Yup. They should talk to the PHBs - the guys they sold their last generation of products to. Forget being the best platform to watch movies or listen to music. Figure out how to make it trivial to do collaboration on documents, read email without spending 30 mins booting up your laptop and connecting to the VPN, and so on. Stick a VGA out onto it that works with any video projector built since 1995, and so on.
Give it a slick client that connects to SharePoint and makes document checkin/out, editing, collaboration, etc dead simple. Make sure it supports any WiFi encryption technology or VPN technology deployed anywhere. Design it so that the guys deploying them can plug a USB dongle into it and turn it on and all the corporate policies and setup is done in less than a minute unattended. Heck, offer to preconfigure them if the company mails the dongle to the factory. Deploy a tool that does all the legal records retention stuff to a semi-broken unit and provisions a new unit with the latest backups. Pay all the big CRM/ERP/etc vendors to build a native client for the thing, and offer management solutions for remote sales forces. Provide dashboards and present a vision of huge IT savings, and all that other stuff that consumer-oriented companies don't bother with.
RIM isn't going to capture the teenage market. MS Exchange doesn't cover that well either. And yet, I don't see the latter going away anytime soon. You just have to understand your niche and be the best at it.
I stand corrected. Jobs knew how to make Apple successful, and did so. However, it doesn't really change my point. Every company has a founder that makes it successful.
I do agree that the MBAs of America seem obsessed for the ultimate excel formula that turns any company into the next Apple. I think the reason Jobs was successful was that in the end he was powerful enough to override the MBAs.
This isn't encryption - it is a password entry. I doubt it is even hashed (maybe it is). The issue is that you need to interact with the device and incur delays between attempts, so you can't try 100k combos per second or whatever.
As somebody else pointed out, there is no terminator on the end, so you only need to account for max-length entries. You can repeat entries however.
The reality is that most patterns are probably only 5-6 dots long, and they're usually geometric. A typical grid point has 3.7 neighbors, so that makes the complexity around 9 * 3.7^4. That is only 1700 combos - less than a 4-digit numeric pin.
Sure, if you carefully move your finger between the dots such that any dot could truly follow any other it might really be 9*8^4, and higher if you use a longer key. But, most people slide their finger around and so after the first dot is selected the number of choices each time is only 3-5.
The purpose of the passcode is to restrict casual access, not a determined attacker who has physical access. I don't want to have to unlock my car while driving with a 47 character alphanumeric password.
A simple solution to problems like this is to: 1. Use full-disk encryption and TPM so that any access attempt has to go through the OS. 2. Have the device go into some level of semi-to-full permanent locking when there are more than n failed unlock attempts. 3. Ensure the OS does not provide backdoors to access when the device is locked.
#1 is easily done - sure it takes work but it is a completely solved problem. The level of security in the TPM varies, but assuming the vendor doesn't backdoor a TPM can be VERY secure. A complex boot decryption password with lots of rounds is another option which can't be backdoored short of a hardware keylogger.
#2 is easy to implement on the lock side. It could just be an instruction to the TPM to forget the key followed by a flash wipe. If you want the ability to restore access then you might need to have a backup hard-to-guess password or something that re-unlocks the TPM, or some other mechanism that relies on plugging in a computer and re-authenticating with some kind of pre-generated secret.
#3 is the big issue. #2 only provides security if it gets triggered. Solutions here are to audit your USB and other interface code, and ensure that stuff isn't listening for new connections/etc if the screen is locked. You might also need hardware tamper switches to detect attempts to open the case while powered on and clear the RAM and re-lock the TPM if that happens.
There are known principles for defeating attacks when the attacker has physical possession of the device, but they do require hardware-level support.
Yup. Like Jobs or not he clearly knew how to make a company successful. Chances are he picked a successor who is likely to have the right stuff, even if he isn't another Jobs.
It all goes downhill once the search committee takes over. They're going to pick the same kind of CEO that any committee of this sort does. If the next guy has anything to do with Jobs, the committee will have likely picked him because how much he ISN'T like Jobs. People with what it takes to make a company a big success rarely resonate with a dozen wall-street types.
The cops were working for the city. They authority they abused was derived from the city. The city -- and thus the citizenry -- is responsible for their actions.
The problem with this sort of logic is that it is almost impossible for concerned citizens to reform this kind of behavior. The people making the decisions wield too much power and our electoral system shields them from most accountability. Financial immunity just means they have virtually no incentive at all to do the right thing. Re-election chances have much more to do with prevailing politics than individual actions, even if the latter is what makes it into the issue ads that the candidate with more money runs.
Yup, either that or fine the mayor in charge at the time.
