competition (assassinations), burning down their competitor's corporate headquarters/manufacturing facilities, stealing their competitor's physical assets, bribing banks/bankers to not give loans to their competitor, threatening businesses that distribute the competitor's product with pulling all of their most popular products (including products their competitor does not make) if the distributors don't drop all of their competitor's products, etc.
The promise of a free market as the solution to the world's ills is a fanciful notion that fools many who have never experienced anything resembling a free market. Those who have experienced it, however, immediately see right through such foolishness. Entrenched monopolies are hard to get rid of even with controls on monopolies. Without those controls, they become unstoppable rather rapidly.
Seems to me that if deserializing from XML is 10x slower than deserializing from any other format, unless you are either dealing with large binary blobs (in which case the performance benefits of being able to say "look for the next tag after 453 bytes" might make an order of magnitude difference) or are managing to transport lots of floating point numbers in a binary fashion, it probably isn't XML that sucks, but rather either your parser or your choice of tag names. You shouldn't be building up a full parse tree with a full XML parser and walking the DOM tree under those circumstances. If you use a lightweight XML dialect with short tag names and only entities for < and &, you can use a very fast, lightweight parser, e.g. something like the following (uncompiled, untested) code:
Ehmm, are you saying premeditated murder is only committed by psychopaths?
Most truly premeditated murders, yes. The vast majority of murders by people who are even remotely sane are crimes of passion (crimes committed under some level of duress), which by definition, are not premeditated.
Hans had a lot of circumstances stacked against him and then he did the worst thing possible and he opened his mouth in court, the guy has a hard time communicating with people that really want to understand him and hear what he has to say....
That's why I don't believe for a minute that it was premeditated. The guy is too incredibly bad at lying to be a psychopath, IMHO.
I wouldn't entirely rule out the "Nina Reiser put a hit out on him and said she would only let him live if he agreed to point the police towards the body of the random person the organized crime syndicate she hired offed a few days earlier who happens to have similar height, build, and gender to Nina Reiser" theory, either....
Until DNA evidence says it is her, I'm not convinced beyond reasonable doubt. Even if/when DNA evidence says it is her body, I'm not convinced beyond reasonable doubt that it was premeditated, which is what is necessary for a Murder I conviction. There's too much reason to believe that the jury convicted him more for his lack of social skills than because of proof of premeditation beyond reasonable doubt....
Besides, with all his reported strangeness, it isn't at all improbable that she died accidentally. There's far more reason to suspect some sort of super-rough, kinky sex gone terribly wrong than premeditated murder.... The preponderance of evidence actually points away from the jury verdict, at least based on what we've seen thus far. Maybe more evidence will come out eventually, but....
Her intent was not to cause harm to MySpace, though. That's technically where the CFaAA is supposed to begin and end.... But yeah, her intent was to cause harm, which makes her either A. a pretty mean-spirited person or B. someone who doesn't think before she acts. Either way, she should be pitied, not given prison time.
Sure that only works for explicit malloc calls. You can also do the same override tricks for any library function if you want, though. The best way, of course, is to use something like dtrace to simply snag the actual malloc and free calls and do all sorts of statistics or, as you noted, a debug malloc library that overrides the symbols at the linker level, though even that won't catch things like C++ new operators, page-sized allocations created using system-call-based allocators like vm_alloc or mmap, etc. That said, you can do a lot with dtrace in that area.
The problem with standard malloc guard areas is that they only detect scribbling at certain key points---when you return from a function, when you do a malloc, etc.---and are way too slow to leave turned on continuously. You can't have the CPU trap on writes to any blocks smaller than an entire page, and hacking malloc() to always return page-sized blocks doesn't really detect buffer overflows unless you leave it set that way and waste a truckload of memory. Those are neat tricks to help debug things, but you still need to do fuzzing of the inputs or similar to have a prayer of catching anything with them.
Whenever C routines automatically allocate a new buffer, someone had better free() it later. You solve the buffer overflow problem and replace it with a memory leak.
Only if you do something particularly sloppy. In the case of the concatenation, the only thing you could possibly leak that you wouldn't otherwise leak would be the first argument, and the common usage would probably be char *strc = stracat(stra, strb); free(stra); or similar. For that matter, you could add a third function, char *stracatfree(char *, char*) that did both at once that you could use if you know you aren't going to need the first value again.
Besides, memory leaks are easily found through trivial testing techniques. Memory always leaks for a substantial percentage of the executions of a function if it is going to leak at all. The exceptions are pretty rare. Even for the rare cases, though, there are plenty of tools that make it easier to find them. Buffer overflows, however, are a pain in the backside to find and fix (particularly when they stomp on a pointer, which in turn causes some other piece of code elsewhere in the app to stomp on a completely unrelated piece of memory), and are far more likely to be found by somebody through accidentally stumbling across it than through any meaningful testing process.... Given a choice, I'd take a million memory leaks over one buffer overflow.
I mean, I can tell you how much memory leaked by doing something like this (WARNING: not thoroughly tested code):
Usually (at least outside of embedded applications) there's not much more effort needed to see if your code is going to leak like a sieve beyond doing this and running it through the major code paths. Just my $0.02.
