It relies upon water under high pressure, flowing through the microchannels coated with surface charge to generate some power.
The exact same highly-pressurized water could be used to drive a turbine connected to a generator.
It will be a matter of whether the efficiency and costs of the new device can be developed to be competitive with conventional hydroelectric power.
My own favorite untapped technology for power generation is taking advantage of the mixing of fresh water and salt water that occurs where rivers flow into the ocean.
There's a tremendous source of untapped energy (look at how much power has to be put into seawater purification to get some idea.)
Perhaps this same microchannel technology could be adapted to harness energy from the mixing of salt and fresh water.
Technically, the QoS (latency) over the low quality lines, seems something of an issue. I read somewhere that anything more than 0.1 sec latency makes conversation noticeably less spontaneous. Maybe it will encourage us all to listen to others without interrupting them:)
Politically, the regulated monopolies that provide local telephone service will be difficult to interface with. They'll demand money for access to the offices (some people have stories about this), but are also obligated to maintain lines and obey a slew of regulations, such as providing service to backwater areas, emergency service, tacking on a whole zoo of fees and taxes.
A Blanket policy against MS, without allowing for a competitive bidding process or even alternative analysis doesn't seem right.
Yes, you're correct. The decision should be based on cost analysis.
Unfortunately, most CIO's won't do a complete cost analysis.
There are costs associated with both Microsoft technology and costs associated with open source software that typically are ignored.
In the case FOSS, what's typically unaccounted are the training and migration costs for a workforce that is largely familiar with Microsoft's way of doing things.
In the case of Microsoft, what's typically unaccounted are the costs of lock-in from using interdependent Microsoft technology and being put on a forced upgrade path that requires hardware upgrades, software upgrades and relicensing on a more frequent basis than a business analysis would demand.
I love FOSS and I think it brings great value to computer users.
But most advocates for legislation take it one step too far by requiring use of FOSS for governmental business. In some particular isolated cases, such as voting machines, complete openness is justifiable.
So, what legislation should do is
Mandate use of complete, free, publicly-documented, open, standard interfaces.
Then, good business policy for purchasing should address thorough consideration of all options and all costs, so long as these interfaces are provided. Multiple vendors, including Microsoft, could compete on providing the highest quality, highest performance, most secure, lowest-priced, best-maintained implementation of a software product.
Mandating use of FOSS is needlessly restrictive; doing so would compound a long history of mistakes that have been made in flimsy sole-source contracts to companies [like Microsoft]. There, we've learned how a "solution" you've bought for your organization, while providing some benefits, simultaneously makes your data and your business processes hostage to someone else's demands for money.
Both purchasing models are not good business, and they can cost the taxpayers in more ways than are typically known at the time of the decision.
If we commercialize the entire infrastructure, prices will rise
Only if the commercial providers aren't competing sufficiently so that demand outstrips supply.
Commercialization works fine with multiple competitors producing higher quality services for lower prices on a standard level playing field. It's only once you lock a deal with a single vendor that prices rise and quality suffers.
The problem is how broad the tree is under.com
If the tree weren't branched so broadly at the root servers no one would care so much to obtain
the monopoly on root servers.
There ought to be bids for servicing smaller chunks of the huge namespace under the.com
Instead of all of
*.com
getting handled authoritatively by one authority, there ought to be, something like sqrt(N) intermediate servers to handle things like (I'll make up some stupid partition):
Remember, you can't advance up the management ladder to the uppermost rungs unless you have nice looking hair!
Gr[ae]y is great, but bald is bad!
On a related issue, I recall a study once that showed how elementary school teachers tended to reward "pretty" students more than "average" or "ugly" students.
If you succeed while being ugly, bald, short, and fat, you're accomplishments are the greater for having done them uphill. Of course, my minority friends will still consider you minor league.
...so it's nearly impossible to check for fraud or malfunction after an actual election.
Were you sleeping in civics class?
Everyone knows that the electoral system is solidly-based upon secure principles of operation. These principles were established long ago and are still operating in the electronic age. The specific details may vary, but the basic mechanics persists.
Positively and undeniably, election winners are exactly those people that have been best able to use money and power to fool the most people.
A pirate has come to mean something too cudly and innocuous. In fact, the loose use of the term to describe otherwise ordinary people engaging in distribution of material copyrighted by others has done much to diminish the proud tradition of "pirate".
From now on, all official RIAA pronouncements will obide by a new naming scheme. Opponents of RIAA will be referred to as "digital terrorists", "hackers", and "pedophiles", preferably in the same sentence.