When the people who make decisions collect pay, and when bad decisions are made the fines are paid by taxpayers, consumers, or shareholders, then you have a system ripe for problems like this.
Police abuse taxpayers, and taxpayers pay the fine. Yeah, that will teach them a lesson!
This seems to be the norm these days - a company that was formerly massively successful begins to die down and fade away, so in a last ditch attempt to cling to live they sue anyone and everyone that has ever done anything remotely similar to them.
Every successful company usually becomes that way under a founder. Or a founder-like figure (maybe the company was obscure for 100 years and then takes the world by storm - the leader in this case is like a founder). Under the founder all is well, and the company generally makes net-positive contributions to society.
Then the founder retires, and his hand-picked successor takes over. They usually start having more of an eye towards whatever the founder hired them for (often marketing, or finance, or whatever). However, they were mentored by the founder and usually are fairly true to the original dream.
After that the next succession is managed by the board's CEO search committee, and everybody after this could care less about visions and dreams, and instead aim to min/max their balance sheet and bonus check. Companies don't sue people - their leaders do. By the time a company reaches the state SGI is in, nobody who had anything to do with creating anything of worth is in charge.
Going by this logic, it does also mean that Apple must be ready to die soon. I mean, they can't have that much money, can they?
Every company is doomed to follow this cycle - it is the nature of wall street. Apple is now operating under the hand-picked successor. He will probably do reasonably well, but one day he will retire. Everything after that will be inertia. Oh, it takes a long time for a huge company to fail, and sometimes you get lucky and the wall street pick might actually turn out to be visionary. However, by-and-large the only innovations at Apple starting with the next CEO will be in the balance sheet.
Sooner or later you can only cut the bottom line so much before the fall in the top line starts. At that point the company will bleed off anything of worth it still has left, until its only function can be that decreed by law - it is still able to collect money, write checks to the decision-makers, and file lawsuits, shielding the decision-makers from personal liability. It is nothing more than a front at that point, but it will continue on.
Apple has always been lawsuit-happy, so who knows - perhaps we won't even last another CEO before the slide starts. It all depends on whether they start rewarding the lawyers more than the innovators like they do at most companies.
I think this depends on what you define as "IT." If you're talking networking and mail servers, then it is just like electricity or any other commodity. However, it is still important - if it isn't done right you're going to lose money - often due to lost opportunities. That doesn't mean that it can't be done by the lowest reputable bidder.
However, IT can also be a source of opportunity in almost any industry. Keeping the hard drives backed up might be a commodity, but chances are the data on those drives is an untapped goldmine of info, and getting the value out of it is not a commodity. I've often seen companies foolishly silo jobs such that gathering metrics or doing analytics is the sole job of somebody with a marketing or science degree. The reality is that somebody with business domain expertise working closely side-by-side with somebody with an IT/database background would get much further.
I tend to work in the sciences, and I've seen numerous cases where automation of routine laboratory work is jealously guarded by people with a chemistry background. Usually this kind of effort yields only modest results - 10% here and there in savings. You never really see it result in much. I've also seen some areas where companies have wisely teamed up scientists, engineers, and software developers in small teams and come up with automation systems that can churn out millions of results in the time it would take a building full of scientists to churn out thousands. Now, not every problem needs that level of investment, but there can be benefits when people with the right skills work together and not in competition for the highest spot in the bell curve.
What would you do if you owned a company and went to the bathroom and saw your best $200k/yr marketer under the sink fixing a leak? You'd probably judge that they're wasting their time. Now, suppose the problem is really bad, and you had to hire a team of plumbers who collectively cost $300k/yr to do the job. Chances are that going with them is still a better choice than having your marketer do the work, because they're going to spend so much less time on it that you save money. IT is no different - it is a skillset and discipline, and if you apply it correctly you make a lot more money than if you don't.
Sales make money, salespeople and marketers cost money. Conclusion, we should get rid of marketers.
Phone calls to customers that result in sales make money. Phones cost money. Conclusion, we should get rid of the phones.
Sometimes you need to spend money to make money. IT is an enabler. Sure, it costs money, but so do phones and electricity.
IT is important for the success of any company. However, IT is generally not important for the success of the CEO and executives. I think that is the real problem. Reward cycles in senior management are so short that they have no incentive at all to keep the company profitable over the longer term.
Typically the term MRI is applied to medical use, and NMR to non-medical use.
Also, MRI is generally used for imaging, and NMR usually is not associated with generating images (at least not images where a pixel on the image corresponds to some point in space inside the magnet).
