perhaps that bactiria [sic] prefers the chemical balance in a bleeding stomach
I believe that h. pylori excretes urease which breaks urea (found naturally in tissues as a byproduct of metabolism) down into ammonia. Ammonia is much more basic than urea, and as a result it reacts with stomach acid to create a more neutral pH, which is better for the organism. Ammonia is also very toxic, which is why the body converts it to urea in the first place (the direct product of metabolism is ammonia, and it is quickly converted to urea to detoxify it before being sent to the kidneys - later it gets partially converted to uric acid which is even less toxic and also requires less water to store/flush (this is what makes bird droppings white - birds do this to the extreme to avoid carrying lots of water in their urine as dead weight when flying)). So, the ammonia is the direct cause of the tissue damage that leads to an ulcer.
The bacteria doesn't really care about blood/pain/holes in stomach - it just wants a break from all that HCl floating around. As long as it nestles in the folds of the tissue it can create a locally moderate environment to grow in.
So, in this case the disease is more a side-effect than a direct goal of the bacteria.
On the other hand, if you look at something like cholera toxin that is something really nasty. Cholera spreads from feces, and its toxin basically makes you run like a faucet (even today it can be quite fatal if not caught early enough - IVs can't fill you up faster than you spill out).
However, you are dead-on that many times disease organisms have the goal of modifying host behavior to encourage spread.
Hmm. If they don't actually distribute linux then they don't need to accept the license. However, this is not likely since I doubt many embedded hardware manufacturers will tell their users to do their own OS installs. As soon as they start imaging hard drives they have to accept the license.
I can't come up with an easy solution to the problem you brought up, though...
Simple - the GPL is a license. You only need a license to do something that would otherwise be illegal - like distributing the copyrighted linux kernel.
Alan hasn't distributed any hardware that requires his signature to a kernel. So, he hasn't violated the GPL.
BadCorp has - they released a version of the kernel which can only be run with Alan's signature. They can only do so if they supply Alan's private key. Note that Alan is under no obligation to provide his key - only BadCorp is. So, unless Alan cooperates with them, they are up the creek.
Obviously this could be hashed out in about 40 different sets of particular circumstances, but I'm guessing the solution will be something along those lines. If Alan isn't doing anything, he doesn't need a license to not do it...
If that were true there would be a simple solution.
You could form a corporation and bid on the cure, develop itself, and sell it for a small fraction of what modern drugs cost. After all, if pharma companies don't have any real expenses then you would easily outcompete them.
The reality is that most purchased leads don't amount to anything, and the ones that do involve substantial R&D costs in addition to whatever the NIH did. Most modern drugs are discovered internally (or from commercial biotechs). At most, the NIH typically just comes up with a concept of some sort.
If it were as easy as you suggest there wouldn't be high drug prices...
It's possible that a governmental body might try to do an end run around the patent, if they decide that AIDS is "epidemic enough" for that sort of thing.
This will likely be the case.
It is probably the reason why most major pharma companies only make half-hearted attempts to discover AIDS medications. After all, the folks who need them most can't afford them, and live in countries that won't pay for them. And, with the huge activist community laws are likely to be slanted against AIDS patent rights.
My guess is that if AIDS is cured it will either be because somebody offers a bounty (of a billion dollars or some crazy figure), or a government agency funds it completely...
Simple. An enterprise DBMS is a HUGE purchase, often going into the 7-figure range if for a big company, but often in the 5-6 figure range for a small company.
Expenses of that scale involve managers. At most companies the IT managers often do not have tons of development experience, and instead tend to have project-management experience and business-side experience. This is how they are chosen by business-side executives to head up IT.
These folks can't show weakness, and so they're going to tend to leave the underlings out of the decisions.
This is certainly not universally true, but it happens often enough to line the pockets of many a salesman...
Actually, math is completely self-consistent given a set of axioms.
The problem is that no set of axioms is sufficient to close the system, as it were. That is, there will always be statements framed in any logical system with a given set of axioms in which it is impossible to prove or disprove their truth.
So, the problem until recently wasn't that somebody had both a proof and a contradiction of Fermat's Last Theorem - it was that we had neither. Granted, this is no longer a good example, but there are many others, and some of them may in fact be unprovable without the addition of axioms, but there is no way to know whether this is the case (at least I don't think there is - perhaps someone more learned than I could indicate that whether Godel allows for the proof of whether or not a statement can be proven, although I guess that would be a statement in itself which might also happen to be unprovable)...