...that war will be waged whereever and whenever it will be perceived to be effective.
It would be ideal if all of the resources being put into preparing for war were redirected towards solving many of the other genuine problems we have, but practically we're not there.
While the idea persists that violence against others is a viable means for achieving one's objectives, everyone must prepare for war in order to defend themselvs, even if they make every effort to wage peace.
The general's observation is correct; he's merely doing his job of looking at how war might be waged effectively.
In the overall scheme, things might be getting better if you consider together: the world's population, the magnitude of past wars, the incredible killing power available today. Despite many regional conflicts, the amount of peace is quite encouraging. But there are still mistakes being made and old wounds festering.
Given the enormous consequences of war and some of the very many bad decisions for going to war, it is incumbent upon voting citizens to become informed of facts rather than inflamed by emotion. When regional powers, civil or religious, use misinformation and play upon people's emotions, including fear and hatred, we're stepping dangerously backwards into an era where conflict was much more common than it is today.
As a recent example, in the context of American policy, the decision to pursue Al-Qaeda in Afghanistan after that government refused to deal with the specific problem was a correct one.
The case made for invading Iraq, on the other hand, was not so justified by the facts and, not surprisingly, emotion-based appeals for that action were plentiful a year ago.
And to further feed an unhealthy addiction to chipr0n, take a look at the 8-way Power5 shown at the Microprocessor Forum with 144 MB of cache - more space in cache than was on my whole disk in the early 1980's - but with a little bit more bw and less latency.
In some ways this would be getting back to Sun's roots.
Recall that prior to the late 1980's when they started to develop SPARC that they relied upon Motorola and its 68000 series processors.
UNIX companies originally thrived because the kernel (BSD, SysV) were easily available, easily portable to whatever was the fastest hardware.
Sun still has great hardware for the big SMP machines that need high data throughput, but its desktop and small server business has gotten eaten by Linux, which is ironic considering how much of UNIX progress has been due to the contributions of Sun.
But the Sun's predicament in finding a new business model is a difficult one. The forces of commoditisation from Linux/x86 are a sea change happening to them (SGI has already suffered a lot from the same forces.)
For the most common purposes that a desktop computer is used these days, a factor of 10 or 100 in CPU speed or many other performance measures doesn't matter. What I see most of the time are web browsers, email clients, word processors and presentation software running on machines that are rarely taxed by those tasks.
The standard workload of the computer needs to be expanded into an area that people find attractive and which requires the kind of special hardware performance and system integration that Sun could deliver for them to reclaim the desktop. That's harder to do now than 10 years ago, with the latest x86 chips so much closer to the best performaning chips than they were. Windows on x86 is Good Enough for most people on today's desktop.
I disagree about "Java is boring". A lot of highly useful and highly profitable lines of business are "boring". Java has already been through the fire of proving itself to be useful and not just some hyped-up vaporous bloatware. Sun should build on Java, in the embedded device market, Wi-Fi. And they should continue to champion useful standards, just like they did with NFS. Customers are likely to view Sun as a nicer player if it is a standards bearer that is generous about opening up. Then, customers will feel more secure that Sun isn't just out to wrap them up in some technology over which dictatorial control and executive fiat could wreak financial havoc. They should follow through completely with Java as a open international standard.
It can get political, emotional and religious if the discussion gets away from the facts.
Defense in depth is sound security strategy; a strategy whose soundness is manifest to people of all political persuasions.
Let Exchange do what it's good at: storing user mail messages in a database, serving IMAP clients and helping do group calendaring.
Switch out sendmail for qmail, which is more secure. Keeping a pure MTA like qmail costs very little in the way of setup and maintence and helps purify the traffic seen by your Exchange servers.
sendmail has built up at least as much of a legend for insecurity as Exchange, probably also amplified by its wide deployment.
Security in depth helps, though.
Sendmail costs nothing but a little time to install, but adds another layer to your corporate email system, one which can be used to handily filter crap that is bad for Windows systems. MyCorp has used both Exchange and sendmail for years. Performance of sendmail on piece of crap hardware is impressive, especially compared with Exchange where we need bunches of servers. To be fair, the Exchage servers are doing a lot of db management of user mailboxes that sendmail, simple MTA, does not.
Even better still, go for something like qmail or exim. Get greater security, great performance, and no mucking with sendmail.cf files.
