And of course some posters on Slashdot ignore the fact that the US Dept of Commerce exercises veto power over ICANN decisions. This has been reported on/. several times, but is conveniently forgotten whenever the current topic comes up.
Hasn't been abused... we think... yet.
Look at the.xxx tld and the commerce departments efforts to delay/block it. Although it was approved, ICANN officially indefinitely delayed the implementation to works out the implementation and policy issues for the tld. I think the US government has already influenced ICANN's decisions significantly.
DDT was highly effective at killing mosquitos, the incidence rate of malaria dropped amazingly after it was put in use. However, someone had the idea that saving a few Peregrine Falcons was more important than tens of thousands of human lives. Too bad.
DDT resistant mosquitos appeared in 1960 and have spread pretty much everywhere. Using more DDT doesn't work since the mosquitos become more and more resistant due to overexpression of cytochrome P450. Meanwhile, things like fish, birds and people who happen to eat those fish or birds get increasing concentrations of DDT and eventually get poisoned or start seeing birth defects.
Unfortunately, we don't go through a few generations every few months and can't quickly develop DDT resistance like mosquitos. The falcons were just an indicator and continuing would have increased the incidence of birth defects in people.
The errors in/. are beside the point; they're obviously not caused by any bugs in the language, but in the code itself. As for caching, if it's transparent to the user, than what's the difference? I don't think that Amazon or a large site such as that is going to get by without some form of caching.
The errors are exactly the point. The errors appear because the slashdot code has to do a lot of caching and other tricks to reduce system load and handle the traffic in a reasonable manner.
Other frameworks that can scale better avoid this by either having better integration/performance so you can add servers and handle the load without resorting to dangerous/complex optimizations or by implementing these optimizations in the framework so they they well tested and debugged as opposed to an ad-hoc solution.
It wasn't compiled. It was hand written in machine code. When I first learned machine code and we had to write things in machine code on the little virtual machines we had I was starting to think of assembly as an easy to use language. Heh. That is when you know you are at the bottom of the barrel;) (It was actually pretty fun).
Now a days, I'm pretty sure the initial compilers are generated using a cross compiler. E.g. you get your gcc compiler on x86 to use the right instruction set and then cross compile gcc for the new architecture. You get a fully featured compiler with minimal work rather than handwriting a basic compiler that you can then use to compile a normal compiler.
n India it is not unheard of that doctors will take organs from persons or bodies, such as valves in the case of bodies, for transplantation to 'customers' who need them.
The deceased may never have signed a donor form, and the family is not informed.
Neither is the recipient - they simply aren't told what type of valve they're getting (artificial being the common assumption).
To paraphrase a statement from one hospital CEO/doctor : "We open them up, take out the valves, sow them back up, and no harm is done. The body gets cremated and nobody will ever know."
That sounds a lot like urban rumors. You don't simply open up a body and harvest organs. Most organs such as kidneys, hearts, etc. require that the donor get prepped before hand while the donor is still alive or within a few minutes of death. The organs need to be matched and harvested within a few minutes of death otherwise the organs are damaged and become useless.
Unless the doctors are systematically matching potential donors on a massive scale, they simply won't have a viable donor ready for potential recipients. Remember, most transplants need to occur within a few hours of harvesting and donors have to match a lot of criteria (MHC, blood types must match; donor can't have diseases such as aids, hiv, colds, infections, donor organ must be free of congenital defects, previous damage, etc.).
There was an episode on Discovery's Animal Face-off [discovery.com] about a Giant Squid versus a Sperm Whale: The winner was the sperm whale, which stunned the squid with its sonic emitter, and then ate it whole.
Given the sperm whale weighs 30-50 tons and the squid weighs a few thousand pounds, I think it's pretty much foregone that the squid loses since it's not poisonous.
A full grown martial artist weighs perhaps 150 pounds, while a fat slashdotter would weight in somewhere between 300-400 pounds. I think, the occasional bruise aside, the slashdotter wins. Every time.