I'm sure Windows will be with us for a long time, but I'm also pretty sure that.NET won't.
Remember COM+, ActiveX, etc.? Every 3-4 years Microsoft comes out with their latest interfaces, buzzwords, etc. In a few years MS will be moving from Visual Fred to Visual Jake, and everybody will be doing backflips to migrate their legacy code.
Is it time to retire some of those COBOL/CSIS mainframes? Sure.
Do we need to rewrite every application we own just because it is more than three years old? No...
A lot of shops still have VB6 sitting around because of the large number of difficult-to-port applications. How many people have GCC v2 lying around for hard-to-compile C apps? Almost none, since the GNU folks are half-decent about backwards compatibility in their development tools. When things break it tends to be minor - as it should be for a programming language.
The bottom line is that programmers shouldn't have to jump through hoops every time MS wants to sell more development stuido licenses, or needs to attract media attention...
Don't forget to factor in backup, redundancy at an offsite datacenter, staffing, etc.
No doubt many companies are getting ripped off, but most likely the guy who made the decision on what products to buy is doing quite well. So, tell me exactly why they should have done anything differently? The first step to good decisions is to punish those who make bad decisions. And that has to start with executive management...
Of course, the cost per GB of many managed SANs ends up running into the $100-200 range (per GB!). This is what leads to 50MB quotas on employee home directories. No, a well-managed SAN doesn't need to cost that much, but that is how much they cost when you buy one from the outfit that will take the CIO out to some nice dinners...
I've found there is a cycle:
1. Company has ancient central IT group populated by dinosaurs and BOFHs. Centrally run servers are underpowered and undermanned.
2. Individual business areas start putting together their own rogue IT groups, and servers spring up all over the company in every closet. The new groups are agile, but not as secure, and the process is not 100% efficient.
3. CIO centralizes the IT groups so that infrastructre is centralized and well-maintained. Costs plummet and service actually increases since the datacenter can be 24x7 monitored unlike the closet down the hall.
4. CIO discovers he can cut his budget 10% without much loss of quality. CIO gets bonus.
5. Repeat step 4 20 times.
6. Goto step 1...
If people stopped at step 3 with the right combination of central infrastructure and business-focused developers/support then you'd probably hit the sweet spot in terms of the best services and cost.
It drives me nuts to see companies that make billions of dollars a year shaving $1000 costs by undersizing server hardware, insufficiently supporting core applications, etc. IT is a force multiplier - every dollar spent in the IT group can enable all kinds of money-making in the business areas - as long as the IT spending goes to the right places. Nickeling and diming IT is like saving money by getting rid of phones at every desk...
Tue May 3 18:12:39 2005 >>> media-video/vlc-0.8.1-r1
merge time: 8 minutes and 24 seconds.
Thu Jan 5 09:49:39 2006 >>> media-video/vlc-0.8.2-r2
merge time: 7 minutes and 46 seconds.
And that is running at a niceness of 10, and optimized -Os, with most of the trimmings USE-wise. I'm not sure what you're running on, but it must be pretty old (this box is going on 2 years old now).
The Fourth Amendment has two clauses - 1. no unreasonable search and seizures, and 2. the warrant clause. Not all searches and seizures require a warrant. For example, searches incident to arrest, "open container" vehicle searches, and (importantly for this discussion) foreign intelligence signals intelligence.
Uh, I don't see the text "arrest," "open container," "vehicle," or "signals intelligence" anywhere in the text of the 4th ammendment.
Obviously there is a need for balance in times of war, but is it really that horrible to ask the president to at least check in with a secret court within 72 hours of starting a tap? The very concept of a secret court in itself is pretty bad, but to not even require that?
Terry stops and pat-downs are reasonable enough since they're limited to looking for obvious signs of weapons while the police stop somebody, and in general police aren't allowed to go digging for admissible evidence. The idea that you can search a vehicle on a whim is completely against the constitution (whether or not it happens to be established precedent).
You either have probable cause or you don't. If you have it, then it is a simple matter to obtain a warrant. If you're hunting a guy who plants bombs in trunks then checking trunks for bombs would also be reasonable since there is a compelling need for public safety, but I'd make anything else found in trucks inadmissable unless there was a warrant.
