Re:RAID5 doesn't need a RAID Card!
on
IDE RAID Examined
·
· Score: 1
Oh, but it is true. The system was rebooted and sat at the raid card bios screen for over 24 hours while the raid parity data was rebuilt. In Linux, the system will reboot in a few minutes (mostly depending on whether or not it needs to fsck). The raid array will be online but parity data will reconstruct in the background.
Re:Drive reliability/backups are major factors
on
IDE RAID Examined
·
· Score: 1
The biggest headache with these arrays is the fact that they can be so big and so very cheap. I built a 1-TB rackmount system for a client over a year ago for about $8k. The same unit now would be about $4k. But there is no way in the world to back them up. A tape juke will cost about twice what the array does! So, we just build a second mirror raid array on a separate box, tie the two together with point-to-point gigabit, and cron an rsync every evening at 1 AM. It works great and we get the extra protection against accidental deletion and file corruption that comes with tape backups...
I think that the era of tape backups is past...at least for budget archiving of large data storgae.
RAID5 doesn't need a RAID Card!
on
IDE RAID Examined
·
· Score: 2, Insightful
I have about 5 TBs of RAID5 storage online at various customer sites. They are all using Linux software RAID and Promise ATA66/100/133 controllers. Even when using two drives per IDE channel, we still see very good performance. An RAID5 system with eight 120-GB 5400-RPM Maxtor drives gives about 55 MB/sec write and 80 MB/sec read performance under Bonnie. Those eight drives were on two Promise ATA100 controllers. Cabling is fairly easy if you use 24" UltraATA cables. And it will get much easier with Serial ATA.
One customer ordered a system from a vendor who insisted on installing an ATA raid card, and it was a remarkable disappointment. Linux was able to indentify the array as a SCSI device and mount it. Then, for some reason, the customer rebooted his system. During the BIOS detection, the raid card started doing parity reconstruction and ran for over 24 hours before finally allowing the system to boot! For comparison, the same sized array would resync in the background under Linux in about 3 hours.
Also, the reconstruction tools built into the raid cards are pretty limited. If you have a problem with a Linux software RAID array, at least you can use the normal low level tools to access the drives and try to diagnose the problems. Just MO.
Assume 100,000 miles over the lifetime of the car, $1.50 per gallon for gas. At 50 MPG, you would spend $3000 on gas over the life of the car. At 40 MPG, it would be $3750. At 30 MPG, $5000. So, that is only a $2000 savings compared to $3000 incremental cost.
Of course, if you double fuel costs, the economics change. Also, if you plan on getting 250,000 miles from your car, it also changes. At the very least, I would say that there is no economic advantage to higher efficiency systems like this. This of course ignores the costs of "environmental impact" which are very difficult to quantify. The key issues is that it makes technological sense and the economics are not too bad.
There is one important point that seems to be often missed. Once a manufacturing process is worked out for the fibers, it will start to become much cheaper. The big cost is hauling all of the first one up into orbit at the beginning.
Once you have ONE of these in orbit, it will be fairly easy to spawn a second or to really strengthen the first. You can tow small fibers of additional nanotubes as reinforcement to the existing ribbon. This is when things can get very interesting. You can build an enormously strong elevator on top of this first seminal one. With multiple strong elevators servicing a single space station, the system could become more like a train track than an elevator, with multiple cars going up and down simultaneously.
Multiple elevators would make the entire system more stable and less suspectible to terrorist attack or weather damage. Imagine putting 1000s of tons of material into geosynch orbit per day. Then we can seriously think about wholesale utilization of space and its resources.
Do we know that the ion density in the ionosphere will not be high enough to cause a catastrophic discharge? The typical way to "generate" lightning strikes is to fire a small rocket into a thundercloud with a small wire attached.
The ionosphere is replentished by solar winds so I don't know if depletion is a greater concern than the lightning strike issue. I haven't looked at the numbers though. I may be wrong.
"There is something to be said for maturity and stability in a language."
