This work uses the SEI CMM (capability maturity model) developed by SEI and used in many areas within the civilian and military software development community. Organizations are assigned levels with specific tasks, roles and processes defined to advance to the next level. All results that are compared to the CMM levels must be verified, quantified and repeatable. A good place to start if you are interested in how the military develops robust softare is the Software Technology Support Center @ http://stsc.hill.af.mil/
Calling all materials physicists! What would be the net effect of using some other material for the heatsink itself? Aluminum is used because it's cheap not because it's a terribly good heatsink material. Can anyone suggest some other material that some smart lad here could manufacture a heatsink out of and then test and record the results, both from a temperature perspective and from an overclockability perspective?
Other than that Mrs. Kennedy how was the motorcade
on
Boo No More
·
· Score: 1
I got an unsolicited email from Boo a few days ago and decided to check it out. The site was frrrrrrustrating, odd, hard to use and didn't seem to offer any real DEALS, that is, their prices were not much better than I could get down the road in person. Other than doing everything wrong I guess it was an interesting exercise. They obviously targetted high end consumers who would be willing to pay more and put up with significant inconvenience to be early adopters and to be able to tell their friends that they web purchased. In the end something like this CAN work a-la L.L. Bean or Land's End but it has to based on a working business process model that already works. All of the back end stuff has to work flawlessly: logistics, returns, service, billing. If we look forward at the future of B2C what we'd find is that the companies that succeed at it will be the same companies that are otherwise successful at performing these functions in the solid world.
There was a PC370 add on card for MCA bus IBM PS2's. There was an RS370-390 add on card for older RS6000's. Both of these had channel IO emulation. There was an RS6000 add on card for PS2's. There was hybrid 370 machine called the 9370 that came with an Intel add on board - again here an MCA type. The AS400 LAN connect server is an Intel board inside. None of these are real elegant.
A few of you have indicated that this instance would be a guest OS, probably under VM. That fact isn't made clear in the announcement and AFAIK that is not a requirement to run Linux on a S390 server. That is, one of the big hurdles to running Linux on S390 is to NOT run it as a VM guest instead running it native on the HW. The big problem is to support native IO, channels and whatnot. The other problem is that Linux does not support any 'native' S390 FS so you have implement some extensions to the OS in order to overlay a Linux FS on top of the basic FS functions on the S390. This, instead of attempting to create PDS's and so on within Linux. All of this work seems to be a follow on to work directed at doing more or less the same thing for AIX on S390 which has been around for several years. In fact OS/390 pretty much IS a Unix variant running on S390 - at least it's certified that way.
BoA & its lawyers were terrified of being able to hijack their URL, pass it through Dialect and have the results returned to some unsuspecting customer in all of its politically incorrect-ositude. They don't want the heat, can't claim they didn't know anything about it and would prefer not to make a headline on CNN about it. Self righteous or not you should expect this sort of thing by now.
....In the argument here and what you have is pretty much the normal day to day operations of any squalid, corrupt, war-torn emerging dictatorship. We hear as much coming out of Sierra Leone or Nigeria eg. "we need to restrict the disruptive elements in our society if there is to be any hope of calling democratic elections". Oh yeah, sure, and all that mayhem, repression and killing, pay no attention to it, ahem.
Once you have the police in control of the media and/or the phones you've fallen into the abyss. Plain and simple.
The graphics card market is breaking into 2 segments. The office/home segment where most of the heavy lifting is going to be handled by the CPU and where the 'graphics card' is simply more or less a port connector. The other segment is the high end 3D market which not only is very expensive to develop and support but holds a comparitively small market. You see the high end of the market going over the top where cards now have >1 ASIC processor, their own power supplies, special cooling, etc. To say nothing of the large investment in driver software and maintenance. At the low end of the market you basically become just a software company. With cycles to burn, current CPU's can be used for most functions previously relegated to peripherals: modem/DSL/cable, sound & video.
I don't get it. GPL implies restrictions and use w/o royalties while patents imply another concept altogether. What would be the point of GPL'ing a patent unless you wanted to make some vague distinction in the use of the patent under one set of licencees vs. some other set of licencees. So if you use a GPL patent code chunk linked to some other GPL code chunk then ok. But if you link a GPL patent code chunk to some other non GPL code chunk then the patent would require enforable royalties? Do I have it right?