Agreed. Signing up with a false name should only be fraud if one party can show financial harm or intent to cause damage. Otherwise, it is simply a breach of contract, which falls squarely into civil, not criminal law. I'd bet money that this case will be laughed out of court. At least on the surface, this screams prosecutorial misconduct.
That said, the Computer Fraud and Abuse Act is pretty broken, particularly with the "PATRIOT" Act "enhancements". They pretty thoroughly make working with computers into a minefield. Nearly everyone on the Internet has probably been on the wrong side of it at least once. Basically, it's a law designed to ensure that everyone is a criminal so that they can screw people over if you get on their bad side. Sadly, this could be interpreted as falling into the list of things that are criminal acts under that law.
What makes this particularly bizarre is that the only reason this is in any way a criminal act is because of the incidental use of MySpace as a vehicle. The same sort of attacks could have driven this person to suicide without that technological help and it would have been legal. In effect, the CFaAA basically boils down to "illegal on the Internet" laws, which is really idiotic. Something legal in person should be legal on the Internet, regardless of the inadvertent side effect of driving some kid to suicide. If you want to make it illegal on the Internet, it should also be illegal to do that same thing IRL. The Internet certainly shouldn't be held to a higher standard, and given the lack of any real verifiable identity on the Internet, should generally be held to a much lower standard.
I'd use C for everything if its string handling weren't so utterly broken. It makes sense from a pure performance, near-the-hardware perspective to handle strings the way it does, but what is desperately needed is POSIX standardization of stracat/stracpy. That one change would make it trivial to eliminate 99% of all buffer overflows in typical apps (and, ideally, kernel code)....
char *stracpy(char *src)
Returns a copy of a string in a newly allocated buffer large enough to hold the result.
char *stracat(char *string1, char *string2...)
Returns a newly allocated string that contains the concatenation of two or more strings.
Sure, I could write my own, and pretty much everybody on the planet has done it at least once in their lives, but it seems ridiculous that such obvious operations require hand holding. Outside of embedded platforms where malloc is expensive, dynamically allocating a new buffer for concatenations should be the norm. For situations where performance takes a big hit because you can't afford to copy the source string every time, you could also add this API:
strbuf_t strbufalloc(char *src [, size_t size_hint])
Returns a copy of a string in a newly allocated buffer large enough to hold the result. The optional second parameter size_hint allows you to specify a suggested size for the buffer.
strbuf *strbufcat(strbuf_t string1, strbuf_t string2...)
Returns a newly allocated string that contains the concatenation of two or more strings. The value of string1 must be a string buffer (strbuf_t). The value of string2 may be either a strbuf_t or char **.
and the routines automatically realloc the internal char * buffer as needed to grow the size to hold incoming content and likely future content based on historical data about prior operations on the structure, the algorithm for which would be an implementation detail outside the scope of the standard. That said, for the rare cases where a simple copy hurts performance badly enough for most folks to care, chances are your application knows its own growth policy better than the OS could guess anyway, so you probably should just roll your own buffer management code. The important bits to add to the standard are those first two functions (stracpy and stracat).
I'd love to see a "computer-in-the-screen" laptop with a snap-on Bluetooth keyboard and built-in hinge stand for the screen so that you don't have a hinge with cables running through it but still have the portability of a laptop (for the most part).
Other things I'd like to see:
Keyboards that recess when the laptop is closed so that they can't touch the screen and create smudges/scratches/creases.
More reliable hinge construction that isn't so subject to metal fatigue. Hinges on my door don't fail after two or three years and they hold a hundred times more weight. Enough with the cheap pot metal already.
More reliable wiring through the hinges. Ideally, don't run wires through the hinges, but instead, use the left hinge itself to carry power and data and the right hinge as a ground. Use high voltage, low current DC through the hinge (post-inverter) and multiplex send and receive data on the power side in a half-duplex fashion isochronously alternating between the two directions. All you have to carry is a stream of pixel data changes and data for the mic/camera/wireless.
Screens that are easier to clean. Smudges suck in sunlight, and plastic screen faces are really hard to clean. Glass, though, is probably too fragile unless you reinforce the screen dramatically. Not sure what would be a good alternative. Maybe some easy-clean coating on the plastic screen. Dunno.
Ultracapacitors instead of batteries (when the energy density gets there).
Better HD reliability
Triple screens with two that slide sideways in opposite directions, then slide forwards/backwards into the same plane and hinge in such a way that the front edges of the screen stay nearly together/borderless, this giving surround visuals and more screen real estate without sacrificing footprint size when closed.
All the hardware and all the software. Not just the hardware. If they stole a $700 copy of Photoshop (retail), for example, that can push the dollar figure way up.
I would also add that your best bet is to contact the upstream ISP for the IP number and inform them of the situation. Let them know that you need to work with police on it and need to know what city it is in so that you can get a local PD to follow up. They won't give you the address without a warrant or court order, but they should be willing to give you the city. Once you know what city it is in, you can then contact their PD and follow up with their computer crimes division, assuming they have one. If they don't, ask who their most computer-savvy officer is. They're bound to have at least one or two people who help maintain their website on the side as a minimum. Try to work with that officer (or if it's a non-officer staffer, try to work with an officer through them since having somebody who understands tech who the officers already know will put you in a better position as far as getting them to trust you).