Just think about why the 4th ammendment was actually written, and perhaps the need for it will become more clear. It was written by folks who actually used to be considered terrorists by the powers in charge, who were on a quest to flush them out by any means necessary. That isn't to condone terrorism, but if somebody really is a public threat it won't be hard to get a warrant to search their belongings...
Re:Unplayable here (was: Re:dont wanna stream?)
on
IT Crowd On-line
·
· Score: 2, Informative
I just install the 64-bit version, and then run a chroot that is purely 32-bit. I don't have much installed there - mainly closed-source stuff, open-source stuff like openoffice which barely builds cleanly 32-bit let alone 64-bit, and semi-open stuff like java which also doesn't do all that well on 64-bit (hello.java works fine, but good luck getting freenet/eclipse/netbeans/etc to work reliably...). Best of both worlds, although my install is using an extra few GB as a result...
Reconciling various theoretical inconsistencies and paradoxes in our current theories is an accomplishment of no small importance, if only because no other theory (predictive or not) has been able to do as much.
Well, if that is the only goal, what's wrong with the "theory" "because God made it that way." That does technically accomplish just as much, and probably is as useful scientifically.
The problem is that if you don't make predictions that can be tested, it really isn't even a theory at all. It is more of a philosophy (like math, or a million other things that aren't concretely tied into the real world). The reason science is so well-accepted is because it is a systematic way of finding out truth that we know pertains to the real world. If we want good speculation and nothing more than I daresay that any number of philosopers/theologians/etc have as much to contribute as what most string theorists have given us.
Not that string theory should just be shunned, but its value is limited at present unless it can actually tell us something meaningful about our universe that we don't already know, and which can be tested. Or at least if it can explain something we already know in a way that no other theory can, with some ability to extrapolate beyond that.
I wouldn't be surpised if string theory eventually develops into something of practical usefulness, but I don't think that day has arrived yet...
Only if you plan on 100% counting the boxes - and not having the machine do any tallying at all. That wasn't the proposal as I read it.
The problem with reading paper ballots is that it is less than 100% accurate and probably would require manual verification. If you want a purely manual election that is fine - the computers would still help by validating the ballots so that you don't have two people marked for one office.
However, if you want the voting machine to keep a running tally and simply audit the boxes, then you can't let the voter touch their ballot. Otherwise some might change their mind before getting to the box and just pocket their ballot. Then you end up with a mismatch between the box and the machine. Sure, the box would override, but the discrepancy would cause audits, and if you're going to 100% audit all the time anyway, why bother having the machine keep a tally? Also - the low-level noise in the ballot counts will tend to mask low levels of fraud (such as adding 100 votes to every precient in the state - it would probably disappear in the noise of people walking out with ballots but could add up to a substantial number of votes).
I'd just have the paper trail behind glass, and make the printer nice and noisy so that party reps would notice if it just started printing votes without somebody in the booth (or make the mechanism dependant on somebody pulling a lever to get the vote in the box).
One little change - the voter wouldn't be able to touch the paper trail - it would be behind glass. Otherwise folks would walk off with them, and in the event of an audit the counts wouldn't match - thus making it impossible to detect true fraud.
Have the paper drop into a box after verification - otherwise it goes in a trash box. Do a random audit of x% of the results, and a systematic audit any time there is cause to do so...
It seems odd that investigators would be so lacadaisical about clinical data. I guess it is because they are independant from the big pharma companies. If a pharma outfit handled their manufacturing and QA testing data in this manner they'd be shut down in a hurry. These operations are heavily regulated in the US, Europe, and Japan - typically any time a number is copied from one place to another (by paper or by typing) somebody has to sign off that they double-checked the entry. I've heard the figure for transcription errors is on the order of 1 in 25 without careful checking.
There are numerous IT systems designed to automate this process - one person enters data, another person checks it and marks it as having been reviewed. An alternative system is double-data-entry - one person types the data, another person types the data again (without seeing what the first entry was), and the comptuter flags for review any entries which differ. Sure, it cuts down productivity, but we're talking about human life here.
