Space travel with chemical fuels isn't feasible. You just can't pack enough energy per unit mass into the fuel. This is a fundamental limitation of chemistry. Liquid oxygen/liquid hydrogen is as good as it can possibly get, and that's been in use for decades.
Only by desperate weight reduction measures, resulting in incredibly fragile vehicles, is anything made to fly into space at all. The vehicles are almost all fuel. Pieces have to be thrown away after launch. Payloads are dinky for the size of the vehicle. Costs are insanely high.
It's been that way for over forty years. It's not getting any better. No combination of parts will fix this fundamentally broken technology.
Space travel won't work until we get a better energy source.
FANG, from 1972, is probably one of the oldest applications you can still download and run. It's a copying utility for UNIVAC mainframes. UNIVAC Exec 8 was way ahead of its time, with full support for threads, multiprocessors, and concurrent I/O from the late 1960s. FANG was one of the first applications to use that concurrency effectively. You could put in a series of commands to operate on multiple files, and it would do them as concurrently as possible, keeping track of any dependencies in the file copies.
Back when the Shuttle was called the "National Space Transportation System" and NASA was claiming that launch costs would come down, NASA used to talk about materials processing in space. That was a long time ago.
The trouble with materials processing in space is that for small things, gravity is dominated by surface tension and other forces like Brownian motion. So biological processing in space never amounted to much. Some early Shuttle flights carried an electrophoresis apparatus designed for zero-G operation to make some kind of diabetes drug. But bioengineering went beyond that approach; today it's easier to engineer some bacterium to crank out whatever you need.
For big objects, there would be some advantages (and many disadvantages) to working in zero G. Handling molten metal in zero G safely would be tough. One molten droplet could puncture anything we currently send into space. With gravity and in air, molten droplets
don't travel very far and cool. In space, they can go a long way.
Steel mills use floors of dirt or refractory brick in molten metal areas; concrete will blow up when its water content boils.
Welding in space has been tried, but on a very small scale, and very nervously.
Lift to orbit is far too expensive to justify flying heavy metal up there for casting and welding.
This is one of those ideas that won't be feasible unless and until lift to orbit costs about what long distance air travel costs now.
Bad publicity made Google fix their open redirector for URLs. Bad publicity will make them fix this.
GMail ought to go back to cell phone authentication for new accounts. Since their capcha was broken, they've become a favorite of spammers.
Blogspot is also a spam haven. Most blogspot blogs are spam, and they can be used as a form of open redirector. Look for spams like: "An IWC watch is a uniquely handcrafted time piece... http://rexefute51720.blogspot.com/"
Complain loudly, publicly, and often. Google needs to take stronger steps to avoid being a spam conduit.
The PC industry is terrified of low-cost laptops. They see $199 laptops in bubble packs at every WalMart, with a profit of about $1 per unit. Dell is in trouble; their custom-build business model is dying. So Microsoft's approach to driving up prices looks attractive.
It won't last, but it might be good for a few years.
California has a similar law, Civil Code section 3344. This covers "publicity rights". Each person's "publicity right" in recognizable images of themself is by law worth at least $750, if used in any manner related to advertising or selling. If you're famous, the price goes up, to cover "actual damages".
So if you're in California and recognizable in Google StreetView, you could put in a claim. It's not worth it unless you're a major celebrity.
Some open plan offices have sound dampening systems; loudspeakers that play white noise at a low level. You couldn't hear them, but you couldn't hear the person three desks away either.
I once stayed in a Howard Johnson's Motor Lodge in Pittsburg that had piped-in noise. At each corridor intersection or bend, there was a speaker. But it wasn't white noise. It was machinery noise - a faint background of whirr, chunka chunka, hiss, whirr, clank. At first I thought someone had just left a microphone open somewhere, but after three days, I realized it had to be intentional. Maybe heavy industry people find it peaceful.
That's very Pittsburgh. I was visiting some robotics people at CMU, and they had desks in a room with a big air compressor. Every ten minutes or so, the air compressor would start, run for about thirty seconds, and shut down with a big hiss. It was too noisy to talk over, so everyone just waited until it stopped. No effort had been made to muffle the thing. This was accepted as normal.
Intel is famous for their workspaces. They pioneered cubicles in the early 1970s.