Do a traceroute to the IP number (or if it's behind a wormhole route, do a traceroute from their IP to www.google.com or something) and see if you see any useful domain names in the trace. If so, it doesn't matter who owns the netblock. Go to whoever owns the domain. If, for example, you can track it to a university campus, you're in even better shape, of course, as they are more likely to work with you without the need to get a warrant if you can show that the computer is stolen and that you are in control over a computer on their network. They are also likely to be technically competent as would an ISP, but unlike an ISP, they have user agreements that almost certainly allow them to investigate their users. Then, ask them to help you work with the campus cops to get your laptop back.
The latter group clump, and they're the ones the advertisers are primarily after, because they are large, easy markets. They're also more easily served by television, and they self-amplify the buzz about whatever the carefully marketed Next Big Thing is.
Flip side, they're more easily targeted by cable TV than broadcast TV, and the networks are getting little real advantage from maintaining all those legacy broadcast TV stations. Frankly, the OTA TV is a money sink for them, and that's one reason they're struggling. Their license agreements with the OTA stations make it much harder for them to expand to new delivery mechanisms, etc.
Broadcast TV, unlike radio, has nothing to offer over alternatives, though. Radio is still around because you can listen to it in your car easily. The number of home radio listeners in the U.S. is rapidly declining in the face of competing technologies.
Broadcast TV... very, very few people watch in cars, and once analog TV goes away, doing so really won't be practical because of the antenna requirements. After analog TV goes away, there will be virtually no one watching TV except through cable. Even now, barely over 10% of TV households in the U.S. lack cable or satellite service. A large percentage of those people are in fringe reception areas and will not even be able to receive HDTV signals because instead of degrading gracefully like analog signals do, digital TV below a certain signal level just plain can't be decoded. HDTV is the beginning of the end for OTA TV, and good riddance.
The way I see it, we've been wasting hundreds of megawatts of power on these transmitters for no good reason for several years. It would be cheaper for the stations to shut down their transmitters, become cable-only/satellite-only stations, and buy DirecTV service for every person who complains than to continue maintaining the broadcast TV infrastructure (nationwide, about $4.2 billion annually to cover one DirecTV set per affected household versus some $61 billion annually in electrical costs for the transmitters by themselves, not counting equipment maintenance costs, keeping a licensed engineer on staff to maintain it, etc.)
I disagree. I think that Microsoft is well aware that it isn't the only game in town. What they don't understand, is how to remedy the situation.
"Remedy the situation" is an interesting choice of words, as it could be interpreted two ways---the way that you probably meant it and the way that is more accurate but less flattering. IMHO, it's not that they don't know how to survive and thrive as one of many players, but rather that they don't know how to get back to a monopoly state. Microsoft's fundamental problem is that their corporate goal does not seem to merely be doing well for themselves as a company, but rather making sure nobody else does/can. It's a completely backwards corporate mentality and will eventually be their downfall in much the same way that treating their customers as likely criminals has hurt them significantly. The goal of a company cannot be to eke out every last possible cent.
Put another way, the goal of a company must be to remain reasonably profitable while behaving responsibly, reasonably, and treating their customers, suppliers, and even their competitors with due respect. Sure, sociopathic corporate behavior serves companies well in the short term, but as Microsoft is seeing now, it eventually comes back to bite them in the you-know-what.
Well, it's not just that. The Internet has helped, sure, but the biggest problem the networks face is declining viewership as cable channels do better and better jobs at hitting more specific niches. You have channels for everything from sci-fi to home improvement. The Internet merely takes that one step farther and creates channels for everything from nude archery to watching people's feet as they walk past aisles of clothing at J.C. Penney.
The point is that as the availability of options increases, the interest in individual options decreases, and younger viewers are far more likely to find those new options and take advantage of them than older viewers simply because they are more connected with other people. You hear about things on TV, the radio, email, around work, etc. Retired people have much more limited ways to find out about these things, and thus are much less likely to end up watching the Smurfs With Green Moustaches Drawn On By Monkeys In Tutus Hour. Therefore, the older demographic will be much slower to transition away from legacy technologies like broadcast TV and towards more niche-oriented content like cable channels, towards more on-demand technologies like iTunes, and towards more peer-generated services like YouTube.
I predicted the death of broadcast TV back in 1995. IIRC, I gave it 10-15 years. It may take a little longer, but I suspect I was a lot closer than the folks who read my essay suspected....
If this story is true, then whatever harebrained idiot thought this one up should have to do penance in the form of having to take the place of one of those undocumented maids for the next twenty years. That said, I don't see anything in that law that suggests that computer repair people have to be licensed PIs. The only people that are covered there are people who are doing forensic analysis on data not available to the general public. If you hire someone to do computer forensics (e.g. investigating the contents of a hard drive), that's a completely different service from merely replacing a defective power supply or even reinstalling Windows. Stretching that law to cover basic computer repairs is a fairly blatant perversion of the law as written and almost certainly won't hold up in court unless I'm either grossly misreading it or the story linked from this one is linking to the wrong law.