Drug development costs a fortune. My feeling is that a BIG part of this cost is the payments to investigators to bring in test subjects - these guys are MDs, and they command hefty fees for their work. I'm not convinced they always give good value for their work in return. If you look at the FDA debarment list you'll find that many of the people listed are in fact investigators. (This is a list of people forbidden to work in any way with data that will be eventually submitted to the FDA or with pharmaceuticals in general.)
My guess is that they slowly depressurize the cabin at a steady time-based rate as you ascend. If you don't have pressurization you depressurize at a height-based rate instead.
Since vertical velocity isn't constant, you'd normally have a period at the start and end of the run which was comfortable, with a painful cruise period in the middle. With a pressurized cabin they can let the air out slowly during the cruise period, and get caught up while the elevator is finished decellerating at the end of the trip to a smooth stop.
That's just a guess though - I'm no elevator expert...
A person who has an asthma attack while driving or suddenly faints for no apparent reason and runs over and kills 5 people is charged with 3rd-degree murder. They could end up in prison.
A person who carefully plans the assassination of their next door neighbor but botches the job is charged with attempted murder. They could end up in prison, and it would depend to some degree on how successful they were.
Clearly the second person is a danger to society. The first person might or might not be - and the harshest penalty really called for would be to deny them a driver's license if their condition were expected to recur. However, in our modern justice system the two crimes are not very different in penalty.
In my mind, the apparent intentions of the criminal should be one of the most important criteria in sentencing. Punishing people whose only crime is being unlucky and giving an easy sentence to somebody whose only virtue is being lucky is completely at odds with the whole notion of justice and the protection of society...
OK, I admit this was a bit off-topic, but it annoys me to no end to see these kinds of laws on the books...
I assume that ac3 compression is smart enough to take advantage of cross-tack similarities? Your 5.1 soundtrack isn't 5 completely isolated audio streams from 5 completely different movie soundtracks. Sure, when there is a surround effect happening they probably diverge a fair amount, but most of the time the surround speakers are just playing sound track - which is mostly the same across all speakers with a few components louder/quieter and maybe a little phase difference. It wouldn't surprise me if the actual amount of loss is all that high as a result.
Somebody should come up with a FLAC-like algorithm for 5.1 surround that takes advantage of this fact. While I'm sure the resulting tracks would be substantially bigger, I doubt they'd require 7-8GB...
Also CSS was "cracked" more because of XING's goof
Uh, it was hardly a goof...
It was inevitable once they allowed software decoders. If you decode in software then the software has to have both the key, and the algorigthm. You can obfuscate it all you want, but it is still there.
Even with hardware you could eventually obtain it, since the key is stored in the hardware somewhere.
Something a little more secure would be sending the encrypted session key online to some 3rd party, who checks your credentials and sends you back the decoded key - if you intercept it you can only crack that single disk. (This is how many itunes crackers work.) However, if if the player displays the movie, then at some point it must have the key.
However, in software it is MUCH easier to access - unless you have trusted computing (then you'd need the appropriate key out of the BIOS just to read the software code off the hard drive).
Except that if you were on an elevator E travelling at the speed of light realtive to a object A, you'd have to travel an infinite distance relative to A in order for any time to pass in E.
You can't travel at the speed of light relative to object A as measured by either E or A. You could travel at the speed of light relative to object A as measured by object B (somebody halfway in between noting both E and A moving in opposite directions at 0.5c), but A and E would measure their speed as being nearly the speed of light but not quite. Relative to object B, time is still ticking fairly normally for both A and E, though relative to each other each sees the other as nearly-frozen (though everything would appear normal to themselves).
The person in the elevator would not see anything odd about travelling at such a high speed. This is logical since from their frame of reference they are standing still, and everything else is flying past them at nearly the speed of light - time is frozen for everything else, and not them by their measurement. Time is not absolute.
We might actually be on the same wavelength here - maybe not. Hard to tell since it is impossible to be clear without qualifying every statement with who is moving where relative to what and who is taking the measurements...
Slashdot Mirror
You might as well be marketing OS-less systems rather than Linux systems.