Let me correct that a bit:
A full grown martial artist weighs perhaps 150 pounds, while a slashdotter would weight in somewhere between 7000-15000 pounds. Both have rough the same proportional levels of muscle. I think, the occasional bruise aside, the slashdotter wins. Every time.
There were *MORE* options for chips back then, not less. MIPS. Alpha. Power. 680x0. AMD 486. Cx 486.
In the consumer market you had a choice between Intel and it's clones (Cyrix, IDT, AMD). By the time the Pentium and PII processors were around only Intel offered reasonable performance. Alpha, power, mips weren't an option for consumer systems. The 680x0 systems were only used for apple systems.
The situation now is about the same, for consumer pcs you have a choice between power and intel/amd. The difference is that the amd cpus are price and performance competitive with intel chips.
However my original comments were about innovations and intel's in the pc market.
Intel has pretty much always "pushed it". Their projected timelines for Mhz improvement has been fairly trackable, and fairly steady since the 1970s.
Look at this graph and you'll notice that transisitor counts and processor introductions started leveling off after the 486 intro and didn't pick up until just after the P3 intro when the athlon came out and started putting up performance numbers that matched or beat the P3 processors.
Are Photonic Processors the next logical step, or will the almighty buck shuffle them aside because of cost?
If photonic processors go into widespread usage, it will probably be because of the almighty buck and companies deciding that they can make more of it by producing photonic processors.
Profits and competition are the main reason for a lot of the recent advances in processor performance. Look at the processor introductions back when 486 and pentium processors were around and Intel didn't have any credible competition.
Let's use my example of MS again.
They gave us IE.
It had bugs, so they patched it.
It had more bugs so they patched it.
It had even more bugs so they patched it some more.
Now they patch it about once a month.
Is a layman able to say "Gee, that must be bad code", or is it a case of where this would happen to any browser?
Can you or I judge? Or is it a case of only peer reviewed code by another computer professional is the only fair way to judge?
It's closer to someone telling the MS that a website is causing IE to install spyware. MS then says try this version of IE, we've put in stack canaries to detect and gracefully exit after buffer overflows. The person says nope that didn't work so MS says alright this version also has randomization of the stack addresses and no-execute bits on data pages.
Now, do you expect the average person to be able to be able judge whether MS's actions were correct ?
ABBYY Finereader does an excellent job. On more difficult pages I had to do some tweaking, but part of that was my inexperience with the product.
My experience was that OCR was not up to the task 2 years ago. However this was with notes and papers where up to 50% of the page was mathematics. Once you start seeing some of the more esoteric or specialized mathematical symbols, I think the OCR just breaks down.
However, even with 95% accuracy on math symbols that would leave a lot of pages to be reviewed for small errors that might significantly alter the meaning. For example, the visual difference difference between a lowercase chi and x or a lowercase nu and v is fairly small but can make a huge difference if they get transposed.
I've helped setup something like this. The best small scale solution would be to get a good flatbed scanner with an automatic document feeder (ADF). You can get decent HP scanners for about $400-700.
Once you have the scanner, you can setup a few scanning profiles that automatically set resolution, color depth, black&white threshold, etc. Then scan the notes into adobe as images. If you scan them in as monochrome images at 100-150dpi you can get fairly small files that are very readable on screen and as printouts.
Finally get a RA or student labor to feed the documents in to adobe and save them in separate files. The adf lets you do 25-100 sheets at a time so the help and start a scan and surf the web or something until the batch is done.
N.B.: Having a flatbed scanner lets you handle odd sized sheets of paper or delicate stuff. Although you can scan and ocr the documents, ocr is probably going to screw things up a bit and you probably don't want to try to read through the documents to catch and correct the ocr errors. Also if you have any math, diagrams, or handwriting in the notes, the ocr program will probably produce unusable junk.
If one person holds a dirty 20 (cm) mirror in pitch darkness about a 100 meters away from me, another person holds a shiny 10 (cm) mirror about 200 meters away from me, and I shine a _powerful_ flashlight at them, I see the smaller mirror is brighter, yet it's NOT bigger, and both appear the same size relative to me.