In any case, assuming the story is legit, let's take this same logic one step further. A maid finds child porn while cleaning some guy's den. We should, therefore, obviously require that every illegal, undocumented maid working in the state of Texas have a PI license. Similarly, every maintenance crew working for a company, every IT employee, every office assistant who might potentially use his/her boss's computer, every school computer lab administrator, every plumber (child porn could be hidden under the sink, you know), every electrician (going to rewire somebody's entertainment center), and every employee at every hard drive refurbishing center.
In short, this same logic would apply equally to large swaths of our population for precisely the same reason, and I predict this law will be struck down swiftly for precisely that reason. It unfairly singles out one small group for regulation out of a much larger group of people for whom the same conditions apply.
Further, as someone said a couple of posts up, the difference between laws requiring a PI license for this and laws requiring a PI license for someone doing an investigation, a medical license for a doctor, etc. is that in all of the cases where such laws have been considered constitutional in the past, the reason for the license was for the protection of the person hiring out for the work to ensure that the person doesn't get shafted, while in this case, the laws are predominantly for the protection of the state and are in direct contradiction to the needs, desires, and best interests of the person hiring out for the work.
As for planting evidence, there's really no more protection against that just because somebody has a PI license. There are plenty of crooked licensed private investigators, lawyers, doctors, etc. At best, there is the additional disincentive of losing your license if caught, but it's not like a computer repair tech can't get a job doing computer repair in a corporate IT department, which presumably would not entail such licensing requirements, or else there are likely to be a lot of high-tech companies (e.g. Apple, Dell, etc.) telling Texas to go f*ck themselves and moving their operations to another state.
More importantly, computer companies that contract out mail-in repairs are likely to eschew Texas from now on. Why? Too much extra expense. Instead of hiring a minimum-wage person and training them in a week, they'll have to hire someone with an expensive license and/or spend months training them at tremendous expense. I know a couple of businesses that are likely to dry up overnight.
Sounds like yet another stupid law written by stupid people for stupid reasons that won't actually fix what it was intended to fix. Since that describes about 98% of all laws passed in my lifetime, could somebody explain why this is news?:-)
What about the slightly slower crystal on the IIsi? It served no purpose other than to create a slower machine.
Well, a smaller machine requires lower thermal output, so there might have been a need to lower the clock speed for that reason, though that's probably a stretch. The obvious purpose, though is product differentiation. Clocking it at the same speed as the IIci would have cannibalized sales of the IIci. Doesn't seem that unreasonable to me.
Slashdot Mirror
Funny you should mention that. I spent a good chunk of this morning getting my MythTV setup back up and running due to continuing bizarre side effects of my last Seagate drive crash... and shivering in horror at the realization that the only drive I had lying around to clone the screeching, dying drive onto was also a Seagate of the same basic model (but PATA instead of SATA) and vintage.... I swear if hard drives get much more unreliable, I may go back to punch cards and VHS. At least VHS tapes didn't suddenly stop working, though you might have to splice them once in a while if the VCR got hungry.
The day the drive crashes and you lose everything.
That day might come sooner than you think....
If you do it right, they'd be pretty useless anyway. Ensure that no two missiles use the same code, order the codes randomly in the document, and don't release the secondary document that provides the lookup table for associating a particular missile with code number 79 on page 5428. :-)
The more common term is "tyranny of the majority", but that sounds like what is being advocated. Indeed, the sheep majority who are all too eager to give up all their rights at the first hint by a charlatan that doing so would help "protect" them is precisely why we have a Constitution and a Bill of Rights.
Granting retroactive immunity to a corporation is generally unconstitutional unless the federal government accompanies that by setting up a fund to compensate victims of the crime, effectively shifting the liability from the company to the government. Since the FISA bill did not do that, it will almost certainly be held unconstitutional. Unfortunately, it will take years for the courts to get around to undoing all the damage, and by the time they do, those same charlatans will have passed even more draconian laws that we'll have to spend even more years to wipe from the books. Any way you look at it, our nation became deeply screwed several years ago, and the only way to get control back in the hands of someone with even a modicum of concern for the well-being of the American people is to throw every single member of Congress out on his/her ass.
It's not a tough call at all. There's no grey area here. A bill is either good or evil. Period. Allowing telecom immunity is tantamount to saying that a guy who raped and murdered a child but spends every weekend volunteering at the homeless shelter and helping underprivileged kids is a great choice for a babysitter because he knows how to watch kids.... A bad rider on a good bill makes it a bad bill. One bad apple spoils the barrel and all that.
More to the point, not only is Obama a hypocrite, everyone who did not vote against this bill voted AGAINST the will of the American people---against the voters who elected them---and voted against the U.S. Constitution. Thus, they are twice hypocrites to the oath they swore before Congress:
Can someone explain how any bill that retroactively grants permission for companies to conspire with illegal actions by the federal government to spy on its citizens and subvert the fourth amendment can possibly be interpreted in any way other than as a direct attack on the U.S. Constitution? Seriously? Anyone?