This is quite true, but there is a huge issue with functionality, stabilit, and performance of the compilers. The Fortran standard is fairly simple, so you would think that compiler makers could get it right. But, we have found that EVERY Linux Fortran compiler revision from EVERY vendor (except maybe Intel) introduces new glitches. There is no gold standard under Linux, though the Intel compiler might become one in another few years. All of the Fortran90+ compilers are closed source. The GCC Fortran compiler is Fortran77 and has no support for things like OpenMP.
The big issue here is lack of market size. The potential market for Fortran compilers is about 1% of that for C or Java. The vendors just don't have enough customers/revenue to provide adequate QA and development, IMO.
I have seen VGA-out boxes for both Xbox and PS2 on www.lik-sang.com. I didn't include an href because the site is down. I have been considering these (with a KVM switch) as a nice way to interface with a front-projector display. Certainly seems cleaner than component-to-VGA-to-DVI rigs.
I agree. After changing to S-video/component instead of composite, the single biggest improvement in visual quality on large television sets is progressive scan or at least line-doubling. My 27" and 36" tube TVs both show serious "tearing" with DVDs and PS2/Xbox games. Unfortunately, neither support 480p. The examples of 480p output on similarly sized screens made a rather dramatic improvement in picture stability, especially when viewed from only a few feet away.
Why not place the PLLs on every other bank of normal DDR slots on a motherboard. Banks 0 and 2 would be normal. Banks 1 and 3 would be 90-degrees off. If you want double bandwidth, install DIMMs in pairs like on normal interleaving motherboards. Trace complexity would remain unchanged, motherboards would only need a few PLLs inlined, and we could all get twice the bandwidth using $40 DDR DIMMs...and not have to wait for new memory modules to drop in price.
It shouldn't surprise us that nuclear explosions can be shaped. The "implosion" triggering system used in Fat Man and most subsequent nuclear devices made use of shaped chemical explosions, as do most anti-armor munitions. These work by the focusing of shock waves through careful shaping of the explosive material, precise denotation time and locations, and "lens" of varying composition to focus the shockwaves.
By constructing the explosive device to reinforce that shockwave, a significant fraction of the released energy can be focused in one direction. By carefully focusing the physical shock, the neutron flux, and x-ray release, it should be possible to build a nuclear explosion that goes "that-a-way" rather than just spherically outward.
No, it is not only a stability issue. The strength of the shock determines the pressure jump. The point is to use the shock as a pressurization mechanism, rather than a spinning compressor section as in a normal turbofan engine. There is certain level of pressurization that needs to occur before this is effective. Just like an IC engine, if you cannot get a high enough compression ratio, you cannot have an efficient engine. In this case, lack of sufficient pressurization can cause reversal of the combusted material back through the intake rather than expanding and accelerating out the nozzle to the rear. This is a problem for both ramjet and scramjet engines
AFAIK, stability is the primary issue in moving from ramjet to scramjet. Supersonic combustion, acoustic coupling, etc. can make for lots of stability problems. Getting good mixing of air and fuel prior to the combustion/shock surface is critical and not all that easy, considering the high inlet velocities.
Also, the presence of a compressor section tends to lead to predictable inlet conditions for the combustor section because effects of turbulence at the intake are muted. For a scramjet, there is no similar buffer, so inlet turbulence adds even more perturbations to the shock/combustion surface. Right now, we have no control systems that can actively control turbulence. This is just a very hard engineering problem...deceptively so even. At least on paper, scramjets are nothing more than a carefully shaped duct and some fuel jets.
At Mach 20, you are definitely going to have atmospheric ionization. That is going to make your "cruise" missile light up a radar display.
But your point is quite valid. It is the issue of detection versus tracking. If I have listening posts a few hundred km from my borders, I would be able to know that something was coming. If I have a fine grained detection network, I can know a lot about its flight path at the point of detection. But, I am not going to have a lot of time to react if it is travelling at Mach-8.
Although, the missile will need to decelerate as it closes. The cruising altitude is going to be quite high, else the thing will burn up. So, it will have to fall much like a MRV.
OK, I guess you could build a decent missile out of one of these, but I don't see how would be superior to a ICMB or MRBM.
I am beginning to realize that I have read entirely too many Clancy novels. 8)
You missed the biggest reason. Cruise missiles only fly at subsonic speeds by design. They are to be quiet and stealthy. Any supersonic signature is detectable with a simple network of microphones. Once you know it is coming, a cruise missle is easier to shoot down than a jet aircraft as cruise missiles typically don't have active countermeasures and certainly don't have a pilot to perform evasive manuevers.