They have some of the world's biggest single-room cube farms. They actually built
new buildings, from the ground up, with 1-acre rooms of tiny grey cubicles. Vast
amounts of money were spent to create this Dilbertland. The
cubicles are so tiny that two people cannot physically sit in one and talk;
one has to sit out on the aisle and block traffic. They look like library
carrels. This isn't a call center; it's where their engineers work.
Klutz Press has a "fun workspace" - the partitions are made out of corrugated sheet metal.
The building (a warehouse) is made of corrugated sheet metal. Lots of toys in the reception area.
Softimage LA went through a period where everything, including partitions, was curved and on wheels. You could fold up the cubicle of someone who was out and push it to the side.
Sony Pictures Imageworks, an animation shop, is a typical cube farm surrounded by offices.
Except for the art department, which has a big open space with drawing boards.
Silicon Valley law firms tend to have rocks. Big rocks. Polished stone surfaces.
Rock gardens. And, for some reason, glass-enclosed conference rooms. Traditional law firms used to go in heavily for wood paneling, but the "high tech" law firms wanted a more modern look. The overall effect is upscale mall, but whatever.
The only thing that makes this hard is a metric of what "fully loaded" means for a server. With generators and boilers, you have a single number which represents output, and you know what the capacity of each unit is, so you know when to start up the next unit. Computer servers are more difficult to characterize.
So you have to measure some values of server load, convert that to a single number, and use it for load measurement purposes. Then it all works just like boiler scheduling.
You don't even need to do much advance planning, as you have to do with boilers and generators, since you can usually start up another server in a minute or so. It takes hours to fire up a big boiler, so you need serious prediction capability.
(The classic power company approach was a chart recorder recording system load. Every day, somebody took the day's load graph and cut out a piece of cardboard to match. The cardboard pieces were accumulated in a rack, and the result was a 3D load graph for the year. It looks like a mountain range. There's an Internet Archive film showing this. That's a worthwhile exercise for your
server farm, and you can probably do it without glue and scissors today. I've seen some of Amazon's server load graphs, which have a huge peak entering the Xmas buying season. In fact, the real reason Amazon is selling "cloud computing" is that their plant is sized for the holiday season and 80% idle the rest of the year.)
Operators of multiple steam boilers have been dealing with this problem for a century.
The number of boilers fired up is adjusted with demand, with the need for some demand prediction because it takes time to get steam up. This was done manually for decades; now it's often automated.
The same thing applies to multiple HVAC compressors. Usually there's a long-term round-robin
switch so that the order of compressor start is rotated on a daily or weekly basis to equalize wear.
More and more, IT is becoming like stationary engineering.
But you still can't download fonts in the browser as part of an HTML document.
That used to work, back in the early days of Mozilla. Microsoft refused to put it in IE, and came up with their own, incompatible system. Mozilla then took theirs out.
It looked like an ADM 3A
on
iMac Turns 10
·
· Score: 1, Interesting
Certainly it didn't look like any other other computer.
There was some guy who had a bicycle with all sorts of electronics on it about twenty years ago. He had an Internet connection using VHF amateur radio. He kept adding on more stuff, and ended up with a trailer on the bike and pedaling in low gear all the time.
Today, it's kind of silly. Devices in the iPhone size class can do almost everything, so building all this gear into a vehicle is kind of pointless.
Way, way overpriced. Four 1.25MW turbines for $90 million, or $18/watt? That's far too high. Compare the Cedar Ridge project, with 41 turbines of 1.65MW capacity each for $180 million, or $2.6/watt. That's a real not-to-exceed number. The American Wind Energy Association likes to talk about $1/watt, but that's seldom achieved.
The easiest way is to authenticate by cell phone number. When you register for a site, your password is sent to your cell phone as an SMS message. One registration per cell phone number.
Yes, it's possible to buy multiple SIM cards to get more phone numbers, but they're not free.
This costs the site about $0.05 for each message sent. For sites that derive some value from having members, it's worth it.
Slashdot would have paid about $50,000 or so in SMS fees by now.
What we're seeing, of course, is the migration of GUIs to browsers. The browser interface isn't very powerful, but it works well enough for most enterprise apps. After all, in most enterprise applications, the real work is being done in the back end, near the database.
The last time I wrote a real GUI application, it was for QNX, where we needed a control panel for a real time system. Anything that looks like a business app I do with a browser in the front. It just isn't worth the effort to deal with the Windows API any more unless you're building
something really elaborate.
In the game community, the trend is towards using Flash for the 2D portions of the GUI. This separates the graphical design from the implementation, and there are good authoring tools for Flash. If you need a "pretty" GUI, and have the artistic talent on tap, that's the way to go.