And why not do this? It certainly doesn't cost more to NOT image the hard drive, right? Unless you have to pay for Windows either way...
perhaps that bactiria [sic] prefers the chemical balance in a bleeding stomach
I believe that h. pylori excretes urease which breaks urea (found naturally in tissues as a byproduct of metabolism) down into ammonia. Ammonia is much more basic than urea, and as a result it reacts with stomach acid to create a more neutral pH, which is better for the organism. Ammonia is also very toxic, which is why the body converts it to urea in the first place (the direct product of metabolism is ammonia, and it is quickly converted to urea to detoxify it before being sent to the kidneys - later it gets partially converted to uric acid which is even less toxic and also requires less water to store/flush (this is what makes bird droppings white - birds do this to the extreme to avoid carrying lots of water in their urine as dead weight when flying)). So, the ammonia is the direct cause of the tissue damage that leads to an ulcer.
The bacteria doesn't really care about blood/pain/holes in stomach - it just wants a break from all that HCl floating around. As long as it nestles in the folds of the tissue it can create a locally moderate environment to grow in.
So, in this case the disease is more a side-effect than a direct goal of the bacteria.
On the other hand, if you look at something like cholera toxin that is something really nasty. Cholera spreads from feces, and its toxin basically makes you run like a faucet (even today it can be quite fatal if not caught early enough - IVs can't fill you up faster than you spill out).
However, you are dead-on that many times disease organisms have the goal of modifying host behavior to encourage spread.
Hmm. If they don't actually distribute linux then they don't need to accept the license. However, this is not likely since I doubt many embedded hardware manufacturers will tell their users to do their own OS installs. As soon as they start imaging hard drives they have to accept the license.
I can't come up with an easy solution to the problem you brought up, though...
Simple - the GPL is a license. You only need a license to do something that would otherwise be illegal - like distributing the copyrighted linux kernel.
Alan hasn't distributed any hardware that requires his signature to a kernel. So, he hasn't violated the GPL.
BadCorp has - they released a version of the kernel which can only be run with Alan's signature. They can only do so if they supply Alan's private key. Note that Alan is under no obligation to provide his key - only BadCorp is. So, unless Alan cooperates with them, they are up the creek.
Obviously this could be hashed out in about 40 different sets of particular circumstances, but I'm guessing the solution will be something along those lines. If Alan isn't doing anything, he doesn't need a license to not do it...
If that were true there would be a simple solution.
You could form a corporation and bid on the cure, develop itself, and sell it for a small fraction of what modern drugs cost. After all, if pharma companies don't have any real expenses then you would easily outcompete them.
The reality is that most purchased leads don't amount to anything, and the ones that do involve substantial R&D costs in addition to whatever the NIH did. Most modern drugs are discovered internally (or from commercial biotechs). At most, the NIH typically just comes up with a concept of some sort.
If it were as easy as you suggest there wouldn't be high drug prices...
It's possible that a governmental body might try to do an end run around the patent, if they decide that AIDS is "epidemic enough" for that sort of thing.
This will likely be the case.
It is probably the reason why most major pharma companies only make half-hearted attempts to discover AIDS medications. After all, the folks who need them most can't afford them, and live in countries that won't pay for them. And, with the huge activist community laws are likely to be slanted against AIDS patent rights.
My guess is that if AIDS is cured it will either be because somebody offers a bounty (of a billion dollars or some crazy figure), or a government agency funds it completely...
Simple. An enterprise DBMS is a HUGE purchase, often going into the 7-figure range if for a big company, but often in the 5-6 figure range for a small company.
Expenses of that scale involve managers. At most companies the IT managers often do not have tons of development experience, and instead tend to have project-management experience and business-side experience. This is how they are chosen by business-side executives to head up IT.
These folks can't show weakness, and so they're going to tend to leave the underlings out of the decisions.
This is certainly not universally true, but it happens often enough to line the pockets of many a salesman...
Actually, math is completely self-consistent given a set of axioms.
The problem is that no set of axioms is sufficient to close the system, as it were. That is, there will always be statements framed in any logical system with a given set of axioms in which it is impossible to prove or disprove their truth.
So, the problem until recently wasn't that somebody had both a proof and a contradiction of Fermat's Last Theorem - it was that we had neither. Granted, this is no longer a good example, but there are many others, and some of them may in fact be unprovable without the addition of axioms, but there is no way to know whether this is the case (at least I don't think there is - perhaps someone more learned than I could indicate that whether Godel allows for the proof of whether or not a statement can be proven, although I guess that would be a statement in itself which might also happen to be unprovable)...