Do I have the names right? This is one of the fictional paradoxes proposed in the recent fictional book on WW2 cryptology. That is - if a problem takes 10 years to solve with current technology and 5 years to solve with the next generation of technology which won't be available for 3 years then you can save 2 years of effort by waiting for 3 years and doing nothing.
Is data gathering eg, steering the 'camera' feasible? Other than the fact that @ 250AU a communications round trip would take ~67hrs does anyone have a rough figure for the amount of power and/or sensitivity a transmitter @ that distance would require for earth bound stations to work with. Would the power requirements negate the design criteria of a light fast sail?
BTW - someone here talked about turning it around and sending it back home. You couldn't do that by stopping it and turning around. You'd need the square of the energy used to get it moving to stop the flyer. I'm not a physicist but it seems that a gentle arc would be a more practical way to turn it around.
In case you didn't cathc this. Yesterday the Justice department announced it had won a suit against most of the major record labels in the US over the prices of CD's. It looks like the record companies were found guilty of price gouging and they will have to repay, to the gov't of course a couple of hundred million dollars. I wonder how this bodes for Napster. I suppose the record labels could appeal on the basis that Napster et.al. force them to overcharge since they are all losing money because of it. Or, it could backlash if people recognize that the existance of Napter and similar technologies is specically BECAUSE of price gouging.
You may have even written a paper yourself about the statistical non correlation between the overall economy and the markets' performance. A bubble is a bubble whether you call it the next technological revolution or latest tulip craze. Does anyone remember the 80's when moribund companies were valued more as dismembered chunks each to be sold off and ultimately dismembered again and sold off again, and so on?
All invesment entails a degree of risk. What IS occuring now is that many companies with no direct experience in the dotcom world are going off and leveraging the hype. Anybody own Starbucks recently? A coffee shop on the net? WTF sense does that make? So in effect what happens is the nominal risk of going off into a new market with a new business model using technology you don't understand well and don't directly control tends to normalize. The risk to an AT&T converges toward that of monkeysoutmyass.com. With valuations this high even the basic dollar value associated with those two risks is roughly the same. So of course the market becomes more volatile. It is experiencing what you statisticians call autocorrelation. That is, my estimate of what a stock will do is based on my estimate of what other analysts think similar risk rated companies will do which is based on what some other analysts think my company will do. Typically wild swings in value are compounded by a lack of understanding whether today's earnings announcement for company A has any bearing at all on company B - - but hey they've both got dotcom in the name so let's treat them the same. Evident this is the importance of 'whisper numbers'. These are earnings and price estimates that are based on unfounded or unattributable information or sources or guesses about guesses about guesses. What you see happening with whisper numbers is that a company will announce projected next quarter earnings, the whisper numbers come in higher or lower. When the time comes for an accounting and that company announces that that quarters earning were in fact what they projected the stock price drops if the whisper number eg. 'the secret real deal' was higher. If the whisper number was lower than what the company publically projected then the market already hammered the price and the announcement of on-target earnings only reaffirms that market actually knew what it was doing in the first place. Rational? No.
Now take the other case, the case of hype over substance. Each investment is evaluated in the context of that set of alternative investments that perform better or worse. For every stock you have to decide whether you think this given stock will outperform any number of other stocks given the investment goals you ascribe to. So if what you want is money market rate + 20% over the next 6 months you determine for yourself whether there is an acceptable likihood of that actually happening. If your goal is instant millionaire then you evaluate that stock in a group of other stocks you think are roughly equal in their ability to make you an instant millionaire. Now as most people know - the bigger you are the harder you hit. So being the first or nearly so in a market even if its only for a few months can give you enormous advantage. With valuations going up this rapidly the first Yahoo is already several billion dollars ahead of the second Yahoo. In fact the difference is so large that the net effect is to crowd out the investment in any other potential competitor. Investing in the second Yahoo is for VC's because for every 2 dozen companies they fund if one pays off even partially like a Yahoo then the loser dogs don't count. IS this rational - actually, yes it is.