Everyone who voted in favor of the FISA legislation is also, IMHO, a traitor against the United States and is guilty of treason:
Their actions are directly aiding and abetting terrorists by reducing the freedoms that those terrorists despise, thus effectively winning the terrorists' war from within our own government without the bad guys having to lift a finger. The whole lot of those Senators and Representatives should have their citizenship revoked and be ejected from this country for their disloyalty to the Constitution and to the American people.
Do your part. Vote to impeach Congress. Whoever the incumbent is, regardless of your party affiliation, vote for the other candidate. We have to send a message to our government that the public will not roll over and allow our rights to be trampled upon. We must do it NOW before it is too late. And elect an independent for President. But please, not Ron Paul....
In general, what eats battery power is writing and erasing flash. If you don't have enough RAM and end up paging to flash, that's going to cost lots of battery life (and SSD lifespan as well). There's also a wide range of power management among flash controllers from those that do little or no power management at all to those that only power up an individual flash part when needed. There's also the problem of computer filesystems being horrible in terms of minimizing writes to the flash. When you have to rewrite an entire 128kB flash block for a 4kB cluster write, you can see why this is inefficient.
More significantly, this means that even small improvements to write caching in the OS can make a huge difference in battery life. I would not be at all surprised if somebody turned around and did the same benchmark on a different OS and finds that the same SSD performs better than the hard drive instead of worse. Indeed, AnandTech did just that on Mac OS X and got very, very different results that showed the SSD providing a significant improvement in battery life.
This is, of course, comparing to drives at the 1.8" size, however. Those same tests with 2.5" drives showed the SSD being slightly worse than the latest hard drives, though still favorable compared to drives from a couple of years ago, and only on the order of 5% worse than the latest drives on average---nowhere near the difference seen in the Tom's Hardware test, and pretty clearly proving wrong what Tom's hardware said about 1.8" SSDs versus 1.8" drives ("As a result, the flash based SSD will lose the power consumption battle against 1.8" mechanical hard drives.").
Thus, the question of SSD power consumption becomes mostly one of how much of the wildly different results is due to better write caching, better hot files clustering, etc. in Mac OS X, and how much of it is due to differences in the workload between the two benchmarks used. Discuss.
Agreed. In a truly free market, most submarkets would quickly devolve into monopolies that would then abuse their monopoly power to ensure that no newcomers could enter the market either by flooding the market with goods at a loss until the newcomer went bankrupt or by using extra money to exhaust crucial resources from the newcomers' suppliers, ensuring that they could not obtain enough of those resources to meet demands. This, of course, assumes that there are still laws preventing what would be the obvious tools of a truly free market---knocking off their competition (assassinations), burning down their competitor's corporate headquarters/manufacturing facilities, stealing their competitor's physical assets, bribing banks/bankers to not give loans to their competitor, threatening businesses that distribute the competitor's product with pulling all of their most popular products (including products their competitor does not make) if the distributors don't drop all of their competitor's products, etc.
The promise of a free market as the solution to the world's ills is a fanciful notion that fools many who have never experienced anything resembling a free market. Those who have experienced it, however, immediately see right through such foolishness. Entrenched monopolies are hard to get rid of even with controls on monopolies. Without those controls, they become unstoppable rather rapidly.
Seems to me that if deserializing from XML is 10x slower than deserializing from any other format, unless you are either dealing with large binary blobs (in which case the performance benefits of being able to say "look for the next tag after 453 bytes" might make an order of magnitude difference) or are managing to transport lots of floating point numbers in a binary fashion, it probably isn't XML that sucks, but rather either your parser or your choice of tag names. You shouldn't be building up a full parse tree with a full XML parser and walking the DOM tree under those circumstances. If you use a lightweight XML dialect with short tag names and only entities for < and &, you can use a very fast, lightweight parser, e.g. something like the following (uncompiled, untested) code:
int curstate; ...
...
...
char *pos =
char *curtag = NULL;
if (*pos == '<') { pos++; curstate = parse_tag_contents(&pos, &curtag); }
int parse_tag_contents(char **current_position_in_file, char **tagname) {
char *startpos = *current_position_in_file, *endpos;
int retval = ENTERING_TAG_CONTEXT;
if (*startpos == '/') { retval = LEAVING_TAG_CONTEXT; startpos++;}
endpos = startpos;
while (*endpos && *endpos != ' ' && *endpos != '\t' && * endpos != '\r' && * endpos != '\n' && * endpos != '>') { endpos++; }
char terminator = *endpos;
*endpos = '\0';
asprintf(tagname, "%s", startpos);
*endpos = terminator;
while (*endpos != '>') { endpos++; }
endpos++;
*current_position_in_file = endpos;
return retval;
}
And you've just done everything you need to do to parse the guts of an XML tag in such a minimal dialect....
Most truly premeditated murders, yes. The vast majority of murders by people who are even remotely sane are crimes of passion (crimes committed under some level of duress), which by definition, are not premeditated.
That's why I don't believe for a minute that it was premeditated. The guy is too incredibly bad at lying to be a psychopath, IMHO.