I'm sure Windows will be with us for a long time, but I'm also pretty sure that .NET won't.
Remember COM+, ActiveX, etc.? Every 3-4 years Microsoft comes out with their latest interfaces, buzzwords, etc. In a few years MS will be moving from Visual Fred to Visual Jake, and everybody will be doing backflips to migrate their legacy code.
Is it time to retire some of those COBOL/CSIS mainframes? Sure.
Do we need to rewrite every application we own just because it is more than three years old? No...
A lot of shops still have VB6 sitting around because of the large number of difficult-to-port applications. How many people have GCC v2 lying around for hard-to-compile C apps? Almost none, since the GNU folks are half-decent about backwards compatibility in their development tools. When things break it tends to be minor - as it should be for a programming language.
The bottom line is that programmers shouldn't have to jump through hoops every time MS wants to sell more development stuido licenses, or needs to attract media attention...
Don't forget to factor in backup, redundancy at an offsite datacenter, staffing, etc.
No doubt many companies are getting ripped off, but most likely the guy who made the decision on what products to buy is doing quite well. So, tell me exactly why they should have done anything differently? The first step to good decisions is to punish those who make bad decisions. And that has to start with executive management...
Of course, the cost per GB of many managed SANs ends up running into the $100-200 range (per GB!). This is what leads to 50MB quotas on employee home directories. No, a well-managed SAN doesn't need to cost that much, but that is how much they cost when you buy one from the outfit that will take the CIO out to some nice dinners...
I've found there is a cycle:
1. Company has ancient central IT group populated by dinosaurs and BOFHs. Centrally run servers are underpowered and undermanned.
2. Individual business areas start putting together their own rogue IT groups, and servers spring up all over the company in every closet. The new groups are agile, but not as secure, and the process is not 100% efficient.
3. CIO centralizes the IT groups so that infrastructre is centralized and well-maintained. Costs plummet and service actually increases since the datacenter can be 24x7 monitored unlike the closet down the hall.
4. CIO discovers he can cut his budget 10% without much loss of quality. CIO gets bonus.
5. Repeat step 4 20 times.
6. Goto step 1...
If people stopped at step 3 with the right combination of central infrastructure and business-focused developers/support then you'd probably hit the sweet spot in terms of the best services and cost.
It drives me nuts to see companies that make billions of dollars a year shaving $1000 costs by undersizing server hardware, insufficiently supporting core applications, etc. IT is a force multiplier - every dollar spent in the IT group can enable all kinds of money-making in the business areas - as long as the IT spending goes to the right places. Nickeling and diming IT is like saving money by getting rid of phones at every desk...
Sounds like a good idea for the next windows worm.
Hey, guys, you gotta check out this new windows theme. It makes your desktop look really cool. To get it working do the following steps:
1. First, send a copy of this email to all of your friends (it won't work unless you do this first).
2.
Sounds like the old Amish virus joke. Soon enough it will be real...
Uh, my numbers are:
# genlop -t vlc
* media-video/vlc
Tue May 3 18:12:39 2005 >>> media-video/vlc-0.8.1-r1
merge time: 8 minutes and 24 seconds.
Thu Jan 5 09:49:39 2006 >>> media-video/vlc-0.8.2-r2
merge time: 7 minutes and 46 seconds.
And that is running at a niceness of 10, and optimized -Os, with most of the trimmings USE-wise. I'm not sure what you're running on, but it must be pretty old (this box is going on 2 years old now).
The Fourth Amendment has two clauses - 1. no unreasonable search and seizures, and 2. the warrant clause. Not all searches and seizures require a warrant. For example, searches incident to arrest, "open container" vehicle searches, and (importantly for this discussion) foreign intelligence signals intelligence.
Uh, I don't see the text "arrest," "open container," "vehicle," or "signals intelligence" anywhere in the text of the 4th ammendment.
Obviously there is a need for balance in times of war, but is it really that horrible to ask the president to at least check in with a secret court within 72 hours of starting a tap? The very concept of a secret court in itself is pretty bad, but to not even require that?
Terry stops and pat-downs are reasonable enough since they're limited to looking for obvious signs of weapons while the police stop somebody, and in general police aren't allowed to go digging for admissible evidence. The idea that you can search a vehicle on a whim is completely against the constitution (whether or not it happens to be established precedent).