Now we move on to third case which is a side effect of the other two. That case is where post IPO companies of different maturities and expertise get dragged in the same direction, usually down, as the rest of the market independant of every other fact. This one is bit more complex because it introduces the effect of institutional investors. Say for example you have two stocks: monkeysoutmyass.com and mail.com. Both went public but one has a rational business model, good management, real earnings growth, generally good relationships with customers while the other company really is 5 guys and a business plan on a bar napkin, a lot of overtime, pursuasive pitches to the analysts and some seed capital. An announcement goes out about some market or sector or company not directly related to this but probably has some tangential effect. Institutional investors decide to rejuggle the portfolios because GiantHardwareCompany.com - a major player and someone who should know announces that next quarter earning will be lower than the previous quarter. Not lower than expected mind you - because this IS the expectation, but just lower, slightly. The reason could be something as obscure as difficulty in procuring controller chips from the one Taiwanese factory that just fell into the ocean from the last earthquake to something closer to home like consumers don't like or understand your recent marketing push to something mundane <and unstated> like our R+D head is going through a midlife crisis and cashed out and moved to Monte Carlo....well you get the point. Immediately the institutional investors pick from their category menus because making individual decisions about thousands of stocks is impossible. So small portions of whole sectors go into the sell basket driving the price down. Ergo your two companies described above both drop below the IPO price for no obvious reason. IS this rational? Sort of; at least it's predictable.
Is the market full of hype, hot air, ignorance, herd mentality, misinformation? Yeah sure. And as long as banks are paying 2.2% on CD's while charging 8-9% on long term consumer debt people will continue to invest in the market because any other option is a GUARANTEED net loss factoring in the CPI and taxes.
I was recently at an intellectual property conference and the two things that shocked me were:
Anything is patentable or at least submittable as a patent. Depending on how you word it even if there may be some issues with prior art, if you describe even a business process, then tat can be patented.
You don't actually need finished product to patent it. You can describe how the code might work. You don't actually need working code to patent what the code does.
This goes to the heart of why this is such a hot topic now. You get a bunch of smart people to describe some of the ideas and solutions they've come up with over the years and patent them. Then you get a bunch of sharp people to scour the landscape and identify something that knowingly or unknowingly uses substantial aspects of your patent. Lastly you go to them with your hand out requesting royalties.
SCO like any other OS company from way-back-when was/is in the business of selling OS code, utilities, some middleware & service. They made some efforts to accommodate third parties and OEMs for utilities and other ware that was either better or more accepted in the marketplace, generally a side effect of being the cheap Xnix alternative for Intel the application vendors actually wrote code for in the late 80's, early 90's. There was no useful NT/2000 SCO could afford to charge lots of money. As the market moved onward to other xNixen, Novell and the like SCO recognized that glucode, directories, filesystems and other infrastructure components were going to be the only avenue for revenue in the future. Look for example at Novell which is now pretty much the NDS company and Banyan which is also just a directory vendor. So what you end up with is SCO, a company with a strong following, fairly good OEM relations, ownership of a great deal of good technology such as SVR4 and a large installed base of small-midsize vertical business applications such as a big piece of the medical office management market. If they read the writing on the wall correctly they'll see that they should start trying to move that space into maintenance mode and work with applications vendors to port this stuff to Linux/Freenix/BSD what-have-you. Beef up the services arm and third party service partners. You see, now that for application vendors and customers in this space W2K is a viable alternative there is no incentive for SCO to charge anything for the xNix OS code itself and instead focus on solid infrastructure components, well built applications and service and maintenance. As almost a seperate function SCO can continue to develop enhancements for xNix and then decide if they can reasonably open source them or not. There are quite a few things that SCO is in a good position to develop for xNix such as better authentication and cleaner user administration. How about distributed directory services for xNix, or enhancements to Samba, or backup/storage management facilities. What about something simple like more robust terminal servers for the thousands of character terminal application installations out there?
WvB was not involved in the Manhattan project as he was still an active Nazi working in Holland to build V2's. But otherwise yes he was the NASA director during the Saturn V effort.
There is a Swiss company that makes explosive CD's (I think posted here !!) So you could use an inverted failsafe code. That is, start the machine and require the correct code else an autoload program triggers the cd to self destruct. Sure you'd have to change the engineering some like building a tiny minidisc and sealing it so it can't be removed from the machine or is packaged with the hard drive.
BTW hardrive encryption works pretty well. My Thinkpad has an option to encrypt the hardrive and while they don't publish the algorithm we've never seen a case where someone has to managed to break it. This is not to say it's unbreakable but seemingly no one's succeeded or bothered to try.