I wouldn't entirely rule out the "Nina Reiser put a hit out on him and said she would only let him live if he agreed to point the police towards the body of the random person the organized crime syndicate she hired offed a few days earlier who happens to have similar height, build, and gender to Nina Reiser" theory, either....
Until DNA evidence says it is her, I'm not convinced beyond reasonable doubt. Even if/when DNA evidence says it is her body, I'm not convinced beyond reasonable doubt that it was premeditated, which is what is necessary for a Murder I conviction. There's too much reason to believe that the jury convicted him more for his lack of social skills than because of proof of premeditation beyond reasonable doubt....
Besides, with all his reported strangeness, it isn't at all improbable that she died accidentally. There's far more reason to suspect some sort of super-rough, kinky sex gone terribly wrong than premeditated murder.... The preponderance of evidence actually points away from the jury verdict, at least based on what we've seen thus far. Maybe more evidence will come out eventually, but....
Her intent was not to cause harm to MySpace, though. That's technically where the CFaAA is supposed to begin and end.... But yeah, her intent was to cause harm, which makes her either A. a pretty mean-spirited person or B. someone who doesn't think before she acts. Either way, she should be pitied, not given prison time.
Sure that only works for explicit malloc calls. You can also do the same override tricks for any library function if you want, though. The best way, of course, is to use something like dtrace to simply snag the actual malloc and free calls and do all sorts of statistics or, as you noted, a debug malloc library that overrides the symbols at the linker level, though even that won't catch things like C++ new operators, page-sized allocations created using system-call-based allocators like vm_alloc or mmap, etc. That said, you can do a lot with dtrace in that area.
The problem with standard malloc guard areas is that they only detect scribbling at certain key points---when you return from a function, when you do a malloc, etc.---and are way too slow to leave turned on continuously. You can't have the CPU trap on writes to any blocks smaller than an entire page, and hacking malloc() to always return page-sized blocks doesn't really detect buffer overflows unless you leave it set that way and waste a truckload of memory. Those are neat tricks to help debug things, but you still need to do fuzzing of the inputs or similar to have a prayer of catching anything with them.
Only if you do something particularly sloppy. In the case of the concatenation, the only thing you could possibly leak that you wouldn't otherwise leak would be the first argument, and the common usage would probably be char *strc = stracat(stra, strb); free(stra); or similar. For that matter, you could add a third function, char *stracatfree(char *, char*) that did both at once that you could use if you know you aren't going to need the first value again.
Besides, memory leaks are easily found through trivial testing techniques. Memory always leaks for a substantial percentage of the executions of a function if it is going to leak at all. The exceptions are pretty rare. Even for the rare cases, though, there are plenty of tools that make it easier to find them. Buffer overflows, however, are a pain in the backside to find and fix (particularly when they stomp on a pointer, which in turn causes some other piece of code elsewhere in the app to stomp on a completely unrelated piece of memory), and are far more likely to be found by somebody through accidentally stumbling across it than through any meaningful testing process.... Given a choice, I'd take a million memory leaks over one buffer overflow.
I mean, I can tell you how much memory leaked by doing something like this (WARNING: not thoroughly tested code):
#include <stdlib.h>
#include <sys/types.h>
#include <stdio.h>
size_t global_leak_counter = 0;
void *countmalloc(size_t a) { void *ret = malloc(a + sizeof(size_t)); global_leak_counter+= (a); *(size_t *)ret = global_leak_counter; return ret+sizeof(size_t); }
void countfree(void *a) { void *x = a-sizeof(size_t); global_leak_counter -= *(size_t *)x; }
#define malloc(a) countmalloc(a)
#define free(a) countfree(a)
int main(int argc, char *argv[])
{
char *x = malloc(4);
char *y = malloc(13);
fprintf(stderr, "Size: %zu bytes\n", global_leak_counter);
free(x);
fprintf(stderr, "Leaked %zu bytes\n", global_leak_counter);
}
Usually (at least outside of embedded applications) there's not much more effort needed to see if your code is going to leak like a sieve beyond doing this and running it through the major code paths. Just my $0.02.
Agreed. Signing up with a false name should only be fraud if one party can show financial harm or intent to cause damage. Otherwise, it is simply a breach of contract, which falls squarely into civil, not criminal law. I'd bet money that this case will be laughed out of court. At least on the surface, this screams prosecutorial misconduct.
That said, the Computer Fraud and Abuse Act is pretty broken, particularly with the "PATRIOT" Act "enhancements". They pretty thoroughly make working with computers into a minefield. Nearly everyone on the Internet has probably been on the wrong side of it at least once. Basically, it's a law designed to ensure that everyone is a criminal so that they can screw people over if you get on their bad side. Sadly, this could be interpreted as falling into the list of things that are criminal acts under that law.
What makes this particularly bizarre is that the only reason this is in any way a criminal act is because of the incidental use of MySpace as a vehicle. The same sort of attacks could have driven this person to suicide without that technological help and it would have been legal. In effect, the CFaAA basically boils down to "illegal on the Internet" laws, which is really idiotic. Something legal in person should be legal on the Internet, regardless of the inadvertent side effect of driving some kid to suicide. If you want to make it illegal on the Internet, it should also be illegal to do that same thing IRL. The Internet certainly shouldn't be held to a higher standard, and given the lack of any real verifiable identity on the Internet, should generally be held to a much lower standard.