You either have probable cause or you don't. If you have it, then it is a simple matter to obtain a warrant. If you're hunting a guy who plants bombs in trunks then checking trunks for bombs would also be reasonable since there is a compelling need for public safety, but I'd make anything else found in trucks inadmissable unless there was a warrant.
Just think about why the 4th ammendment was actually written, and perhaps the need for it will become more clear. It was written by folks who actually used to be considered terrorists by the powers in charge, who were on a quest to flush them out by any means necessary. That isn't to condone terrorism, but if somebody really is a public threat it won't be hard to get a warrant to search their belongings...
I just install the 64-bit version, and then run a chroot that is purely 32-bit. I don't have much installed there - mainly closed-source stuff, open-source stuff like openoffice which barely builds cleanly 32-bit let alone 64-bit, and semi-open stuff like java which also doesn't do all that well on 64-bit (hello.java works fine, but good luck getting freenet/eclipse/netbeans/etc to work reliably...). Best of both worlds, although my install is using an extra few GB as a result...
Reconciling various theoretical inconsistencies and paradoxes in our current theories is an accomplishment of no small importance, if only because no other theory (predictive or not) has been able to do as much.
Well, if that is the only goal, what's wrong with the "theory" "because God made it that way." That does technically accomplish just as much, and probably is as useful scientifically.
The problem is that if you don't make predictions that can be tested, it really isn't even a theory at all. It is more of a philosophy (like math, or a million other things that aren't concretely tied into the real world). The reason science is so well-accepted is because it is a systematic way of finding out truth that we know pertains to the real world. If we want good speculation and nothing more than I daresay that any number of philosopers/theologians/etc have as much to contribute as what most string theorists have given us.
Not that string theory should just be shunned, but its value is limited at present unless it can actually tell us something meaningful about our universe that we don't already know, and which can be tested. Or at least if it can explain something we already know in a way that no other theory can, with some ability to extrapolate beyond that.
I wouldn't be surpised if string theory eventually develops into something of practical usefulness, but I don't think that day has arrived yet...
Only if you plan on 100% counting the boxes - and not having the machine do any tallying at all. That wasn't the proposal as I read it.
The problem with reading paper ballots is that it is less than 100% accurate and probably would require manual verification. If you want a purely manual election that is fine - the computers would still help by validating the ballots so that you don't have two people marked for one office.
However, if you want the voting machine to keep a running tally and simply audit the boxes, then you can't let the voter touch their ballot. Otherwise some might change their mind before getting to the box and just pocket their ballot. Then you end up with a mismatch between the box and the machine. Sure, the box would override, but the discrepancy would cause audits, and if you're going to 100% audit all the time anyway, why bother having the machine keep a tally? Also - the low-level noise in the ballot counts will tend to mask low levels of fraud (such as adding 100 votes to every precient in the state - it would probably disappear in the noise of people walking out with ballots but could add up to a substantial number of votes).
I'd just have the paper trail behind glass, and make the printer nice and noisy so that party reps would notice if it just started printing votes without somebody in the booth (or make the mechanism dependant on somebody pulling a lever to get the vote in the box).
One little change - the voter wouldn't be able to touch the paper trail - it would be behind glass. Otherwise folks would walk off with them, and in the event of an audit the counts wouldn't match - thus making it impossible to detect true fraud.
Have the paper drop into a box after verification - otherwise it goes in a trash box. Do a random audit of x% of the results, and a systematic audit any time there is cause to do so...
It seems odd that investigators would be so lacadaisical about clinical data. I guess it is because they are independant from the big pharma companies. If a pharma outfit handled their manufacturing and QA testing data in this manner they'd be shut down in a hurry. These operations are heavily regulated in the US, Europe, and Japan - typically any time a number is copied from one place to another (by paper or by typing) somebody has to sign off that they double-checked the entry. I've heard the figure for transcription errors is on the order of 1 in 25 without careful checking.
There are numerous IT systems designed to automate this process - one person enters data, another person checks it and marks it as having been reviewed. An alternative system is double-data-entry - one person types the data, another person types the data again (without seeing what the first entry was), and the comptuter flags for review any entries which differ. Sure, it cuts down productivity, but we're talking about human life here.