Actually most of the Nazi scientists that formed the core of the American space program were hurried out of Germany at the end of the war and hidden away in military projects in the US. This was called 'Project Paperclip'. The reason they were hidden away was not because we were afraid that the Russians would get them. Instead it was because we KNEW they were actively being sought by Allied agencies for prosecution as war criminals for their activities during the war. It was a well known open secret that von Braun and company were ardent, respected, honored and low numbered Nazi party members, that they made extensive use of slave labor in camp Dora and other places, that the entire Nazi space medicine program was founded on torture and murder of victims in the camps and that the US gov't under Paperclip which was a joint project of the CIA and DIA was hiding these people. In several instances Paperclip arranged to 'kidnap' back several scientists who had actually been captured and detained in the US and overseas.
No really, punning aside if I take the screen, keyboard and other 'external' junk off my IBM Thinkpad 600E what I'm left with could probably be packaged inside a mid-sized cell phone. With a little ingenuity it could be crammed inside a package small enough to be just a fat spot in the power cord. What good is a tiny PC if you have to rely on someone leaving a monitor and a keyboard laying around?
Are there even 3744 useful productive hours in a 4 year degree or even a BS+MS? When I went to school you got a major a minor and filled out the other 40-50% with fluff. If we compare this to a typical I'net consulting engagement @ 80% utilization eg. billable hours then you're talking about a 90 hour work week for 1 solid year. If even 10% of all people can maintain that then they are probably already working @ a dotcom startup or Bain or McKinsey and being paid in the high 5's low 6's. If they aren't then when they finish the first question will be "Well why didn't you just start your own company?"
Most PCs are manufactured in Taiwan, SE Asia, Mexico and E. Europe. Labor is cheaper because they pay less and have fewer benefits and generally have less restrictive labor laws. Moreover the plant itself is probably subject to fewer environmental and safety regulations. The countries they are located in do not have strong labor movements or a culture of litigation. The only exceptions to this scenario would be a government mandated direct intervention, for example an attempt to alleviate high unemployment in Ireland and Scotland and the consequent tax breaks that flow from any large company moving in to hire young well educated people. It's really a synthetic case because the company's addtional cost burden of doing business in an industrialized regulated country; eg. Western, is born by the taxpayers as the cost of keeping people employed. Why else is there such a stink in the US over low-middle tech manufacturing sites just on the other side of the Mexican border being given special status under the old NAFTA? It's the best of both worlds from the company's perspective. Cheap local labor, weak regulation and a close proximity to the advanced technology, distribution and financial infrastructures in the US.
Well actually what I have is 2 IBM Netfinity 3500-M10 servers with a quarter gig ram and 3x9gig SCSI3 drives off a 2940U2W adapter, each. They don't officially support OC but can be made to work with some careful adjustments. My point is that upgrading now involves 2xcost of whatever Pent. you want to put in here vs. 1xcost of whatever you put in here. That's the hurdle rate eg. the least granular cost is 2x. You will always have to purchase two of something to continue to upgrade. Yes yes we all agree that 2x something is faster than 1x something but the cost committment is twice as large vs. using a single CPU machine, putting in a fast processor, overclocking it to say 70-75% of the raw combined speed of an SMP machine, running it until you want to either replace THAT with another CPU and/or something breaks/fries. With the money you save you could buy some solid state disk and blow the wheels off most anything.
SMP rarely if ever makes sense for most uses if you're using your own money. I have a couple of state-of-the-art P2-233 2way machines circa late 1998 that are now way too expensive to upgrade. At the time if you needed to be @ the edge of performance this is how you did it. But, and this is the problem, if I had waited maybe 2 months I wouldn't have had to get an SMP at all. You don't get any price/peformance improvement trying to upgrade an SMP box; either you have to get 2 less-than-the-latest CPUs for a dollar premium or you have to toss the MB and start over. Overall your best bet is to get the cheapest CPU, while spending some amount of money to overclock it a certain amount, certainly less than the absolutely achievable max speed since cooling and retro-engineering will cost too much. Then run that CPU as fast as possible until it fries and replace it with the cheapest CPU you can REASONABLY overclock at that point. Heck if you melt the MB or anything else just get another one since you don't have to pay the premium for SMP MBs or worry about drivers and whatnot. In real world tests of interactive performance people generally can't tell the difference between +- 15% and generally don't experience any benefit until +20% performance. So in the end the best bet is to effectively upgrade CPU performance to only 80% of the max achievable.
If I'm not mistaken E.D.'s latest incarnation is as a VC . Isn't this a conflict of interest or do we not care about who's interest is served in the standards process anymore.If I'm wrong and E.D. is not a VC and doesn't have a stake in seeing that some standards are more standard than others then I retract everything.