At the distances the planets are from us, both objects look like specks. They will probably be larger on something like Gemini but there won't be a difference in sizes due to the distance.
In addition, we already know how far away the objects are due to measurements of it's position. E.g. once you get a few observations you can plug that information into Kepler's equations and get an orbit from it. Once we know that the object is further away from pluto and still brighter, we can figure out that the object has to be either larger than pluto or a perfect mirror. One of the pages gives various size estimates based on reflectivity.
St. Louis, Louisville, Cincinnati, and Pittsburg would not change location if the branch of the delta being used by the river changed. Just New Orleans.
The barges being used to ship goods to and from those cities would need to get a new port however and that would get expensive after a few relocations.
Interesting - you'd think that in situations like this, the BIOS could mediate the startup process such that everything wasn't hammering the PSU all at once. I wondered if this is what was happening on occasion, since my CPU would start up, but the drive would power up a few seconds afterward.
Scsi cards have had this for at least 5 years. You can set them to delay spin up for 2*scsi id , so the drive with id 1 spins up at 2s, the drive with id 2 spins up after 4 s, etc. It's useful when a scsi chain can have up to 14 drives.
*Rojak, along with Satay, Roti Canai, Teh Tarik, etc are ALL Malaysian dishes, despite what the lying Singaporeans have been claiming.
With Roti Canai at least, the name of the dish is straight from Hindi. The few times I eaten it, it basically resembles Indian/South Asia food so I'm fairly sure that it's Indian in origin.
That makes no sense. Just make it have a register "frequency" and have it reject out of band stuff ON DIE.
Different regions have different frequencies that are out of bounds. For example, Canada and the US allow different frequencies making certain channels out of bounds. Similar restrictions apply to the EU and Asia.
Hardware vendors don't want to have different cards/chipsets for each country so they set frequencies in firmware/software.
The feds threatened to arrest the Americans who defaced Chinese websites after the PLA-Air Force brough our AWAC down early in Bush's first term.
I am not familiar with this incident... could you please elaborate?
Look at this. An us plane monitoring chinese communications had an "incident" with a chinese fighter that was intercepting it. The plane was forced to land at a chinese airbase and the crew was detained for a while. I'm not sure if the plane was returned.
Re:FreeBSD
on
Why FreeBSD
·
· Score: 2, Insightful
Well, a package format will often include as part of its specs a declaration of what info will be included in the package. RPM DOES NOT, as part of the spec, include package dependencies. There are ways to track deps in RPM land, but NO RPMs do not include dependency tracking as a base feature, last time I checked.
You should check again, the rpm specs and packages have had a requires section for dependencies since the very beginning. More recent versions have also included a build-req field to allow developers to specify what is needed to rebuild the rpm from source.
Unfortunately, this will no longer work. The FCC regs have been changed to allow for a 'blanket' license for specific multiple uses such as radar guns.
But that doesn't mean you folks can't give it a try. You may be able to intimidate the cop enough for him to not write the ticket.
You don't want to do this. Although you can try, you might end up guilty of perjury if you tell the judge something that is untrue. A perjury conviction is a lot worse than a speeding ticket.
Although I dislike the use of the word hacking in this context it technically isn't wrong to use it. One definition of hacking is "In a similar vein, a "hack" may refer to works outside of computer programming. For example, a math hack means a clever solution to a mathematical problem. The GNU General Public License has been described as a copyright hack because it cleverly uses the copyright laws for a purpose the lawmakers did not foresee."
The term hack is pretty much limited to a small subculture (i.e. certain groups of programmers). Although you can describe an elegant solution to a math problem as a math hack, a mathematician probably wouldn't know what you're talking about.
I think your prediction of hack and hacking gaining widespread use is overly optimistic. It will probably remain within the programming subculture and won't spread. Most terms and slang from subcultures share the same fate.
Your point of view would make sense, except the ending to that thing is quite bizarre. I think the creator really had something in mind, but it is unclear what that is.....