Um, just as a point of reference, ICBMs travel much faster than Mach 7. On suborbital trajectories from the other side of the globe, you might see a time of flight of 30 minutes or less. Think 20,000 km/hour.
Scramjets are not really interesting as strategic weapons. Extra-atmospheric vehicles (MVRs) are faster and proven 30-year-old tech. Scramjets are going to be useless for cruise missiles, because a Mach-7 shock cone will standout rather nicely even if the missile itself were stealthy. Depending on the altitude, it could also cause ionization of the atmosphere which would show up on radar!
Military applications here are going to be reactive in nature...fighter-bombers that can reach any corner of the globe in two-hours is a big selling point, as is the (literally) stratospheric flight ceilings such crafts would have. But I don't know what form a scramjet-based weapons system might need to take or what niche it might fill.
there will probably be "low-speed" versions for shorter distances
Probably not. To understand why requires some knowledge of how a scramjet differs from a normal turbofan engine. There are no spinning parts in a scramjet or ramjet engine. The (sc)ram engine requires a strong standing shock to me maintained in the intake. This standing shock replaces the compressor section of a normal turbo fan. There is a minimum speed which will produce a sufficiently strong, stable shock that will allow this to work.
The SC part is for supersonic combustion which makes that standing shock also replace the combustor portion of the turbo fan. Chemical reactions and transonic fluid dynamics can interact in very complicated ways. This can make this supersonic combustion unstable. The best way to stabilize it is to go faster and increase the strength of the shock.
So, to sum up, operating scramjets at lower speeds is more difficult, so if anything, we will probably see them operating at the highest possible speeds that the airframe and aerodynamics will allow.
This struck me. I am just recently not "in my 20's" anymore. My cousin is an undergrad and he is in his early twenties. There is a huge difference. Undergrad days for me were done with a 286 and then later a very expensive 486. Programming was in TurboPascal, Fortran77, QuickBasic (ICK!), and dBase (dobule ICK!). For him, it is Java, OO-this and that, C++ if he is feeling old-school. So much has changed.
Structural changes have occured in the way that "we do what it is we do." How can anyone have specific longterm goals? I have had some success in the past several years by doing what I am good at, what I enjoy, and chasing after opportunities. I run my own company, but I can honestly say that I don't know what my plans are for 5 years from now, let alone 20. I guess I am not saying anything that you haven't...but let me try.
Planning a carreer is a nonlinear optimization problem. There is a functional that denotes value. It rolls up the financial, social, and intellectual rewards that you get from your carreer. This value functional must be integrated over a path from t=18 years to t=55 years. The path that we take should optimize that total value. But, the value space is highly nonlinear. There will be huge numbers of local extrema...making the best of the current situation. There will be paths that seem to be going nowhere and then lead to huge gains or falls.
So, in handling these types of optimization problems, we need to use common sense (gradient methods) to make the most of what we have at hand. But to keep from being locked into the first local extrema we find, we need to shake things up (entropy methods). Philosophically, I think the same recommendations hold for steering your life and carreer. Choose the best path in front of you, but never stop shaking things up, lest you get stuck.
Seeing the end is not nearly as important as seeing ALL of the now.
I opened this page and I got a Microsoft.NET ad in the middle of the story. I think Slashdot needs some psychotherapy...it is a touch schitzo.
Re:...isn't this type of thing still legal in the
on
Shake-up At SonicBlue
·
· Score: 3, Insightful
But I think that even the minority shareholders have the right to sue for damages, if only for not having the same loan "opportunities" made available to them. The board is reponsible to represent the best interest of ALL of the shareholders, not just their buddies.
It looks very very bad. Any judge/jury would be sympathetic. It just takes an excuse of a lawsuit and the board will be wiped out. The real problem is for employees of that company. How would you like to find out that the captain of your ship is drilling holes in the hull?
But, the board has civil liability. There is a mechanism in place to resolve this. The shareholders can and should sue them for far more than they stole.