I'd use C for everything if its string handling weren't so utterly broken. It makes sense from a pure performance, near-the-hardware perspective to handle strings the way it does, but what is desperately needed is POSIX standardization of stracat/stracpy. That one change would make it trivial to eliminate 99% of all buffer overflows in typical apps (and, ideally, kernel code)....
char *stracpy(char *src) Returns a copy of a string in a newly allocated buffer large enough to hold the result. char *stracat(char *string1, char *string2Sure, I could write my own, and pretty much everybody on the planet has done it at least once in their lives, but it seems ridiculous that such obvious operations require hand holding. Outside of embedded platforms where malloc is expensive, dynamically allocating a new buffer for concatenations should be the norm. For situations where performance takes a big hit because you can't afford to copy the source string every time, you could also add this API:
strbuf_t strbufalloc(char *src [, size_t size_hint]) Returns a copy of a string in a newly allocated buffer large enough to hold the result. The optional second parameter size_hint allows you to specify a suggested size for the buffer. strbuf *strbufcat(strbuf_t string1, strbuf_t string2Wherein strbuf is opaquely defined as:
struct strbuf_private { // null-terminated string
char *content;
size_t length;
void *private_history_data;
};
struct strbuf;
typedef struct strbuf *strbuf_t;
and the routines automatically realloc the internal char * buffer as needed to grow the size to hold incoming content and likely future content based on historical data about prior operations on the structure, the algorithm for which would be an implementation detail outside the scope of the standard. That said, for the rare cases where a simple copy hurts performance badly enough for most folks to care, chances are your application knows its own growth policy better than the OS could guess anyway, so you probably should just roll your own buffer management code. The important bits to add to the standard are those first two functions (stracpy and stracat).
I'd love to see a "computer-in-the-screen" laptop with a snap-on Bluetooth keyboard and built-in hinge stand for the screen so that you don't have a hinge with cables running through it but still have the portability of a laptop (for the most part).
Other things I'd like to see:
Maybe it's just me?
All the hardware and all the software. Not just the hardware. If they stole a $700 copy of Photoshop (retail), for example, that can push the dollar figure way up.
I would also add that your best bet is to contact the upstream ISP for the IP number and inform them of the situation. Let them know that you need to work with police on it and need to know what city it is in so that you can get a local PD to follow up. They won't give you the address without a warrant or court order, but they should be willing to give you the city. Once you know what city it is in, you can then contact their PD and follow up with their computer crimes division, assuming they have one. If they don't, ask who their most computer-savvy officer is. They're bound to have at least one or two people who help maintain their website on the side as a minimum. Try to work with that officer (or if it's a non-officer staffer, try to work with an officer through them since having somebody who understands tech who the officers already know will put you in a better position as far as getting them to trust you).
Do a traceroute to the IP number (or if it's behind a wormhole route, do a traceroute from their IP to www.google.com or something) and see if you see any useful domain names in the trace. If so, it doesn't matter who owns the netblock. Go to whoever owns the domain. If, for example, you can track it to a university campus, you're in even better shape, of course, as they are more likely to work with you without the need to get a warrant if you can show that the computer is stolen and that you are in control over a computer on their network. They are also likely to be technically competent as would an ISP, but unlike an ISP, they have user agreements that almost certainly allow them to investigate their users. Then, ask them to help you work with the campus cops to get your laptop back.
Flip side, they're more easily targeted by cable TV than broadcast TV, and the networks are getting little real advantage from maintaining all those legacy broadcast TV stations. Frankly, the OTA TV is a money sink for them, and that's one reason they're struggling. Their license agreements with the OTA stations make it much harder for them to expand to new delivery mechanisms, etc.
Broadcast TV, unlike radio, has nothing to offer over alternatives, though. Radio is still around because you can listen to it in your car easily. The number of home radio listeners in the U.S. is rapidly declining in the face of competing technologies.
Broadcast TV... very, very few people watch in cars, and once analog TV goes away, doing so really won't be practical because of the antenna requirements. After analog TV goes away, there will be virtually no one watching TV except through cable. Even now, barely over 10% of TV households in the U.S. lack cable or satellite service. A large percentage of those people are in fringe reception areas and will not even be able to receive HDTV signals because instead of degrading gracefully like analog signals do, digital TV below a certain signal level just plain can't be decoded. HDTV is the beginning of the end for OTA TV, and good riddance.
The way I see it, we've been wasting hundreds of megawatts of power on these transmitters for no good reason for several years. It would be cheaper for the stations to shut down their transmitters, become cable-only/satellite-only stations, and buy DirecTV service for every person who complains than to continue maintaining the broadcast TV infrastructure (nationwide, about $4.2 billion annually to cover one DirecTV set per affected household versus some $61 billion annually in electrical costs for the transmitters by themselves, not counting equipment maintenance costs, keeping a licensed engineer on staff to maintain it, etc.)