That's been the trouble with these "peer to peer" protocols. The routing algorithms have been horribly inefficient. It's quite possible to have the same data flowing in both directions on the same pipe. Multiple copies, even.
It might be cheaper for the telecom industry (which is big) to buy out the music industry (which is tiny) and just cache the RIAA's entire output on local servers. Just cacheing the top 100 releases or so might cut traffic in half.
(This won't scale to movies, though. Movies are bigger and more expensive to make.)
It's time for one of the major desktop manufacturers to cut a deal with Apple to make Mac desktop machines. It's time for Apple to exit desktops anyway; laptops are taking over in the personal market. But in business, where there are desks, desktops will be around for years to come. Since they're just x86 machines, there's no technical obstacle.
The problem is that the Slashdot article has links to places that get Roland the Plogger ad revenue, but doesn't have a link to the original paper. This is typical Roland the Plogger behavior.
As is typical of a Roland the Plogger article, there's no link to the original article, but there's a link to his ad-laden blog.
Here's
the abstract:
Hierarchical structure and the prediction of missing links in networks
Nature 453, 98 (2008). doi:10.1038/nature06830
Authors: Aaron Clauset, Cristopher Moore
& M. E. J. Newman
Networks have in recent years emerged as an invaluable tool for describing and quantifying complex systems in many branches of science. Recent studies suggest that networks often exhibit hierarchical organization, in which vertices divide into groups that further subdivide into groups of groups, and so forth over multiple scales. In many cases the groups are found to correspond to known functional units, such as ecological niches in food webs, modules in biochemical networks (protein interaction networks, metabolic networks or genetic regulatory networks) or communities in social networks. Here we present a general technique for inferring hierarchical structure from network data and show that the existence of hierarchy can simultaneously explain and quantitatively reproduce many commonly observed topological properties of networks, such as right-skewed degree distributions, high clustering coefficients and short path lengths. We further show that knowledge of hierarchical structure can be used to predict missing connections in partly known networks with high accuracy, and for more general network structures than competing techniques. Taken together, our results suggest that hierarchy is a central organizing principle of complex networks, capable of offering insight into many network phenomena.
So now, unlike Roland, we now have a clue what's being talked about. It's a scheme for finding some structure in networks and inferring what links might be missing.
It's a serviceable, but mediocre superhero movie. Worse than any of the Batman or Spiderman franchises. Worse than the second Fantastic 4 movie. On a par, maybe, with "Supergirl".
It's better than the other comic book derivative with Howard Hughes, "The Rocketeer", but that's not saying much.
Marvel Studios isn't the next Pixar. More like the next DisneyToons, the "Crap Sequel Division" of Disney (Cinderella III, Bambi II, Brother Bear 2, The Lion King 1½, The Lion King 2, Lady and the Tramp 2) working to milk the last possible dollar out of each franchise.
Stay home and play through GTA IV. Wait for "Speed Racer" next week.
There are lots of little IPC packages. Too many, actually. OpenRPC is somewhat dated, but OK. The last time I had to do this, I used Python, CPickle and pipes. But that wasn't a IPC-intensive application. When we did a robot vehicle for the DARPA Grand Challenge, there were about 20 processes communicating via QNX MsgSend/MsgReceive, and that worked out well, including dealing with hard real-time constraints, heavy computation on the same CPU as low-level hard real time control loops, and video over IPC. That's why I'm impressed with QNX RPC.
A major problem in the Unix/Linux world is that there's no decent IPC system that's always there, always on, and used for almost everything. At least Windows has OLE. Gnome and OpenOffice both use CORBA, but have each has its own, incompatible, CORBA ORB. That kind of thing is part of the problem.
Slashdot Mirror
Space travel with chemical fuels isn't feasible. You just can't pack enough energy per unit mass into the fuel. This is a fundamental limitation of chemistry. Liquid oxygen/liquid hydrogen is as good as it can possibly get, and that's been in use for decades.
Only by desperate weight reduction measures, resulting in incredibly fragile vehicles, is anything made to fly into space at all. The vehicles are almost all fuel. Pieces have to be thrown away after launch. Payloads are dinky for the size of the vehicle. Costs are insanely high.
It's been that way for over forty years. It's not getting any better. No combination of parts will fix this fundamentally broken technology.
Space travel won't work until we get a better energy source.