Nothing's invulnerable, and there's still a decision with two layered MTAs as to how to layer things properly.
My own take is that the application/system/platform with the best security record and the one that is less common is the front you want to expose to to the network at large. Expose the Exchange servers more to the inside users than to to the outside world.
shouldn't fight the technology, in my opinion, but some of the uses of it.
Yes, but some of the nefarious uses for TCPM will be put into place only because the implementation of the technology is permitted. It's the camel getting its nose under the tent.
Imagine a rollout of IE 7.0 on Windows with built-in mechanisms for authentication based on your hardware. Initially, no big deal. Then, later, some of your favorite websites start returning errors because they can't "authenticate you". Most people will blindly blunder forward locking their personal information into their machine very tightly, in a way that marketing folks, online sellers of copyrighted material, and repressive governments are just drooling to get.
Once ISP's are required to proxy all port 80 transactions and won't initiate unless your machine can do the TCPA handshake, we'll be stranded. Only we'll know a lot sooner how we ended up shit creek than the average joe that uses "whatever came with My Computer".
Slashdot Mirror
You mean it's not
?Indirectly, it does.
If you look closely at the bottom of the story you'll see something like:
Bigger numbers mean more popular stories and larger ratios of the first to the second number mean more readable posts.
Slash wish list:
rights of personhood to corporations.
Then, mergers might well be consider marriage, eh? [In reference to all the civil union controversy of late.]
Then, spin-off companies would be children, subject to juvenile courts.
Then, corporation executives that cease late-stage development of a viable spin-off company might be prosecuted under abortion laws?
hydroelectric
There's the big questions about this technology.
It relies upon water under high pressure, flowing through the microchannels coated with surface charge to generate some power.
The exact same highly-pressurized water could be used to drive a turbine connected to a generator.
It will be a matter of whether the efficiency and costs of the new device can be developed to be competitive with conventional hydroelectric power.
My own favorite untapped technology for power generation is taking advantage of the mixing of fresh water and salt water that occurs where rivers flow into the ocean.
There's a tremendous source of untapped energy (look at how much power has to be put into seawater purification to get some idea.)
Perhaps this same microchannel technology could be adapted to harness energy from the mixing of salt and fresh water.
There are many issues to be worked out
Technically, the QoS (latency) over the low quality lines, seems something of an issue. I read somewhere that anything more than 0.1 sec latency makes conversation noticeably less spontaneous. Maybe it will encourage us all to listen to others without interrupting them:)
Politically, the regulated monopolies that provide local telephone service will be difficult to interface with. They'll demand money for access to the offices (some people have stories about this), but are also obligated to maintain lines and obey a slew of regulations, such as providing service to backwater areas, emergency service, tacking on a whole zoo of fees and taxes.
A Blanket policy against MS, without allowing for a competitive bidding process or even alternative analysis doesn't seem right.
Yes, you're correct. The decision should be based on cost analysis.
Unfortunately, most CIO's won't do a complete cost analysis.
There are costs associated with both Microsoft technology and costs associated with open source software that typically are ignored.
In the case FOSS, what's typically unaccounted are the training and migration costs for a workforce that is largely familiar with Microsoft's way of doing things.
In the case of Microsoft, what's typically unaccounted are the costs of lock-in from using interdependent Microsoft technology and being put on a forced upgrade path that requires hardware upgrades, software upgrades and relicensing on a more frequent basis than a business analysis would demand.
I love FOSS and I think it brings great value to computer users.
But most advocates for legislation take it one step too far by requiring use of FOSS for governmental business. In some particular isolated cases, such as voting machines, complete openness is justifiable.
So, what legislation should do is
- Mandate use of complete, free, publicly-documented, open, standard interfaces.
Then, good business policy for purchasing should address thorough consideration of all options and all costs, so long as these interfaces are provided. Multiple vendors, including Microsoft, could compete on providing the highest quality, highest performance, most secure, lowest-priced, best-maintained implementation of a software product.Mandating use of FOSS is needlessly restrictive; doing so would compound a long history of mistakes that have been made in flimsy sole-source contracts to companies [like Microsoft]. There, we've learned how a "solution" you've bought for your organization, while providing some benefits, simultaneously makes your data and your business processes hostage to someone else's demands for money.
Both purchasing models are not good business, and they can cost the taxpayers in more ways than are typically known at the time of the decision.
Blaster, Welchia, whatever cause my computer to crash, yawn.
Software in nuclear power plants shutsdown, OK.