From my understanding, the producers ran out of money for the ending and so tossed to together stuff they were able to afford in order to have a final episode at all. The way the two movies tried come up with an ending but failed suggests that the creator didn't really have a clear idea of what he wanted and was adding things because they sounded cool earlier in the series.
Slashdot Mirror
Look at the .xxx tld and the commerce departments efforts to delay/block it. Although it was approved, ICANN officially indefinitely delayed the implementation to works out the implementation and policy issues for the tld. I think the US government has already influenced ICANN's decisions significantly.
DDT resistant mosquitos appeared in 1960 and have spread pretty much everywhere. Using more DDT doesn't work since the mosquitos become more and more resistant due to overexpression of cytochrome P450. Meanwhile, things like fish, birds and people who happen to eat those fish or birds get increasing concentrations of DDT and eventually get poisoned or start seeing birth defects.
Unfortunately, we don't go through a few generations every few months and can't quickly develop DDT resistance like mosquitos. The falcons were just an indicator and continuing would have increased the incidence of birth defects in people.
The errors are exactly the point. The errors appear because the slashdot code has to do a lot of caching and other tricks to reduce system load and handle the traffic in a reasonable manner.
Other frameworks that can scale better avoid this by either having better integration/performance so you can add servers and handle the load without resorting to dangerous/complex optimizations or by implementing these optimizations in the framework so they they well tested and debugged as opposed to an ad-hoc solution.
Now a days, I'm pretty sure the initial compilers are generated using a cross compiler. E.g. you get your gcc compiler on x86 to use the right instruction set and then cross compile gcc for the new architecture. You get a fully featured compiler with minimal work rather than handwriting a basic compiler that you can then use to compile a normal compiler.
Could it be a first post?
Because they can be pulling down 150,000-250,000 as a airline pilot as opposed to 80,000-125,000 as a pilot in the military?
That sounds a lot like urban rumors. You don't simply open up a body and harvest organs. Most organs such as kidneys, hearts, etc. require that the donor get prepped before hand while the donor is still alive or within a few minutes of death. The organs need to be matched and harvested within a few minutes of death otherwise the organs are damaged and become useless.
Unless the doctors are systematically matching potential donors on a massive scale, they simply won't have a viable donor ready for potential recipients. Remember, most transplants need to occur within a few hours of harvesting and donors have to match a lot of criteria (MHC, blood types must match; donor can't have diseases such as aids, hiv, colds, infections, donor organ must be free of congenital defects, previous damage, etc.).
Given the sperm whale weighs 30-50 tons and the squid weighs a few thousand pounds, I think it's pretty much foregone that the squid loses since it's not poisonous.
Let me correct that a bit:
A full grown martial artist weighs perhaps 150 pounds, while a slashdotter would weight in somewhere between 7000-15000 pounds. Both have rough the same proportional levels of muscle. I think, the occasional bruise aside, the slashdotter wins. Every time.
In the consumer market you had a choice between Intel and it's clones (Cyrix, IDT, AMD). By the time the Pentium and PII processors were around only Intel offered reasonable performance. Alpha, power, mips weren't an option for consumer systems. The 680x0 systems were only used for apple systems.
The situation now is about the same, for consumer pcs you have a choice between power and intel/amd. The difference is that the amd cpus are price and performance competitive with intel chips.
However my original comments were about innovations and intel's in the pc market.
Look at this graph and you'll notice that transisitor counts and processor introductions started leveling off after the 486 intro and didn't pick up until just after the P3 intro when the athlon came out and started putting up performance numbers that matched or beat the P3 processors.
If photonic processors go into widespread usage, it will probably be because of the almighty buck and companies deciding that they can make more of it by producing photonic processors.
Profits and competition are the main reason for a lot of the recent advances in processor performance. Look at the processor introductions back when 486 and pentium processors were around and Intel didn't have any credible competition.