Drug development costs a fortune. My feeling is that a BIG part of this cost is the payments to investigators to bring in test subjects - these guys are MDs, and they command hefty fees for their work. I'm not convinced they always give good value for their work in return. If you look at the FDA debarment list you'll find that many of the people listed are in fact investigators. (This is a list of people forbidden to work in any way with data that will be eventually submitted to the FDA or with pharmaceuticals in general.)
My guess is that they slowly depressurize the cabin at a steady time-based rate as you ascend. If you don't have pressurization you depressurize at a height-based rate instead.
Since vertical velocity isn't constant, you'd normally have a period at the start and end of the run which was comfortable, with a painful cruise period in the middle. With a pressurized cabin they can let the air out slowly during the cruise period, and get caught up while the elevator is finished decellerating at the end of the trip to a smooth stop.
That's just a guess though - I'm no elevator expert...
Does this seem crazy to anybody else?
A person who has an asthma attack while driving or suddenly faints for no apparent reason and runs over and kills 5 people is charged with 3rd-degree murder. They could end up in prison.
A person who carefully plans the assassination of their next door neighbor but botches the job is charged with attempted murder. They could end up in prison, and it would depend to some degree on how successful they were.
Clearly the second person is a danger to society. The first person might or might not be - and the harshest penalty really called for would be to deny them a driver's license if their condition were expected to recur. However, in our modern justice system the two crimes are not very different in penalty.
In my mind, the apparent intentions of the criminal should be one of the most important criteria in sentencing. Punishing people whose only crime is being unlucky and giving an easy sentence to somebody whose only virtue is being lucky is completely at odds with the whole notion of justice and the protection of society...
OK, I admit this was a bit off-topic, but it annoys me to no end to see these kinds of laws on the books...
Well, with sufficient funding I'm sure this guy will manage to discover the series circuit.
You can get a million volts out of batteries if you're determined enough. Getting it to actually power anything, though, is going to be the trick.
I wonder what happens when the aluminum in his "tree" battery runs out?
I assume that ac3 compression is smart enough to take advantage of cross-tack similarities? Your 5.1 soundtrack isn't 5 completely isolated audio streams from 5 completely different movie soundtracks. Sure, when there is a surround effect happening they probably diverge a fair amount, but most of the time the surround speakers are just playing sound track - which is mostly the same across all speakers with a few components louder/quieter and maybe a little phase difference. It wouldn't surprise me if the actual amount of loss is all that high as a result.
Somebody should come up with a FLAC-like algorithm for 5.1 surround that takes advantage of this fact. While I'm sure the resulting tracks would be substantially bigger, I doubt they'd require 7-8GB...
Also CSS was "cracked" more because of XING's goof
Uh, it was hardly a goof...
It was inevitable once they allowed software decoders. If you decode in software then the software has to have both the key, and the algorigthm. You can obfuscate it all you want, but it is still there.
Even with hardware you could eventually obtain it, since the key is stored in the hardware somewhere.
Something a little more secure would be sending the encrypted session key online to some 3rd party, who checks your credentials and sends you back the decoded key - if you intercept it you can only crack that single disk. (This is how many itunes crackers work.) However, if if the player displays the movie, then at some point it must have the key.
However, in software it is MUCH easier to access - unless you have trusted computing (then you'd need the appropriate key out of the BIOS just to read the software code off the hard drive).
Except that if you were on an elevator E travelling at the speed of light realtive to a object A, you'd have to travel an infinite distance relative to A in order for any time to pass in E.
You can't travel at the speed of light relative to object A as measured by either E or A. You could travel at the speed of light relative to object A as measured by object B (somebody halfway in between noting both E and A moving in opposite directions at 0.5c), but A and E would measure their speed as being nearly the speed of light but not quite. Relative to object B, time is still ticking fairly normally for both A and E, though relative to each other each sees the other as nearly-frozen (though everything would appear normal to themselves).
The person in the elevator would not see anything odd about travelling at such a high speed. This is logical since from their frame of reference they are standing still, and everything else is flying past them at nearly the speed of light - time is frozen for everything else, and not them by their measurement. Time is not absolute.
We might actually be on the same wavelength here - maybe not. Hard to tell since it is impossible to be clear without qualifying every statement with who is moving where relative to what and who is taking the measurements...