So, what should the government do to correct this? Make all procedings of the board open to shareholders? Require extensive reports of corporate financial activity? Require board members to be elected by the shareholders? Guess what...they already do? So, my point still stands. What should change? In what sense should this be "illegal"?
Knee-jerk responses such as your don't really deserve a response, but I was feeling generous.
I don't disagree with this being corruption. I think that it is bad business practice and probably illegal. The board of directors is tasked with representing the interests of the share holders. At the very least, the shareholder have the basis for a civil case against the board in this matter.
But I don't see the value of government intervention. Perhaps I would be one of the free-marketeers you would seek to bludgeon. In my mind, companies are responsible to their customers and to their shareholders. They must product safe and worthwhile products to deliver to their customers, or there will be no more customers. They must deliver profits to their shareholders, or they will have no shareholders and hence no capital or corporate value.
The downturn that is underway is the market's fault. Everyone who has money in a 401k mutual fund is in some sense responsible. Just like third-world countries that suffer from hyperinflation. Who causes hyperinflation...they all do. Value is always based on perception at some level. The value of these companies were misstated and we believed them. The longterm value of Internet stocks were misstated (mostly in good faith) and the market believed them. When the true value of each came to light, then everyone cries. Who is creating this awful market. We are.
So, yes, people need to go jail or be sued out of existence over all of this. Shareholders must demand accountability. But, I don't see how the government can "fix" any of this beyond insisting on fair and consistent public account practices.
People need to know what they are investing in. They need to know the risks. The individuals within companies that lie and defraud must be help accountable.
It is not quite that simple. No lending agency in the world is going to give you a line of credit or venture capital without some sort of financial guarantee. When I got the line of credit for my company, I had to put my house up as collateral as well as have justification (in the form of past and present POs) for the loan amount.
This situation is different. SonicBlue seems to have had cash reserves and/or available credit and elected to make unguaranteed loans to shareholders. It looks very shady and might be illegal, at least from a tax accounting point of view. But, you do need to have an established business with cash or at least a great line of BS to feed VCs to get them to front you the money.
Slashdot Mirror
Oh, but it is true. The system was rebooted and sat at the raid card bios screen for over 24 hours while the raid parity data was rebuilt. In Linux, the system will reboot in a few minutes (mostly depending on whether or not it needs to fsck). The raid array will be online but parity data will reconstruct in the background.
The biggest headache with these arrays is the fact that they can be so big and so very cheap. I built a 1-TB rackmount system for a client over a year ago for about $8k. The same unit now would be about $4k. But there is no way in the world to back them up. A tape juke will cost about twice what the array does! So, we just build a second mirror raid array on a separate box, tie the two together with point-to-point gigabit, and cron an rsync every evening at 1 AM. It works great and we get the extra protection against accidental deletion and file corruption that comes with tape backups...
I think that the era of tape backups is past...at least for budget archiving of large data storgae.
I have about 5 TBs of RAID5 storage online at various customer sites. They are all using Linux software RAID and Promise ATA66/100/133 controllers. Even when using two drives per IDE channel, we still see very good performance. An RAID5 system with eight 120-GB 5400-RPM Maxtor drives gives about 55 MB/sec write and 80 MB/sec read performance under Bonnie. Those eight drives were on two Promise ATA100 controllers. Cabling is fairly easy if you use 24" UltraATA cables. And it will get much easier with Serial ATA.
One customer ordered a system from a vendor who insisted on installing an ATA raid card, and it was a remarkable disappointment. Linux was able to indentify the array as a SCSI device and mount it. Then, for some reason, the customer rebooted his system. During the BIOS detection, the raid card started doing parity reconstruction and ran for over 24 hours before finally allowing the system to boot! For comparison, the same sized array would resync in the background under Linux in about 3 hours.
Also, the reconstruction tools built into the raid cards are pretty limited. If you have a problem with a Linux software RAID array, at least you can use the normal low level tools to access the drives and try to diagnose the problems. Just MO.
I laughed out loud...bravo!
Probably not.
Assume 100,000 miles over the lifetime of the car, $1.50 per gallon for gas. At 50 MPG, you would spend $3000 on gas over the life of the car. At 40 MPG, it would be $3750. At 30 MPG, $5000. So, that is only a $2000 savings compared to $3000 incremental cost.