Most MRI images ask the question "is there water in this spot or not?" Most NMR data visualizations represent whether there are certain relationships between particular atoms in some molecule. For example, an NMR might ask the question "for each particular backbone 15N atom in some protein, follow the bond to the attached hydrogen atom and then look for any other hydrogen atom within about 10 angstroms or so, and then identify any 15N atoms attached to those." From that you can elucidate the 3D structure of the protein (well, in reality it is just one out of a large set of clues you need to do this).
I have read a little on some more elaborate functional MRI studies that start to move in this direction. For example, you can ask the question "show me all the water in this spot that wasn't there a moment before." Or you can ask "show me where all the water in this particular spot went after half a second."
Yup, if it is just junk downloaded off of TPB, well then your recovery solution is to re-download it off of TPB.
If this is raw DV video of your vacations, convert it to mpeg2 or something a bit less disk-heavy, and burn it to DVD, or store it on a hard drive in a remote location. That stuff doesn't change much.
For stuff like documents there are lots of good cloud-based solutions - you should be talking 100's of MB at most there, not TB. You need to be smart and determine what is really worth saving.
Yup, I use cloud backup (homebrew solution), and while I have TB of stuff, I only back up the stuff I really need. If I lose TV episodes on my DVR I'll live, but I don't want to lose documents/records/etc. Configuration settings are also useful to backup -/etc, and such. The goods thing is that most of the really essential stuff is very compact - I cast a bit wider net and only backup a few GB (daily incrementals, and so on).
If you can't afford to back it up properly, then plan to live without it. Technology breaks...
Sure, but why deploy security patches if you're behind a firewall, and why put the computer in a cabinet if employees aren't allowed to bring in usb drives?
This is called defense in depth. It means that even if an attacker gets past a layer of security their ability to exploit is greatly reduced. If you put all your energy into a single layer of protection then if somebody subverts it they have the run of the place.
The extra layers of security, coupled with guards/etc, mean that an attacker is going to take a long time to do anything. Even if they are armed with zero-days against every layer of your software security they still have to break into cabinets and solder around glued up USB ports/etc, and that makes them easy to catch with roaming patrols.
Sure, nothing is perfect, but layers of security are the best way to contain problems.
I can't really speak to wine but I can talk a little about bread (a side hobby of mine on occasion). It isn't that hard to make bread in your home that is substantially better than just about anything you'll find in a marketplace without travelling a fair bit. Now, you won't get it just by dumping flour in a bread machine, but you can automate quite a bit of it (most of the difference comes from extended fermentation at the right spots in the process, or cold fermentation).
The reason that you can do so much better at home is due to a few factors: 1. Techniques like cold fermentation aren't practical if you scale up. You probably have room for two pounds of dough in your fridge at home, and that fridge can probably cool it down in an hour (especially if you used cold water to make it, and getting a few hundred ml of that isn't expensive at all). Then warming up two pounds of dough consists of setting it on a counter for a few hours. If you try to do the same thing with 2000 pounds of dough it is a whole different ball game chilling that much water, finding space to store it, and then warming it up without running a network of heating tubes through the massive pile of dough.
2. You can make it when you want to eat it. You aren't making 300 loaves and then having to compromise on composition and preservatives to get it to last 3 days on a store shelf. I'll be the first to admit that one of my home made loaves tastes far worse than a store loaf if you allow both to sit for a day before eating it. I typically freeze anything I won't immediately eat (which destroys texture, but leaves taste generally intact).
It doesn't surprise me that other techniques that employ fermentation like wine have similar benefits from being done at home. You can do the extra step that is simple on a few bottles but not on a few thousand, or which can't be done by a machine.
Tend to agree here. I think another issue is likely to be insurance payments. A new pulse sequence might as well be a new class of drug as far as getting paid for it goes. That's another group of people who don't understand NMR theory to try to explain what you're doing to.
The high fields are another big issue. MRIs must be big money since they're popping up all over the place, but I imagine that much of that money comes from (possibly unnecessary) shoulder pain images and such. Maybe a stroke victim might benefit from a diffusion MRI of the brain, but in my local hospital they just keep you around for a day and take the picture with a more conventional test on a cheaper spectrometer. Then again, I'm not sure what intervention options actually exist that would create a benefit from earlier detection, but maybe if we actually reliably detected non-hemorrhagic evolving strokes in the ER somebody might figure out a treatment option.