Slashdot Mirror
This work uses the SEI CMM (capability maturity model) developed by SEI and used in many areas within the civilian and military software development community. Organizations are assigned levels with specific tasks, roles and processes defined to advance to the next level. All results that are compared to the CMM levels must be verified, quantified and repeatable. A good place to start if you are interested in how the military develops robust softare is the Software Technology Support Center @ http://stsc.hill.af.mil/
Calling all materials physicists! What would be the net effect of using some other material for the heatsink itself? Aluminum is used because it's cheap not because it's a terribly good heatsink material. Can anyone suggest some other material that some smart lad here could manufacture a heatsink out of and then test and record the results, both from a temperature perspective and from an overclockability perspective?
I got an unsolicited email from Boo a few days ago and decided to check it out. The site was frrrrrrustrating, odd, hard to use and didn't seem to offer any real DEALS, that is, their prices were not much better than I could get down the road in person. Other than doing everything wrong I guess it was an interesting exercise. They obviously targetted high end consumers who would be willing to pay more and put up with significant inconvenience to be early adopters and to be able to tell their friends that they web purchased. In the end something like this CAN work a-la L.L. Bean or Land's End but it has to based on a working business process model that already works. All of the back end stuff has to work flawlessly: logistics, returns, service, billing. If we look forward at the future of B2C what we'd find is that the companies that succeed at it will be the same companies that are otherwise successful at performing these functions in the solid world.
There was a PC370 add on card for MCA bus IBM PS2's. There was an RS370-390 add on card for older RS6000's. Both of these had channel IO emulation. There was an RS6000 add on card for PS2's. There was hybrid 370 machine called the 9370 that came with an Intel add on board - again here an MCA type. The AS400 LAN connect server is an Intel board inside. None of these are real elegant.
A few of you have indicated that this instance would be a guest OS, probably under VM. That fact isn't made clear in the announcement and AFAIK that is not a requirement to run Linux on a S390 server. That is, one of the big hurdles to running Linux on S390 is to NOT run it as a VM guest instead running it native on the HW. The big problem is to support native IO, channels and whatnot. The other problem is that Linux does not support any 'native' S390 FS so you have implement some extensions to the OS in order to overlay a Linux FS on top of the basic FS functions on the S390. This, instead of attempting to create PDS's and so on within Linux. All of this work seems to be a follow on to work directed at doing more or less the same thing for AIX on S390 which has been around for several years. In fact OS/390 pretty much IS a Unix variant running on S390 - at least it's certified that way.
BoA & its lawyers were terrified of being able to hijack their URL, pass it through Dialect and have the results returned to some unsuspecting customer in all of its politically incorrect-ositude. They don't want the heat, can't claim they didn't know anything about it and would prefer not to make a headline on CNN about it. Self righteous or not you should expect this sort of thing by now.
....In the argument here and what you have is pretty much the normal day to day operations of any squalid, corrupt, war-torn emerging dictatorship. We hear as much coming out of Sierra Leone or Nigeria eg. "we need to restrict the disruptive elements in our society if there is to be any hope of calling democratic elections". Oh yeah, sure, and all that mayhem, repression and killing, pay no attention to it, ahem.
Once you have the police in control of the media and/or the phones you've fallen into the abyss. Plain and simple.
The graphics card market is breaking into 2 segments. The office/home segment where most of the heavy lifting is going to be handled by the CPU and where the 'graphics card' is simply more or less a port connector. The other segment is the high end 3D market which not only is very expensive to develop and support but holds a comparitively small market. You see the high end of the market going over the top where cards now have >1 ASIC processor, their own power supplies, special cooling, etc. To say nothing of the large investment in driver software and maintenance. At the low end of the market you basically become just a software company. With cycles to burn, current CPU's can be used for most functions previously relegated to peripherals: modem/DSL/cable, sound & video.
I don't get it. GPL implies restrictions and use w/o royalties while patents imply another concept altogether. What would be the point of GPL'ing a patent unless you wanted to make some vague distinction in the use of the patent under one set of licencees vs. some other set of licencees. So if you use a GPL patent code chunk linked to some other GPL code chunk then ok. But if you link a GPL patent code chunk to some other non GPL code chunk then the patent would require enforable royalties? Do I have it right?
Do I have the names right? This is one of the fictional paradoxes proposed in the recent fictional book on WW2 cryptology. That is - if a problem takes 10 years to solve with current technology and 5 years to solve with the next generation of technology which won't be available for 3 years then you can save 2 years of effort by waiting for 3 years and doing nothing.