FANG, from 1972, is probably one of the oldest applications you can still download and run. It's a copying utility for UNIVAC mainframes. UNIVAC Exec 8 was way ahead of its time, with full support for threads, multiprocessors, and concurrent I/O from the late 1960s. FANG was one of the first applications to use that concurrency effectively. You could put in a series of commands to operate on multiple files, and it would do them as concurrently as possible, keeping track of any dependencies in the file copies.
Back when the Shuttle was called the "National Space Transportation System" and NASA was claiming that launch costs would come down, NASA used to talk about materials processing in space. That was a long time ago.
The trouble with materials processing in space is that for small things, gravity is dominated by surface tension and other forces like Brownian motion. So biological processing in space never amounted to much. Some early Shuttle flights carried an electrophoresis apparatus designed for zero-G operation to make some kind of diabetes drug. But bioengineering went beyond that approach; today it's easier to engineer some bacterium to crank out whatever you need.
For big objects, there would be some advantages (and many disadvantages) to working in zero G. Handling molten metal in zero G safely would be tough. One molten droplet could puncture anything we currently send into space. With gravity and in air, molten droplets don't travel very far and cool. In space, they can go a long way. Steel mills use floors of dirt or refractory brick in molten metal areas; concrete will blow up when its water content boils. Welding in space has been tried, but on a very small scale, and very nervously.
Lift to orbit is far too expensive to justify flying heavy metal up there for casting and welding. This is one of those ideas that won't be feasible unless and until lift to orbit costs about what long distance air travel costs now.
Bad publicity made Google fix their open redirector for URLs. Bad publicity will make them fix this.
GMail ought to go back to cell phone authentication for new accounts. Since their capcha was broken, they've become a favorite of spammers.
Blogspot is also a spam haven. Most blogspot blogs are spam, and they can be used as a form of open redirector. Look for spams like: "An IWC watch is a uniquely handcrafted time piece ... http://rexefute51720.blogspot.com/"
Complain loudly, publicly, and often. Google needs to take stronger steps to avoid being a spam conduit.
The PC industry is terrified of low-cost laptops. They see $199 laptops in bubble packs at every WalMart, with a profit of about $1 per unit. Dell is in trouble; their custom-build business model is dying. So Microsoft's approach to driving up prices looks attractive.
It won't last, but it might be good for a few years.
That's really what Intel looks like. And those are the better cubicles. Some sections have smaller ones.
California has a similar law, Civil Code section 3344. This covers "publicity rights". Each person's "publicity right" in recognizable images of themself is by law worth at least $750, if used in any manner related to advertising or selling. If you're famous, the price goes up, to cover "actual damages".
So if you're in California and recognizable in Google StreetView, you could put in a claim. It's not worth it unless you're a major celebrity.
Aargh. I spent a half hour listening to this drivel. Next time, post a transcript.
Most of what he says is either trivial or a weird attempt to reconcile patent and copyright law with the writings of Ayn Rand.
Some open plan offices have sound dampening systems; loudspeakers that play white noise at a low level. You couldn't hear them, but you couldn't hear the person three desks away either.
I once stayed in a Howard Johnson's Motor Lodge in Pittsburg that had piped-in noise. At each corridor intersection or bend, there was a speaker. But it wasn't white noise. It was machinery noise - a faint background of whirr, chunka chunka, hiss, whirr, clank. At first I thought someone had just left a microphone open somewhere, but after three days, I realized it had to be intentional. Maybe heavy industry people find it peaceful.
That's very Pittsburgh. I was visiting some robotics people at CMU, and they had desks in a room with a big air compressor. Every ten minutes or so, the air compressor would start, run for about thirty seconds, and shut down with a big hiss. It was too noisy to talk over, so everyone just waited until it stopped. No effort had been made to muffle the thing. This was accepted as normal.
Intel is famous for their workspaces. They pioneered cubicles in the early 1970s. They have some of the world's biggest single-room cube farms. They actually built new buildings, from the ground up, with 1-acre rooms of tiny grey cubicles. Vast amounts of money were spent to create this Dilbertland. The cubicles are so tiny that two people cannot physically sit in one and talk; one has to sit out on the aisle and block traffic. They look like library carrels. This isn't a call center; it's where their engineers work.
Klutz Press has a "fun workspace" - the partitions are made out of corrugated sheet metal. The building (a warehouse) is made of corrugated sheet metal. Lots of toys in the reception area.