I personally know somebody who had intermittent and somewhat subtle stroke symptoms, and the doctors were pretty mixed on whether anything was wrong, though they admitted her (not really with any kind of aggressive treatment beyond a CT scan and if anything her normal antiplatelet medication was delayed due to the usual hospital practice of stopping all medications until all the paperwork is finished 12 hours later). The next day she had a vocabulary of about 12 words and a golfball-sized bright spot on an MRI. I'm not really sure if more rapid action could have changed the outcome since she had the symptoms when she woke up and was outside the 3 hour window for most direct intervention, but it was incredibly frustrating trying to convince the doctors that something was wrong and at the very least a diffusion MRI might have shown something right away and ended the debate. Then again, I'm sure the MRI machine was busy with a long line of shoulder injuries.
So, whether they would have made any difference in this particular case is debatable, but I think it is pretty clear that there are a lot of potential applications for MRI that are untapped, and most of the barriers are non-technological in nature.
You don't punish businesses, because business are incapable of making bad decisions.
You punish executives, because they are.
Otherwise, anytime an executive makes a crooked decision and the business profits they take a big bonus, and anytime the business gets punished the shareholders take a loss. That is a recipe for the situation we're in now.
Besides spacial resolution, higher field MRI would allow more chemical information to be probed. At sufficient field strengths it becomes more practical to image nuclei other than hydrogen. You can also study stuff like diffusion and movement of blood/etc. Nuclear Magnetic Resonance has many applications beyond the fairly simple pictures generated by most MRI machines, but the biggest limitation has been the field sizes of magnets large enough to stuff a person into.
Imaging other nuclei than hydrogen would get around your microwave issue - their resonance frequencies are much lower. Of course, with the lower frequencies comes lower resolution as well.
However, generating magnetic fields with capacitor pulses, non-superconducting electromagnets, and explosions (another technique) aren't very useful for MRI since they tend to generate very unstable and non-uniform fields. I remember reading an article about this facility about 12 years ago and they were using "conventional" electromagnets that were water-cooled. They were pumping some insane amount of water through the magnet and it would come out way warmer than it went in. I think the power draw was a number of megawatts just for the magnet coil.
14 years ago I routinely used a 12T NMR, and we had a 17T model in the basement (time on that one was a bit harder to get).
However, to be useful for most forms of NMR the field has to be very uniform and stable over long periods of time. You can't do that with pulses or some of the other techniques used at this high-field lab. There are of course all kinds of other things you can do there.
MRIs tend to be much weaker than NMRs. The problem is the bore size of the magnet. Scaling up the size of a magnet is very difficult, and it takes a lot more energy to make a weak field the size of the earth than a strong one that you can fit a skinny test tube inside. This is similar to the difference between temperature and heat. A match and a bonfire might be the same temperature, but the bonfire puts out way more heat.
Medical MRI tends to be only a few T at most. Really big ones are in the very low teens, and are VERY expensive to build. Of course, MRIs have spacial resolution and NMRs typically do not. An NMR probes fairly complex chemical relationships but does not generate a spacial image. An MRI probes fairly simple chemical relationships (often just the presence of water or a contrast agent), but it takes a 3D picture.
The other more modern trend is building bigger NMRs but instead of making them more powerful using extra magnets to cancel out the field outside of the dewer. This makes them easier to site - and people don't get injured by flying tools if somebody is careless. High-field NMRs can be very dangerous when performing operations like filling with the aid of gas cylinders (with very long hoses). Shielding or not, another big danger with either NMRs or MRIs is ventilation. If something causes the magnet to quench you can get huge volumes of He/N2 liberation which will quickly displace all O2 in even a large room.
Here is my problem with these kinds of studies. Did they enroll thousands of people for many years specifically to test whether aspirin prevents cancer? I doubt it. They probably were testing for prevention of heart disease and found additionally that it reduces cancer.
So, what's wrong with that, you ask. Xkcd as usual sums it up.
Well, how likely is it that the conclusion is right? The scientists will no doubt show that it is 99% likely that they are right based on confidence limits and all that. However, that doesn't account for how many hypothesis were tested. If you give me a big data set with lots of individual maladies within it, I can show a 99% confidence relationship to SOMETHING in that list for sure.
The only way I trust a clinical trial is if the endpoint of the trial was declared BEFORE the trial was started, and then THAT endpoint was reached. Repeat this whole trial again looking specifically for cancer and I'll believe the results, even if EVERY aspect of the trial is conducted in the same way.
This sort of argument might seem strange to somebody not versed in statistics, but hypothesis is mining is a big problem.
Slashdot Mirror
Well, if they gave each of the departing engineers 5x their annual salary in a golden parachute I doubt the engineers would be complaining.
Yup. They should talk to the PHBs - the guys they sold their last generation of products to. Forget being the best platform to watch movies or listen to music. Figure out how to make it trivial to do collaboration on documents, read email without spending 30 mins booting up your laptop and connecting to the VPN, and so on. Stick a VGA out onto it that works with any video projector built since 1995, and so on.