Is data gathering eg, steering the 'camera' feasible? Other than the fact that @ 250AU a communications round trip would take ~67hrs does anyone have a rough figure for the amount of power and/or sensitivity a transmitter @ that distance would require for earth bound stations to work with. Would the power requirements negate the design criteria of a light fast sail?
BTW - someone here talked about turning it around and sending it back home. You couldn't do that by stopping it and turning around. You'd need the square of the energy used to get it moving to stop the flyer. I'm not a physicist but it seems that a gentle arc would be a more practical way to turn it around.
In case you didn't cathc this. Yesterday the Justice department announced it had won a suit against most of the major record labels in the US over the prices of CD's. It looks like the record companies were found guilty of price gouging and they will have to repay, to the gov't of course a couple of hundred million dollars. I wonder how this bodes for Napster. I suppose the record labels could appeal on the basis that Napster et.al. force them to overcharge since they are all losing money because of it. Or, it could backlash if people recognize that the existance of Napter and similar technologies is specically BECAUSE of price gouging.
You may have even written a paper yourself about the statistical non correlation between the overall economy and the markets' performance. A bubble is a bubble whether you call it the next technological revolution or latest tulip craze. Does anyone remember the 80's when moribund companies were valued more as dismembered chunks each to be sold off and ultimately dismembered again and sold off again, and so on?
All invesment entails a degree of risk. What IS occuring now is that many companies with no direct experience in the dotcom world are going off and leveraging the hype. Anybody own Starbucks recently? A coffee shop on the net? WTF sense does that make? So in effect what happens is the nominal risk of going off into a new market with a new business model using technology you don't understand well and don't directly control tends to normalize. The risk to an AT&T converges toward that of monkeysoutmyass.com. With valuations this high even the basic dollar value associated with those two risks is roughly the same. So of course the market becomes more volatile. It is experiencing what you statisticians call autocorrelation. That is, my estimate of what a stock will do is based on my estimate of what other analysts think similar risk rated companies will do which is based on what some other analysts think my company will do. Typically wild swings in value are compounded by a lack of understanding whether today's earnings announcement for company A has any bearing at all on company B - - but hey they've both got dotcom in the name so let's treat them the same. Evident this is the importance of 'whisper numbers'. These are earnings and price estimates that are based on unfounded or unattributable information or sources or guesses about guesses about guesses. What you see happening with whisper numbers is that a company will announce projected next quarter earnings, the whisper numbers come in higher or lower. When the time comes for an accounting and that company announces that that quarters earning were in fact what they projected the stock price drops if the whisper number eg. 'the secret real deal' was higher. If the whisper number was lower than what the company publically projected then the market already hammered the price and the announcement of on-target earnings only reaffirms that market actually knew what it was doing in the first place. Rational? No.
Now take the other case, the case of hype over substance. Each investment is evaluated in the context of that set of alternative investments that perform better or worse. For every stock you have to decide whether you think this given stock will outperform any number of other stocks given the investment goals you ascribe to. So if what you want is money market rate + 20% over the next 6 months you determine for yourself whether there is an acceptable likihood of that actually happening. If your goal is instant millionaire then you evaluate that stock in a group of other stocks you think are roughly equal in their ability to make you an instant millionaire. Now as most people know - the bigger you are the harder you hit. So being the first or nearly so in a market even if its only for a few months can give you enormous advantage. With valuations going up this rapidly the first Yahoo is already several billion dollars ahead of the second Yahoo. In fact the difference is so large that the net effect is to crowd out the investment in any other potential competitor. Investing in the second Yahoo is for VC's because for every 2 dozen companies they fund if one pays off even partially like a Yahoo then the loser dogs don't count. IS this rational - actually, yes it is.