Of course, if you double fuel costs, the economics change. Also, if you plan on getting 250,000 miles from your car, it also changes. At the very least, I would say that there is no economic advantage to higher efficiency systems like this. This of course ignores the costs of "environmental impact" which are very difficult to quantify. The key issues is that it makes technological sense and the economics are not too bad.
There is one important point that seems to be often missed. Once a manufacturing process is worked out for the fibers, it will start to become much cheaper. The big cost is hauling all of the first one up into orbit at the beginning.
Once you have ONE of these in orbit, it will be fairly easy to spawn a second or to really strengthen the first. You can tow small fibers of additional nanotubes as reinforcement to the existing ribbon. This is when things can get very interesting. You can build an enormously strong elevator on top of this first seminal one. With multiple strong elevators servicing a single space station, the system could become more like a train track than an elevator, with multiple cars going up and down simultaneously.
Multiple elevators would make the entire system more stable and less suspectible to terrorist attack or weather damage. Imagine putting 1000s of tons of material into geosynch orbit per day. Then we can seriously think about wholesale utilization of space and its resources.
Do we know that the ion density in the ionosphere will not be high enough to cause a catastrophic discharge? The typical way to "generate" lightning strikes is to fire a small rocket into a thundercloud with a small wire attached.
The ionosphere is replentished by solar winds so I don't know if depletion is a greater concern than the lightning strike issue. I haven't looked at the numbers though. I may be wrong.
CompSci 202C, I think, at Penn State. Spring 1991. Fortran77 with Applications
Engineers and Scientists do love their Fortran. I thought it was a little clunky as I was using Turbo Pascal before, and it spoiled me I guess.
"There is something to be said for maturity and stability in a language."
This is quite true, but there is a huge issue with functionality, stabilit, and performance of the compilers. The Fortran standard is fairly simple, so you would think that compiler makers could get it right. But, we have found that EVERY Linux Fortran compiler revision from EVERY vendor (except maybe Intel) introduces new glitches. There is no gold standard under Linux, though the Intel compiler might become one in another few years. All of the Fortran90+ compilers are closed source. The GCC Fortran compiler is Fortran77 and has no support for things like OpenMP.
The big issue here is lack of market size. The potential market for Fortran compilers is about 1% of that for C or Java. The vendors just don't have enough customers/revenue to provide adequate QA and development, IMO.
Promise UltraATA Controller...$25 for ATA100, $40 for ATA133.
I have seen VGA-out boxes for both Xbox and PS2 on www.lik-sang.com. I didn't include an href because the site is down. I have been considering these (with a KVM switch) as a nice way to interface with a front-projector display. Certainly seems cleaner than component-to-VGA-to-DVI rigs.
Has anyone tried these VGA breakout boxes?
I agree. After changing to S-video/component instead of composite, the single biggest improvement in visual quality on large television sets is progressive scan or at least line-doubling. My 27" and 36" tube TVs both show serious "tearing" with DVDs and PS2/Xbox games. Unfortunately, neither support 480p. The examples of 480p output on similarly sized screens made a rather dramatic improvement in picture stability, especially when viewed from only a few feet away.
Why not place the PLLs on every other bank of normal DDR slots on a motherboard. Banks 0 and 2 would be normal. Banks 1 and 3 would be 90-degrees off. If you want double bandwidth, install DIMMs in pairs like on normal interleaving motherboards. Trace complexity would remain unchanged, motherboards would only need a few PLLs inlined, and we could all get twice the bandwidth using $40 DDR DIMMs...and not have to wait for new memory modules to drop in price.
Just a thought.
It shouldn't surprise us that nuclear explosions can be shaped. The "implosion" triggering system used in Fat Man and most subsequent nuclear devices made use of shaped chemical explosions, as do most anti-armor munitions. These work by the focusing of shock waves through careful shaping of the explosive material, precise denotation time and locations, and "lens" of varying composition to focus the shockwaves.
By constructing the explosive device to reinforce that shockwave, a significant fraction of the released energy can be focused in one direction. By carefully focusing the physical shock, the neutron flux, and x-ray release, it should be possible to build a nuclear explosion that goes "that-a-way" rather than just spherically outward.