"Remedy the situation" is an interesting choice of words, as it could be interpreted two ways---the way that you probably meant it and the way that is more accurate but less flattering. IMHO, it's not that they don't know how to survive and thrive as one of many players, but rather that they don't know how to get back to a monopoly state. Microsoft's fundamental problem is that their corporate goal does not seem to merely be doing well for themselves as a company, but rather making sure nobody else does/can. It's a completely backwards corporate mentality and will eventually be their downfall in much the same way that treating their customers as likely criminals has hurt them significantly. The goal of a company cannot be to eke out every last possible cent.
Put another way, the goal of a company must be to remain reasonably profitable while behaving responsibly, reasonably, and treating their customers, suppliers, and even their competitors with due respect. Sure, sociopathic corporate behavior serves companies well in the short term, but as Microsoft is seeing now, it eventually comes back to bite them in the you-know-what.
Well, it's not just that. The Internet has helped, sure, but the biggest problem the networks face is declining viewership as cable channels do better and better jobs at hitting more specific niches. You have channels for everything from sci-fi to home improvement. The Internet merely takes that one step farther and creates channels for everything from nude archery to watching people's feet as they walk past aisles of clothing at J.C. Penney.
The point is that as the availability of options increases, the interest in individual options decreases, and younger viewers are far more likely to find those new options and take advantage of them than older viewers simply because they are more connected with other people. You hear about things on TV, the radio, email, around work, etc. Retired people have much more limited ways to find out about these things, and thus are much less likely to end up watching the Smurfs With Green Moustaches Drawn On By Monkeys In Tutus Hour. Therefore, the older demographic will be much slower to transition away from legacy technologies like broadcast TV and towards more niche-oriented content like cable channels, towards more on-demand technologies like iTunes, and towards more peer-generated services like YouTube.
I predicted the death of broadcast TV back in 1995. IIRC, I gave it 10-15 years. It may take a little longer, but I suspect I was a lot closer than the folks who read my essay suspected....
If this story is true, then whatever harebrained idiot thought this one up should have to do penance in the form of having to take the place of one of those undocumented maids for the next twenty years. That said, I don't see anything in that law that suggests that computer repair people have to be licensed PIs. The only people that are covered there are people who are doing forensic analysis on data not available to the general public. If you hire someone to do computer forensics (e.g. investigating the contents of a hard drive), that's a completely different service from merely replacing a defective power supply or even reinstalling Windows. Stretching that law to cover basic computer repairs is a fairly blatant perversion of the law as written and almost certainly won't hold up in court unless I'm either grossly misreading it or the story linked from this one is linking to the wrong law.
In any case, assuming the story is legit, let's take this same logic one step further. A maid finds child porn while cleaning some guy's den. We should, therefore, obviously require that every illegal, undocumented maid working in the state of Texas have a PI license. Similarly, every maintenance crew working for a company, every IT employee, every office assistant who might potentially use his/her boss's computer, every school computer lab administrator, every plumber (child porn could be hidden under the sink, you know), every electrician (going to rewire somebody's entertainment center), and every employee at every hard drive refurbishing center.
In short, this same logic would apply equally to large swaths of our population for precisely the same reason, and I predict this law will be struck down swiftly for precisely that reason. It unfairly singles out one small group for regulation out of a much larger group of people for whom the same conditions apply.
Further, as someone said a couple of posts up, the difference between laws requiring a PI license for this and laws requiring a PI license for someone doing an investigation, a medical license for a doctor, etc. is that in all of the cases where such laws have been considered constitutional in the past, the reason for the license was for the protection of the person hiring out for the work to ensure that the person doesn't get shafted, while in this case, the laws are predominantly for the protection of the state and are in direct contradiction to the needs, desires, and best interests of the person hiring out for the work.
As for planting evidence, there's really no more protection against that just because somebody has a PI license. There are plenty of crooked licensed private investigators, lawyers, doctors, etc. At best, there is the additional disincentive of losing your license if caught, but it's not like a computer repair tech can't get a job doing computer repair in a corporate IT department, which presumably would not entail such licensing requirements, or else there are likely to be a lot of high-tech companies (e.g. Apple, Dell, etc.) telling Texas to go f*ck themselves and moving their operations to another state.
More importantly, computer companies that contract out mail-in repairs are likely to eschew Texas from now on. Why? Too much extra expense. Instead of hiring a minimum-wage person and training them in a week, they'll have to hire someone with an expensive license and/or spend months training them at tremendous expense. I know a couple of businesses that are likely to dry up overnight.
Sounds like yet another stupid law written by stupid people for stupid reasons that won't actually fix what it was intended to fix. Since that describes about 98% of all laws passed in my lifetime, could somebody explain why this is news? :-)
Now, now, don't cowtow (sic) to the moderators. :-)
Well, a smaller machine requires lower thermal output, so there might have been a need to lower the clock speed for that reason, though that's probably a stretch. The obvious purpose, though is product differentiation. Clocking it at the same speed as the IIci would have cannibalized sales of the IIci. Doesn't seem that unreasonable to me.