Now we move on to third case which is a side effect of the other two. That case is where post IPO companies of different maturities and expertise get dragged in the same direction, usually down, as the rest of the market independant of every other fact. This one is bit more complex because it introduces the effect of institutional investors. Say for example you have two stocks: monkeysoutmyass.com and mail.com. Both went public but one has a rational business model, good management, real earnings growth, generally good relationships with customers while the other company really is 5 guys and a business plan on a bar napkin, a lot of overtime, pursuasive pitches to the analysts and some seed capital. An announcement goes out about some market or sector or company not directly related to this but probably has some tangential effect. Institutional investors decide to rejuggle the portfolios because GiantHardwareCompany.com - a major player and someone who should know announces that next quarter earning will be lower than the previous quarter. Not lower than expected mind you - because this IS the expectation, but just lower, slightly. The reason could be something as obscure as difficulty in procuring controller chips from the one Taiwanese factory that just fell into the ocean from the last earthquake to something closer to home like consumers don't like or understand your recent marketing push to something mundane <and unstated> like our R+D head is going through a midlife crisis and cashed out and moved to Monte Carlo....well you get the point. Immediately the institutional investors pick from their category menus because making individual decisions about thousands of stocks is impossible. So small portions of whole sectors go into the sell basket driving the price down. Ergo your two companies described above both drop below the IPO price for no obvious reason. IS this rational? Sort of; at least it's predictable.
Is the market full of hype, hot air, ignorance, herd mentality, misinformation? Yeah sure. And as long as banks are paying 2.2% on CD's while charging 8-9% on long term consumer debt people will continue to invest in the market because any other option is a GUARANTEED net loss factoring in the CPI and taxes.
I was recently at an intellectual property conference and the two things that shocked me were:
Anything is patentable or at least submittable as a patent. Depending on how you word it even if there may be some issues with prior art, if you describe even a business process, then tat can be patented.
You don't actually need finished product to patent it. You can describe how the code might work. You don't actually need working code to patent what the code does.
This goes to the heart of why this is such a hot topic now. You get a bunch of smart people to describe some of the ideas and solutions they've come up with over the years and patent them. Then you get a bunch of sharp people to scour the landscape and identify something that knowingly or unknowingly uses substantial aspects of your patent. Lastly you go to them with your hand out requesting royalties.
SCO like any other OS company from way-back-when was/is in the business of selling OS code, utilities, some middleware & service. They made some efforts to accommodate third parties and OEMs for utilities and other ware that was either better or more accepted in the marketplace, generally a side effect of being the cheap Xnix alternative for Intel the application vendors actually wrote code for in the late 80's, early 90's. There was no useful NT/2000 SCO could afford to charge lots of money. As the market moved onward to other xNixen, Novell and the like SCO recognized that glucode, directories, filesystems and other infrastructure components were going to be the only avenue for revenue in the future. Look for example at Novell which is now pretty much the NDS company and Banyan which is also just a directory vendor. So what you end up with is SCO, a company with a strong following, fairly good OEM relations, ownership of a great deal of good technology such as SVR4 and a large installed base of small-midsize vertical business applications such as a big piece of the medical office management market. If they read the writing on the wall correctly they'll see that they should start trying to move that space into maintenance mode and work with applications vendors to port this stuff to Linux/Freenix/BSD what-have-you. Beef up the services arm and third party service partners. You see, now that for application vendors and customers in this space W2K is a viable alternative there is no incentive for SCO to charge anything for the xNix OS code itself and instead focus on solid infrastructure components, well built applications and service and maintenance. As almost a seperate function SCO can continue to develop enhancements for xNix and then decide if they can reasonably open source them or not. There are quite a few things that SCO is in a good position to develop for xNix such as better authentication and cleaner user administration. How about distributed directory services for xNix, or enhancements to Samba, or backup/storage management facilities. What about something simple like more robust terminal servers for the thousands of character terminal application installations out there?
WvB was not involved in the Manhattan project as he was still an active Nazi working in Holland to build V2's. But otherwise yes he was the NASA director during the Saturn V effort.
There is a Swiss company that makes explosive CD's (I think posted here !!) So you could use an inverted failsafe code. That is, start the machine and require the correct code else an autoload program triggers the cd to self destruct. Sure you'd have to change the engineering some like building a tiny minidisc and sealing it so it can't be removed from the machine or is packaged with the hard drive.
BTW hardrive encryption works pretty well. My Thinkpad has an option to encrypt the hardrive and while they don't publish the algorithm we've never seen a case where someone has to managed to break it. This is not to say it's unbreakable but seemingly no one's succeeded or bothered to try.
Because as we all know you can't shut off an SRB. So there must have a liquid fuel component or the story is incorrect.