No, it is not only a stability issue. The strength of the shock determines the pressure jump. The point is to use the shock as a pressurization mechanism, rather than a spinning compressor section as in a normal turbofan engine. There is certain level of pressurization that needs to occur before this is effective. Just like an IC engine, if you cannot get a high enough compression ratio, you cannot have an efficient engine. In this case, lack of sufficient pressurization can cause reversal of the combusted material back through the intake rather than expanding and accelerating out the nozzle to the rear. This is a problem for both ramjet and scramjet engines
AFAIK, stability is the primary issue in moving from ramjet to scramjet. Supersonic combustion, acoustic coupling, etc. can make for lots of stability problems. Getting good mixing of air and fuel prior to the combustion/shock surface is critical and not all that easy, considering the high inlet velocities.
Also, the presence of a compressor section tends to lead to predictable inlet conditions for the combustor section because effects of turbulence at the intake are muted. For a scramjet, there is no similar buffer, so inlet turbulence adds even more perturbations to the shock/combustion surface. Right now, we have no control systems that can actively control turbulence. This is just a very hard engineering problem...deceptively so even. At least on paper, scramjets are nothing more than a carefully shaped duct and some fuel jets.
IAAMechanicalEngineer.
At Mach 20, you are definitely going to have atmospheric ionization. That is going to make your "cruise" missile light up a radar display.
But your point is quite valid. It is the issue of detection versus tracking. If I have listening posts a few hundred km from my borders, I would be able to know that something was coming. If I have a fine grained detection network, I can know a lot about its flight path at the point of detection. But, I am not going to have a lot of time to react if it is travelling at Mach-8.
Although, the missile will need to decelerate as it closes. The cruising altitude is going to be quite high, else the thing will burn up. So, it will have to fall much like a MRV.
OK, I guess you could build a decent missile out of one of these, but I don't see how would be superior to a ICMB or MRBM.
I am beginning to realize that I have read entirely too many Clancy novels. 8)
You missed the biggest reason. Cruise missiles only fly at subsonic speeds by design. They are to be quiet and stealthy. Any supersonic signature is detectable with a simple network of microphones. Once you know it is coming, a cruise missle is easier to shoot down than a jet aircraft as cruise missiles typically don't have active countermeasures and certainly don't have a pilot to perform evasive manuevers.
Um, just as a point of reference, ICBMs travel much faster than Mach 7. On suborbital trajectories from the other side of the globe, you might see a time of flight of 30 minutes or less. Think 20,000 km/hour.
Scramjets are not really interesting as strategic weapons. Extra-atmospheric vehicles (MVRs) are faster and proven 30-year-old tech. Scramjets are going to be useless for cruise missiles, because a Mach-7 shock cone will standout rather nicely even if the missile itself were stealthy. Depending on the altitude, it could also cause ionization of the atmosphere which would show up on radar!
Military applications here are going to be reactive in nature...fighter-bombers that can reach any corner of the globe in two-hours is a big selling point, as is the (literally) stratospheric flight ceilings such crafts would have. But I don't know what form a scramjet-based weapons system might need to take or what niche it might fill.
there will probably be "low-speed" versions for shorter distances
Probably not. To understand why requires some knowledge of how a scramjet differs from a normal turbofan engine. There are no spinning parts in a scramjet or ramjet engine. The (sc)ram engine requires a strong standing shock to me maintained in the intake. This standing shock replaces the compressor section of a normal turbo fan. There is a minimum speed which will produce a sufficiently strong, stable shock that will allow this to work.
The SC part is for supersonic combustion which makes that standing shock also replace the combustor portion of the turbo fan. Chemical reactions and transonic fluid dynamics can interact in very complicated ways. This can make this supersonic combustion unstable. The best way to stabilize it is to go faster and increase the strength of the shock.
So, to sum up, operating scramjets at lower speeds is more difficult, so if anything, we will probably see them operating at the highest possible speeds that the airframe and aerodynamics will allow.
he's "in his 20s". Are you 20 or 29?