Give it a slick client that connects to SharePoint and makes document checkin/out, editing, collaboration, etc dead simple. Make sure it supports any WiFi encryption technology or VPN technology deployed anywhere. Design it so that the guys deploying them can plug a USB dongle into it and turn it on and all the corporate policies and setup is done in less than a minute unattended. Heck, offer to preconfigure them if the company mails the dongle to the factory. Deploy a tool that does all the legal records retention stuff to a semi-broken unit and provisions a new unit with the latest backups. Pay all the big CRM/ERP/etc vendors to build a native client for the thing, and offer management solutions for remote sales forces. Provide dashboards and present a vision of huge IT savings, and all that other stuff that consumer-oriented companies don't bother with.
RIM isn't going to capture the teenage market. MS Exchange doesn't cover that well either. And yet, I don't see the latter going away anytime soon. You just have to understand your niche and be the best at it.
Fair enough - if you're a lawyer you might not be crippled by a lawsuit.
Not really sure it changes the fact that the system is pretty broken.
I stand corrected. Jobs knew how to make Apple successful, and did so. However, it doesn't really change my point. Every company has a founder that makes it successful.
I do agree that the MBAs of America seem obsessed for the ultimate excel formula that turns any company into the next Apple. I think the reason Jobs was successful was that in the end he was powerful enough to override the MBAs.
This isn't encryption - it is a password entry. I doubt it is even hashed (maybe it is). The issue is that you need to interact with the device and incur delays between attempts, so you can't try 100k combos per second or whatever.
As somebody else pointed out, there is no terminator on the end, so you only need to account for max-length entries. You can repeat entries however.
The reality is that most patterns are probably only 5-6 dots long, and they're usually geometric. A typical grid point has 3.7 neighbors, so that makes the complexity around 9 * 3.7^4. That is only 1700 combos - less than a 4-digit numeric pin.
Sure, if you carefully move your finger between the dots such that any dot could truly follow any other it might really be 9*8^4, and higher if you use a longer key. But, most people slide their finger around and so after the first dot is selected the number of choices each time is only 3-5.
The purpose of the passcode is to restrict casual access, not a determined attacker who has physical access. I don't want to have to unlock my car while driving with a 47 character alphanumeric password.
A simple solution to problems like this is to:
1. Use full-disk encryption and TPM so that any access attempt has to go through the OS.
2. Have the device go into some level of semi-to-full permanent locking when there are more than n failed unlock attempts.
3. Ensure the OS does not provide backdoors to access when the device is locked.
#1 is easily done - sure it takes work but it is a completely solved problem. The level of security in the TPM varies, but assuming the vendor doesn't backdoor a TPM can be VERY secure. A complex boot decryption password with lots of rounds is another option which can't be backdoored short of a hardware keylogger.
#2 is easy to implement on the lock side. It could just be an instruction to the TPM to forget the key followed by a flash wipe. If you want the ability to restore access then you might need to have a backup hard-to-guess password or something that re-unlocks the TPM, or some other mechanism that relies on plugging in a computer and re-authenticating with some kind of pre-generated secret.
#3 is the big issue. #2 only provides security if it gets triggered. Solutions here are to audit your USB and other interface code, and ensure that stuff isn't listening for new connections/etc if the screen is locked. You might also need hardware tamper switches to detect attempts to open the case while powered on and clear the RAM and re-lock the TPM if that happens.
There are known principles for defeating attacks when the attacker has physical possession of the device, but they do require hardware-level support.
Yup. Like Jobs or not he clearly knew how to make a company successful. Chances are he picked a successor who is likely to have the right stuff, even if he isn't another Jobs.
It all goes downhill once the search committee takes over. They're going to pick the same kind of CEO that any committee of this sort does. If the next guy has anything to do with Jobs, the committee will have likely picked him because how much he ISN'T like Jobs. People with what it takes to make a company a big success rarely resonate with a dozen wall-street types.
The cops were working for the city. They authority they abused was derived from the city. The city -- and thus the citizenry -- is responsible for their actions.
The problem with this sort of logic is that it is almost impossible for concerned citizens to reform this kind of behavior. The people making the decisions wield too much power and our electoral system shields them from most accountability. Financial immunity just means they have virtually no incentive at all to do the right thing. Re-election chances have much more to do with prevailing politics than individual actions, even if the latter is what makes it into the issue ads that the candidate with more money runs.
Yeah, good luck holding a job after spending who knows how long in jail, followed by having to make court appearances over a five year trial...
Yup, either that or fine the mayor in charge at the time.