Software in electric grid gets flaky - I'm cool - eat the ice cream by candlelight.
Software in grandpa's life support system has to reboot - well, them's the breaks.
Commercial airliner avionice system goes down - so sorry.
But...
Software messin' with my beer? SHIT, WE GOT A CRISIS HERE!
If we commercialize the entire infrastructure, prices will rise
Only if the commercial providers aren't competing sufficiently so that demand outstrips supply.
Commercialization works fine with multiple competitors producing higher quality services for lower prices on a standard level playing field. It's only once you lock a deal with a single vendor that prices rise and quality suffers.
The problem is how broad the tree is under .com
If the tree weren't branched so broadly at the root servers no one would care so much to obtain the monopoly on root servers.
There ought to be bids for servicing smaller chunks of the huge namespace under the .com
Instead of all of
getting handled authoritatively by one authority, there ought to be, something like sqrt(N) intermediate servers to handle things like (I'll make up some stupid partition):etc.
Just a suggestions for breaking the logjam.
Remember, you can't advance up the management ladder to the uppermost rungs unless you have nice looking hair!
Gr[ae]y is great, but bald is bad!
On a related issue, I recall a study once that showed how elementary school teachers tended to reward "pretty" students more than "average" or "ugly" students.
If you succeed while being ugly, bald, short, and fat, you're accomplishments are the greater for having done them uphill. Of course, my minority friends will still consider you minor league.
Does Cygwin run on WinCE?
Were you sleeping in civics class?
Everyone knows that the electoral system is solidly-based upon secure principles of operation. These principles were established long ago and are still operating in the electronic age. The specific details may vary, but the basic mechanics persists.
Positively and undeniably, election winners are exactly those people that have been best able to use money and power to fool the most people.
A pirate has come to mean something too cudly and innocuous. In fact, the loose use of the term to describe otherwise ordinary people engaging in distribution of material copyrighted by others has done much to diminish the proud tradition of "pirate".
From now on, all official RIAA pronouncements will obide by a new naming scheme. Opponents of RIAA will be referred to as "digital terrorists", "hackers", and "pedophiles", preferably in the same sentence.
...that war will be waged whereever and whenever it will be perceived to be effective.
It would be ideal if all of the resources being put into preparing for war were redirected towards solving many of the other genuine problems we have, but practically we're not there.
While the idea persists that violence against others is a viable means for achieving one's objectives, everyone must prepare for war in order to defend themselvs, even if they make every effort to wage peace.
The general's observation is correct; he's merely doing his job of looking at how war might be waged effectively.
In the overall scheme, things might be getting better if you consider together: the world's population, the magnitude of past wars, the incredible killing power available today. Despite many regional conflicts, the amount of peace is quite encouraging. But there are still mistakes being made and old wounds festering.
Given the enormous consequences of war and some of the very many bad decisions for going to war, it is incumbent upon voting citizens to become informed of facts rather than inflamed by emotion. When regional powers, civil or religious, use misinformation and play upon people's emotions, including fear and hatred, we're stepping dangerously backwards into an era where conflict was much more common than it is today.
As a recent example, in the context of American policy, the decision to pursue Al-Qaeda in Afghanistan after that government refused to deal with the specific problem was a correct one.
The case made for invading Iraq, on the other hand, was not so justified by the facts and, not surprisingly, emotion-based appeals for that action were plentiful a year ago.
With a limited budget
be inclined to buy the software
If it works for drugs I don't see any reason why it can't work for software!
In some cases there are 12 step program that can get you the help you need.
Like say, the Alpha?
Yes, like the Alpha, RIP.
And to further feed an unhealthy addiction to chipr0n, take a look at the 8-way Power5 shown at the Microprocessor Forum with 144 MB of cache - more space in cache than was on my whole disk in the early 1980's - but with a little bit more bw and less latency.
A super cool, fast and cheap workstation.
In some ways this would be getting back to Sun's roots.
Recall that prior to the late 1980's when they started to develop SPARC that they relied upon Motorola and its 68000 series processors.
UNIX companies originally thrived because the kernel (BSD, SysV) were easily available, easily portable to whatever was the fastest hardware.
Sun still has great hardware for the big SMP machines that need high data throughput, but its desktop and small server business has gotten eaten by Linux, which is ironic considering how much of UNIX progress has been due to the contributions of Sun.
But the Sun's predicament in finding a new business model is a difficult one. The forces of commoditisation from Linux/x86 are a sea change happening to them (SGI has already suffered a lot from the same forces.)