Actually most of the Nazi scientists that formed the core of the American space program were hurried out of Germany at the end of the war and hidden away in military projects in the US. This was called 'Project Paperclip'. The reason they were hidden away was not because we were afraid that the Russians would get them. Instead it was because we KNEW they were actively being sought by Allied agencies for prosecution as war criminals for their activities during the war. It was a well known open secret that von Braun and company were ardent, respected, honored and low numbered Nazi party members, that they made extensive use of slave labor in camp Dora and other places, that the entire Nazi space medicine program was founded on torture and murder of victims in the camps and that the US gov't under Paperclip which was a joint project of the CIA and DIA was hiding these people. In several instances Paperclip arranged to 'kidnap' back several scientists who had actually been captured and detained in the US and overseas.
No really, punning aside if I take the screen, keyboard and other 'external' junk off my IBM Thinkpad 600E what I'm left with could probably be packaged inside a mid-sized cell phone. With a little ingenuity it could be crammed inside a package small enough to be just a fat spot in the power cord. What good is a tiny PC if you have to rely on someone leaving a monitor and a keyboard laying around?
Are there even 3744 useful productive hours in a 4 year degree or even a BS+MS? When I went to school you got a major a minor and filled out the other 40-50% with fluff. If we compare this to a typical I'net consulting engagement @ 80% utilization eg. billable hours then you're talking about a 90 hour work week for 1 solid year. If even 10% of all people can maintain that then they are probably already working @ a dotcom startup or Bain or McKinsey and being paid in the high 5's low 6's. If they aren't then when they finish the first question will be "Well why didn't you just start your own company?"
Most PCs are manufactured in Taiwan, SE Asia, Mexico and E. Europe. Labor is cheaper because they pay less and have fewer benefits and generally have less restrictive labor laws. Moreover the plant itself is probably subject to fewer environmental and safety regulations. The countries they are located in do not have strong labor movements or a culture of litigation. The only exceptions to this scenario would be a government mandated direct intervention, for example an attempt to alleviate high unemployment in Ireland and Scotland and the consequent tax breaks that flow from any large company moving in to hire young well educated people. It's really a synthetic case because the company's addtional cost burden of doing business in an industrialized regulated country; eg. Western, is born by the taxpayers as the cost of keeping people employed. Why else is there such a stink in the US over low-middle tech manufacturing sites just on the other side of the Mexican border being given special status under the old NAFTA? It's the best of both worlds from the company's perspective. Cheap local labor, weak regulation and a close proximity to the advanced technology, distribution and financial infrastructures in the US.
Well actually what I have is 2 IBM Netfinity 3500-M10 servers with a quarter gig ram and 3x9gig SCSI3 drives off a 2940U2W adapter, each. They don't officially support OC but can be made to work with some careful adjustments. My point is that upgrading now involves 2xcost of whatever Pent. you want to put in here vs. 1xcost of whatever you put in here. That's the hurdle rate eg. the least granular cost is 2x. You will always have to purchase two of something to continue to upgrade. Yes yes we all agree that 2x something is faster than 1x something but the cost committment is twice as large vs. using a single CPU machine, putting in a fast processor, overclocking it to say 70-75% of the raw combined speed of an SMP machine, running it until you want to either replace THAT with another CPU and/or something breaks/fries. With the money you save you could buy some solid state disk and blow the wheels off most anything.
SMP rarely if ever makes sense for most uses if you're using your own money. I have a couple of state-of-the-art P2-233 2way machines circa late 1998 that are now way too expensive to upgrade. At the time if you needed to be @ the edge of performance this is how you did it. But, and this is the problem, if I had waited maybe 2 months I wouldn't have had to get an SMP at all. You don't get any price/peformance improvement trying to upgrade an SMP box; either you have to get 2 less-than-the-latest CPUs for a dollar premium or you have to toss the MB and start over. Overall your best bet is to get the cheapest CPU, while spending some amount of money to overclock it a certain amount, certainly less than the absolutely achievable max speed since cooling and retro-engineering will cost too much. Then run that CPU as fast as possible until it fries and replace it with the cheapest CPU you can REASONABLY overclock at that point. Heck if you melt the MB or anything else just get another one since you don't have to pay the premium for SMP MBs or worry about drivers and whatnot. In real world tests of interactive performance people generally can't tell the difference between +- 15% and generally don't experience any benefit until +20% performance. So in the end the best bet is to effectively upgrade CPU performance to only 80% of the max achievable.
If I'm not mistaken E.D.'s latest incarnation is as a VC . Isn't this a conflict of interest or do we not care about who's interest is served in the standards process anymore.If I'm wrong and E.D. is not a VC and doesn't have a stake in seeing that some standards are more standard than others then I retract everything.