When the people who make decisions collect pay, and when bad decisions are made the fines are paid by taxpayers, consumers, or shareholders, then you have a system ripe for problems like this.
Police abuse taxpayers, and taxpayers pay the fine. Yeah, that will teach them a lesson!
This seems to be the norm these days - a company that was formerly massively successful begins to die down and fade away, so in a last ditch attempt to cling to live they sue anyone and everyone that has ever done anything remotely similar to them.
Every successful company usually becomes that way under a founder. Or a founder-like figure (maybe the company was obscure for 100 years and then takes the world by storm - the leader in this case is like a founder). Under the founder all is well, and the company generally makes net-positive contributions to society.
Then the founder retires, and his hand-picked successor takes over. They usually start having more of an eye towards whatever the founder hired them for (often marketing, or finance, or whatever). However, they were mentored by the founder and usually are fairly true to the original dream.
After that the next succession is managed by the board's CEO search committee, and everybody after this could care less about visions and dreams, and instead aim to min/max their balance sheet and bonus check. Companies don't sue people - their leaders do. By the time a company reaches the state SGI is in, nobody who had anything to do with creating anything of worth is in charge.
Going by this logic, it does also mean that Apple must be ready to die soon. I mean, they can't have that much money, can they?
Every company is doomed to follow this cycle - it is the nature of wall street. Apple is now operating under the hand-picked successor. He will probably do reasonably well, but one day he will retire. Everything after that will be inertia. Oh, it takes a long time for a huge company to fail, and sometimes you get lucky and the wall street pick might actually turn out to be visionary. However, by-and-large the only innovations at Apple starting with the next CEO will be in the balance sheet.
Sooner or later you can only cut the bottom line so much before the fall in the top line starts. At that point the company will bleed off anything of worth it still has left, until its only function can be that decreed by law - it is still able to collect money, write checks to the decision-makers, and file lawsuits, shielding the decision-makers from personal liability. It is nothing more than a front at that point, but it will continue on.
Apple has always been lawsuit-happy, so who knows - perhaps we won't even last another CEO before the slide starts. It all depends on whether they start rewarding the lawyers more than the innovators like they do at most companies.
I think this depends on what you define as "IT." If you're talking networking and mail servers, then it is just like electricity or any other commodity. However, it is still important - if it isn't done right you're going to lose money - often due to lost opportunities. That doesn't mean that it can't be done by the lowest reputable bidder.
However, IT can also be a source of opportunity in almost any industry. Keeping the hard drives backed up might be a commodity, but chances are the data on those drives is an untapped goldmine of info, and getting the value out of it is not a commodity. I've often seen companies foolishly silo jobs such that gathering metrics or doing analytics is the sole job of somebody with a marketing or science degree. The reality is that somebody with business domain expertise working closely side-by-side with somebody with an IT/database background would get much further.
I tend to work in the sciences, and I've seen numerous cases where automation of routine laboratory work is jealously guarded by people with a chemistry background. Usually this kind of effort yields only modest results - 10% here and there in savings. You never really see it result in much. I've also seen some areas where companies have wisely teamed up scientists, engineers, and software developers in small teams and come up with automation systems that can churn out millions of results in the time it would take a building full of scientists to churn out thousands. Now, not every problem needs that level of investment, but there can be benefits when people with the right skills work together and not in competition for the highest spot in the bell curve.
What would you do if you owned a company and went to the bathroom and saw your best $200k/yr marketer under the sink fixing a leak? You'd probably judge that they're wasting their time. Now, suppose the problem is really bad, and you had to hire a team of plumbers who collectively cost $300k/yr to do the job. Chances are that going with them is still a better choice than having your marketer do the work, because they're going to spend so much less time on it that you save money. IT is no different - it is a skillset and discipline, and if you apply it correctly you make a lot more money than if you don't.
Sales make money. You cost money.
Sales make money, salespeople and marketers cost money. Conclusion, we should get rid of marketers.
Phone calls to customers that result in sales make money. Phones cost money. Conclusion, we should get rid of the phones.
Sometimes you need to spend money to make money. IT is an enabler. Sure, it costs money, but so do phones and electricity.
IT is important for the success of any company. However, IT is generally not important for the success of the CEO and executives. I think that is the real problem. Reward cycles in senior management are so short that they have no incentive at all to keep the company profitable over the longer term.
Typically the term MRI is applied to medical use, and NMR to non-medical use.
Also, MRI is generally used for imaging, and NMR usually is not associated with generating images (at least not images where a pixel on the image corresponds to some point in space inside the magnet).