For the most common purposes that a desktop computer is used these days, a factor of 10 or 100 in CPU speed or many other performance measures doesn't matter. What I see most of the time are web browsers, email clients, word processors and presentation software running on machines that are rarely taxed by those tasks.
The standard workload of the computer needs to be expanded into an area that people find attractive and which requires the kind of special hardware performance and system integration that Sun could deliver for them to reclaim the desktop. That's harder to do now than 10 years ago, with the latest x86 chips so much closer to the best performaning chips than they were. Windows on x86 is Good Enough for most people on today's desktop.
I disagree about "Java is boring". A lot of highly useful and highly profitable lines of business are "boring". Java has already been through the fire of proving itself to be useful and not just some hyped-up vaporous bloatware. Sun should build on Java, in the embedded device market, Wi-Fi. And they should continue to champion useful standards, just like they did with NFS. Customers are likely to view Sun as a nicer player if it is a standards bearer that is generous about opening up. Then, customers will feel more secure that Sun isn't just out to wrap them up in some technology over which dictatorial control and executive fiat could wreak financial havoc. They should follow through completely with Java as a open international standard.
Windows metrically equivalent fonts
Cigar time. That is the one key most signifant practical barrier to widespread adoption of Open Office.
If metric equivalent fonts, especially non Latin fonts, were widely available, there would be a tidal wave of OOo adoption.
since data could be bussed to the CPU fast enough
Got that right.
The whole design of systems is going in the direction where main memory will be considered as slow as disks once were.
It will be considered as much a sin to miss the L2 cache as it is to swap.
The chips that will be speed kings will be the ones that can afford huge fast caches.
security debate (which can get political).
It can get political, emotional and religious if the discussion gets away from the facts.
Defense in depth is sound security strategy; a strategy whose soundness is manifest to people of all political persuasions.
Let Exchange do what it's good at: storing user mail messages in a database, serving IMAP clients and helping do group calendaring.
Switch out sendmail for qmail, which is more secure. Keeping a pure MTA like qmail costs very little in the way of setup and maintence and helps purify the traffic seen by your Exchange servers.
when I started to read the story I couldn't help but think of writing style = "Jon Katz".
not using sendmail
sendmail has built up at least as much of a legend for insecurity as Exchange, probably also amplified by its wide deployment.
Security in depth helps, though.
Sendmail costs nothing but a little time to install, but adds another layer to your corporate email system, one which can be used to handily filter crap that is bad for Windows systems. MyCorp has used both Exchange and sendmail for years. Performance of sendmail on piece of crap hardware is impressive, especially compared with Exchange where we need bunches of servers. To be fair, the Exchage servers are doing a lot of db management of user mailboxes that sendmail, simple MTA, does not.
Even better still, go for something like qmail or exim. Get greater security, great performance, and no mucking with sendmail.cf files.
Nothing's invulnerable, and there's still a decision with two layered MTAs as to how to layer things properly.
My own take is that the application/system/platform with the best security record and the one that is less common is the front you want to expose to to the network at large. Expose the Exchange servers more to the inside users than to to the outside world.
I'm still waiting for my dream to come true - where networked vehicle control eliminates the need for traffic signals and stop signs.
Imagine sailing down city streets at freeway speeds, with perpedicular streams of traffic flowing through another through the magic of precise timing.
Of course a firewall will offer some protection but shouldn't be relied on.
Check.
Unfinished poetry composition from RPC...
Ah! I get it now!
You said Libraries of Congress.
Initially I was confused, thinking it was just "Congress".
As you know, political bodies emit information in a continuous unending unquantifiable stream, as in
shouldn't fight the technology, in my opinion, but some of the uses of it.
Yes, but some of the nefarious uses for TCPM will be put into place only because the implementation of the technology is permitted. It's the camel getting its nose under the tent.
Imagine a rollout of IE 7.0 on Windows with built-in mechanisms for authentication based on your hardware. Initially, no big deal. Then, later, some of your favorite websites start returning errors because they can't "authenticate you". Most people will blindly blunder forward locking their personal information into their machine very tightly, in a way that marketing folks, online sellers of copyrighted material, and repressive governments are just drooling to get.
Once ISP's are required to proxy all port 80 transactions and won't initiate unless your machine can do the TCPA handshake, we'll be stranded. Only we'll know a lot sooner how we ended up shit creek than the average joe that uses "whatever came with My Computer".