It's closer to someone telling the MS that a website is causing IE to install spyware. MS then says try this version of IE, we've put in stack canaries to detect and gracefully exit after buffer overflows. The person says nope that didn't work so MS says alright this version also has randomization of the stack addresses and no-execute bits on data pages.
Now, do you expect the average person to be able to be able judge whether MS's actions were correct ?
My experience was that OCR was not up to the task 2 years ago. However this was with notes and papers where up to 50% of the page was mathematics. Once you start seeing some of the more esoteric or specialized mathematical symbols, I think the OCR just breaks down.
However, even with 95% accuracy on math symbols that would leave a lot of pages to be reviewed for small errors that might significantly alter the meaning. For example, the visual difference difference between a lowercase chi and x or a lowercase nu and v is fairly small but can make a huge difference if they get transposed.
I've helped setup something like this. The best small scale solution would be to get a good flatbed scanner with an automatic document feeder (ADF). You can get decent HP scanners for about $400-700.
Once you have the scanner, you can setup a few scanning profiles that automatically set resolution, color depth, black&white threshold, etc. Then scan the notes into adobe as images. If you scan them in as monochrome images at 100-150dpi you can get fairly small files that are very readable on screen and as printouts.
Finally get a RA or student labor to feed the documents in to adobe and save them in separate files. The adf lets you do 25-100 sheets at a time so the help and start a scan and surf the web or something until the batch is done.
N.B.: Having a flatbed scanner lets you handle odd sized sheets of paper or delicate stuff. Although you can scan and ocr the documents, ocr is probably going to screw things up a bit and you probably don't want to try to read through the documents to catch and correct the ocr errors. Also if you have any math, diagrams, or handwriting in the notes, the ocr program will probably produce unusable junk.
At the distances the planets are from us, both objects look like specks. They will probably be larger on something like Gemini but there won't be a difference in sizes due to the distance.
In addition, we already know how far away the objects are due to measurements of it's position. E.g. once you get a few observations you can plug that information into Kepler's equations and get an orbit from it. Once we know that the object is further away from pluto and still brighter, we can figure out that the object has to be either larger than pluto or a perfect mirror. One of the pages gives various size estimates based on reflectivity.
The barges being used to ship goods to and from those cities would need to get a new port however and that would get expensive after a few relocations.
Scsi cards have had this for at least 5 years. You can set them to delay spin up for 2*scsi id , so the drive with id 1 spins up at 2s, the drive with id 2 spins up after 4 s, etc. It's useful when a scsi chain can have up to 14 drives.
Different regions have different frequencies that are out of bounds. For example, Canada and the US allow different frequencies making certain channels out of bounds. Similar restrictions apply to the EU and Asia.
Hardware vendors don't want to have different cards/chipsets for each country so they set frequencies in firmware/software.
Look at this. An us plane monitoring chinese communications had an "incident" with a chinese fighter that was intercepting it. The plane was forced to land at a chinese airbase and the crew was detained for a while. I'm not sure if the plane was returned.
You should check again, the rpm specs and packages have had a requires section for dependencies since the very beginning. More recent versions have also included a build-req field to allow developers to specify what is needed to rebuild the rpm from source.
You don't want to do this. Although you can try, you might end up guilty of perjury if you tell the judge something that is untrue. A perjury conviction is a lot worse than a speeding ticket.
The term hack is pretty much limited to a small subculture (i.e. certain groups of programmers). Although you can describe an elegant solution to a math problem as a math hack, a mathematician probably wouldn't know what you're talking about.
I think your prediction of hack and hacking gaining widespread use is overly optimistic. It will probably remain within the programming subculture and won't spread. Most terms and slang from subcultures share the same fate.
Why yes I did. The shuttle has a single pilot and a commander. The other crew members are mission specialists and just sit back for the ride.
From my understanding, the producers ran out of money for the ending and so tossed to together stuff they were able to afford in order to have a final episode at all. The way the two movies tried come up with an ending but failed suggests that the creator didn't really have a clear idea of what he wanted and was adding things because they sounded cool earlier in the series.