This struck me. I am just recently not "in my 20's" anymore. My cousin is an undergrad and he is in his early twenties. There is a huge difference. Undergrad days for me were done with a 286 and then later a very expensive 486. Programming was in TurboPascal, Fortran77, QuickBasic (ICK!), and dBase (dobule ICK!). For him, it is Java, OO-this and that, C++ if he is feeling old-school. So much has changed.
Structural changes have occured in the way that "we do what it is we do." How can anyone have specific longterm goals? I have had some success in the past several years by doing what I am good at, what I enjoy, and chasing after opportunities. I run my own company, but I can honestly say that I don't know what my plans are for 5 years from now, let alone 20. I guess I am not saying anything that you haven't...but let me try.
Planning a carreer is a nonlinear optimization problem. There is a functional that denotes value. It rolls up the financial, social, and intellectual rewards that you get from your carreer. This value functional must be integrated over a path from t=18 years to t=55 years. The path that we take should optimize that total value. But, the value space is highly nonlinear. There will be huge numbers of local extrema...making the best of the current situation. There will be paths that seem to be going nowhere and then lead to huge gains or falls.
So, in handling these types of optimization problems, we need to use common sense (gradient methods) to make the most of what we have at hand. But to keep from being locked into the first local extrema we find, we need to shake things up (entropy methods). Philosophically, I think the same recommendations hold for steering your life and carreer. Choose the best path in front of you, but never stop shaking things up, lest you get stuck.
Seeing the end is not nearly as important as seeing ALL of the now.
I opened this page and I got a Microsoft .NET ad in the middle of the story. I think Slashdot needs some psychotherapy...it is a touch schitzo.
But I think that even the minority shareholders have the right to sue for damages, if only for not having the same loan "opportunities" made available to them. The board is reponsible to represent the best interest of ALL of the shareholders, not just their buddies.
It looks very very bad. Any judge/jury would be sympathetic. It just takes an excuse of a lawsuit and the board will be wiped out. The real problem is for employees of that company. How would you like to find out that the captain of your ship is drilling holes in the hull?
But, the board has civil liability. There is a mechanism in place to resolve this. The shareholders can and should sue them for far more than they stole.
So, what should the government do to correct this? Make all procedings of the board open to shareholders? Require extensive reports of corporate financial activity? Require board members to be elected by the shareholders? Guess what...they already do? So, my point still stands. What should change? In what sense should this be "illegal"?
Knee-jerk responses such as your don't really deserve a response, but I was feeling generous.
I don't disagree with this being corruption. I think that it is bad business practice and probably illegal. The board of directors is tasked with representing the interests of the share holders. At the very least, the shareholder have the basis for a civil case against the board in this matter.
But I don't see the value of government intervention. Perhaps I would be one of the free-marketeers you would seek to bludgeon. In my mind, companies are responsible to their customers and to their shareholders. They must product safe and worthwhile products to deliver to their customers, or there will be no more customers. They must deliver profits to their shareholders, or they will have no shareholders and hence no capital or corporate value.
The downturn that is underway is the market's fault. Everyone who has money in a 401k mutual fund is in some sense responsible. Just like third-world countries that suffer from hyperinflation. Who causes hyperinflation...they all do. Value is always based on perception at some level. The value of these companies were misstated and we believed them. The longterm value of Internet stocks were misstated (mostly in good faith) and the market believed them. When the true value of each came to light, then everyone cries. Who is creating this awful market. We are.
So, yes, people need to go jail or be sued out of existence over all of this. Shareholders must demand accountability. But, I don't see how the government can "fix" any of this beyond insisting on fair and consistent public account practices.
People need to know what they are investing in. They need to know the risks. The individuals within companies that lie and defraud must be help accountable.
It is not quite that simple. No lending agency in the world is going to give you a line of credit or venture capital without some sort of financial guarantee. When I got the line of credit for my company, I had to put my house up as collateral as well as have justification (in the form of past and present POs) for the loan amount.
This situation is different. SonicBlue seems to have had cash reserves and/or available credit and elected to make unguaranteed loans to shareholders. It looks very shady and might be illegal, at least from a tax accounting point of view. But, you do need to have an established business with cash or at least a great line of BS to feed VCs to get them to front you the money.
IANAAccountant.