Most MRI images ask the question "is there water in this spot or not?" Most NMR data visualizations represent whether there are certain relationships between particular atoms in some molecule. For example, an NMR might ask the question "for each particular backbone 15N atom in some protein, follow the bond to the attached hydrogen atom and then look for any other hydrogen atom within about 10 angstroms or so, and then identify any 15N atoms attached to those." From that you can elucidate the 3D structure of the protein (well, in reality it is just one out of a large set of clues you need to do this).
I have read a little on some more elaborate functional MRI studies that start to move in this direction. For example, you can ask the question "show me all the water in this spot that wasn't there a moment before." Or you can ask "show me where all the water in this particular spot went after half a second."
as long as you keep it in solution, it won't blow up on you.
Yeah, but keeping something 100% in solution isn't as easy as it sounds. All it takes is a little residue in the threads of a cap to dry out...
Only as the dries and forms crystals - no problem while it's in solution...
Yeah, kind of like what happens as your reservoir in the super soaker gets depleted and forms a film on the inside...
Yup, if it is just junk downloaded off of TPB, well then your recovery solution is to re-download it off of TPB.
If this is raw DV video of your vacations, convert it to mpeg2 or something a bit less disk-heavy, and burn it to DVD, or store it on a hard drive in a remote location. That stuff doesn't change much.
For stuff like documents there are lots of good cloud-based solutions - you should be talking 100's of MB at most there, not TB. You need to be smart and determine what is really worth saving.
Yup, I use cloud backup (homebrew solution), and while I have TB of stuff, I only back up the stuff I really need. If I lose TV episodes on my DVR I'll live, but I don't want to lose documents/records/etc. Configuration settings are also useful to backup - /etc, and such. The goods thing is that most of the really essential stuff is very compact - I cast a bit wider net and only backup a few GB (daily incrementals, and so on).
If you can't afford to back it up properly, then plan to live without it. Technology breaks...
Sure, but why deploy security patches if you're behind a firewall, and why put the computer in a cabinet if employees aren't allowed to bring in usb drives?
This is called defense in depth. It means that even if an attacker gets past a layer of security their ability to exploit is greatly reduced. If you put all your energy into a single layer of protection then if somebody subverts it they have the run of the place.
The extra layers of security, coupled with guards/etc, mean that an attacker is going to take a long time to do anything. Even if they are armed with zero-days against every layer of your software security they still have to break into cabinets and solder around glued up USB ports/etc, and that makes them easy to catch with roaming patrols.
Sure, nothing is perfect, but layers of security are the best way to contain problems.
I can't really speak to wine but I can talk a little about bread (a side hobby of mine on occasion). It isn't that hard to make bread in your home that is substantially better than just about anything you'll find in a marketplace without travelling a fair bit. Now, you won't get it just by dumping flour in a bread machine, but you can automate quite a bit of it (most of the difference comes from extended fermentation at the right spots in the process, or cold fermentation).
The reason that you can do so much better at home is due to a few factors:
1. Techniques like cold fermentation aren't practical if you scale up. You probably have room for two pounds of dough in your fridge at home, and that fridge can probably cool it down in an hour (especially if you used cold water to make it, and getting a few hundred ml of that isn't expensive at all). Then warming up two pounds of dough consists of setting it on a counter for a few hours. If you try to do the same thing with 2000 pounds of dough it is a whole different ball game chilling that much water, finding space to store it, and then warming it up without running a network of heating tubes through the massive pile of dough.
2. You can make it when you want to eat it. You aren't making 300 loaves and then having to compromise on composition and preservatives to get it to last 3 days on a store shelf. I'll be the first to admit that one of my home made loaves tastes far worse than a store loaf if you allow both to sit for a day before eating it. I typically freeze anything I won't immediately eat (which destroys texture, but leaves taste generally intact).
It doesn't surprise me that other techniques that employ fermentation like wine have similar benefits from being done at home. You can do the extra step that is simple on a few bottles but not on a few thousand, or which can't be done by a machine.
Tend to agree here. I think another issue is likely to be insurance payments. A new pulse sequence might as well be a new class of drug as far as getting paid for it goes. That's another group of people who don't understand NMR theory to try to explain what you're doing to.
The high fields are another big issue. MRIs must be big money since they're popping up all over the place, but I imagine that much of that money comes from (possibly unnecessary) shoulder pain images and such. Maybe a stroke victim might benefit from a diffusion MRI of the brain, but in my local hospital they just keep you around for a day and take the picture with a more conventional test on a cheaper spectrometer. Then again, I'm not sure what intervention options actually exist that would create a benefit from earlier detection, but maybe if we actually reliably detected non-hemorrhagic evolving strokes in the ER somebody might figure out a treatment option.
I personally know somebody who had intermittent and somewhat subtle stroke symptoms, and the doctors were pretty mixed on whether anything was wrong, though they admitted her (not really with any kind of aggressive treatment beyond a CT scan and if anything her normal antiplatelet medication was delayed due to the usual hospital practice of stopping all medications until all the paperwork is finished 12 hours later). The next day she had a vocabulary of about 12 words and a golfball-sized bright spot on an MRI. I'm not really sure if more rapid action could have changed the outcome since she had the symptoms when she woke up and was outside the 3 hour window for most direct intervention, but it was incredibly frustrating trying to convince the doctors that something was wrong and at the very least a diffusion MRI might have shown something right away and ended the debate. Then again, I'm sure the MRI machine was busy with a long line of shoulder injuries.
So, whether they would have made any difference in this particular case is debatable, but I think it is pretty clear that there are a lot of potential applications for MRI that are untapped, and most of the barriers are non-technological in nature.
You don't punish businesses, because business are incapable of making bad decisions.
You punish executives, because they are.
Otherwise, anytime an executive makes a crooked decision and the business profits they take a big bonus, and anytime the business gets punished the shareholders take a loss. That is a recipe for the situation we're in now.
Besides spacial resolution, higher field MRI would allow more chemical information to be probed. At sufficient field strengths it becomes more practical to image nuclei other than hydrogen. You can also study stuff like diffusion and movement of blood/etc. Nuclear Magnetic Resonance has many applications beyond the fairly simple pictures generated by most MRI machines, but the biggest limitation has been the field sizes of magnets large enough to stuff a person into.
Imaging other nuclei than hydrogen would get around your microwave issue - their resonance frequencies are much lower. Of course, with the lower frequencies comes lower resolution as well.
However, generating magnetic fields with capacitor pulses, non-superconducting electromagnets, and explosions (another technique) aren't very useful for MRI since they tend to generate very unstable and non-uniform fields. I remember reading an article about this facility about 12 years ago and they were using "conventional" electromagnets that were water-cooled. They were pumping some insane amount of water through the magnet and it would come out way warmer than it went in. I think the power draw was a number of megawatts just for the magnet coil.
14 years ago I routinely used a 12T NMR, and we had a 17T model in the basement (time on that one was a bit harder to get).
However, to be useful for most forms of NMR the field has to be very uniform and stable over long periods of time. You can't do that with pulses or some of the other techniques used at this high-field lab. There are of course all kinds of other things you can do there.
MRIs tend to be much weaker than NMRs. The problem is the bore size of the magnet. Scaling up the size of a magnet is very difficult, and it takes a lot more energy to make a weak field the size of the earth than a strong one that you can fit a skinny test tube inside. This is similar to the difference between temperature and heat. A match and a bonfire might be the same temperature, but the bonfire puts out way more heat.
Medical MRI tends to be only a few T at most. Really big ones are in the very low teens, and are VERY expensive to build. Of course, MRIs have spacial resolution and NMRs typically do not. An NMR probes fairly complex chemical relationships but does not generate a spacial image. An MRI probes fairly simple chemical relationships (often just the presence of water or a contrast agent), but it takes a 3D picture.
The other more modern trend is building bigger NMRs but instead of making them more powerful using extra magnets to cancel out the field outside of the dewer. This makes them easier to site - and people don't get injured by flying tools if somebody is careless. High-field NMRs can be very dangerous when performing operations like filling with the aid of gas cylinders (with very long hoses). Shielding or not, another big danger with either NMRs or MRIs is ventilation. If something causes the magnet to quench you can get huge volumes of He/N2 liberation which will quickly displace all O2 in even a large room.
Here is my problem with these kinds of studies. Did they enroll thousands of people for many years specifically to test whether aspirin prevents cancer? I doubt it. They probably were testing for prevention of heart disease and found additionally that it reduces cancer.
So, what's wrong with that, you ask. Xkcd as usual sums it up.
Well, how likely is it that the conclusion is right? The scientists will no doubt show that it is 99% likely that they are right based on confidence limits and all that. However, that doesn't account for how many hypothesis were tested. If you give me a big data set with lots of individual maladies within it, I can show a 99% confidence relationship to SOMETHING in that list for sure.
The only way I trust a clinical trial is if the endpoint of the trial was declared BEFORE the trial was started, and then THAT endpoint was reached. Repeat this whole trial again looking specifically for cancer and I'll believe the results, even if EVERY aspect of the trial is conducted in the same way.
This sort of argument might seem strange to somebody not versed in statistics, but hypothesis is mining is a big problem.