Softimage LA went through a period where everything, including partitions, was curved and on wheels. You could fold up the cubicle of someone who was out and push it to the side.
Sony Pictures Imageworks, an animation shop, is a typical cube farm surrounded by offices. Except for the art department, which has a big open space with drawing boards.
Silicon Valley law firms tend to have rocks. Big rocks. Polished stone surfaces. Rock gardens. And, for some reason, glass-enclosed conference rooms. Traditional law firms used to go in heavily for wood paneling, but the "high tech" law firms wanted a more modern look. The overall effect is upscale mall, but whatever.
The only thing that makes this hard is a metric of what "fully loaded" means for a server. With generators and boilers, you have a single number which represents output, and you know what the capacity of each unit is, so you know when to start up the next unit. Computer servers are more difficult to characterize.
So you have to measure some values of server load, convert that to a single number, and use it for load measurement purposes. Then it all works just like boiler scheduling.
You don't even need to do much advance planning, as you have to do with boilers and generators, since you can usually start up another server in a minute or so. It takes hours to fire up a big boiler, so you need serious prediction capability.
(The classic power company approach was a chart recorder recording system load. Every day, somebody took the day's load graph and cut out a piece of cardboard to match. The cardboard pieces were accumulated in a rack, and the result was a 3D load graph for the year. It looks like a mountain range. There's an Internet Archive film showing this. That's a worthwhile exercise for your server farm, and you can probably do it without glue and scissors today. I've seen some of Amazon's server load graphs, which have a huge peak entering the Xmas buying season. In fact, the real reason Amazon is selling "cloud computing" is that their plant is sized for the holiday season and 80% idle the rest of the year.)
Operators of multiple steam boilers have been dealing with this problem for a century. The number of boilers fired up is adjusted with demand, with the need for some demand prediction because it takes time to get steam up. This was done manually for decades; now it's often automated.
The same thing applies to multiple HVAC compressors. Usually there's a long-term round-robin switch so that the order of compressor start is rotated on a daily or weekly basis to equalize wear.
More and more, IT is becoming like stationary engineering.
But you still can't download fonts in the browser as part of an HTML document.
That used to work, back in the early days of Mozilla. Microsoft refused to put it in IE, and came up with their own, incompatible system. Mozilla then took theirs out.
Certainly it didn't look like any other other computer.
Yes, it did:
There was some guy who had a bicycle with all sorts of electronics on it about twenty years ago. He had an Internet connection using VHF amateur radio. He kept adding on more stuff, and ended up with a trailer on the bike and pedaling in low gear all the time.
Today, it's kind of silly. Devices in the iPhone size class can do almost everything, so building all this gear into a vehicle is kind of pointless.
The Art of Electronics was great, and went through two editions, but the last was in 1989. Nothing quite as good has been published since.
Way, way overpriced. Four 1.25MW turbines for $90 million, or $18/watt? That's far too high. Compare the Cedar Ridge project, with 41 turbines of 1.65MW capacity each for $180 million, or $2.6/watt. That's a real not-to-exceed number. The American Wind Energy Association likes to talk about $1/watt, but that's seldom achieved.
$18/watt is either wrong or a rip-off.
The easiest way is to authenticate by cell phone number. When you register for a site, your password is sent to your cell phone as an SMS message. One registration per cell phone number. Yes, it's possible to buy multiple SIM cards to get more phone numbers, but they're not free.
This costs the site about $0.05 for each message sent. For sites that derive some value from having members, it's worth it.
Slashdot would have paid about $50,000 or so in SMS fees by now.
What we're seeing, of course, is the migration of GUIs to browsers. The browser interface isn't very powerful, but it works well enough for most enterprise apps. After all, in most enterprise applications, the real work is being done in the back end, near the database.
The last time I wrote a real GUI application, it was for QNX, where we needed a control panel for a real time system. Anything that looks like a business app I do with a browser in the front. It just isn't worth the effort to deal with the Windows API any more unless you're building something really elaborate.
In the game community, the trend is towards using Flash for the 2D portions of the GUI. This separates the graphical design from the implementation, and there are good authoring tools for Flash. If you need a "pretty" GUI, and have the artistic talent on tap, that's the way to go.
That's been the trouble with these "peer to peer" protocols. The routing algorithms have been horribly inefficient. It's quite possible to have the same data flowing in both directions on the same pipe. Multiple copies, even.
It might be cheaper for the telecom industry (which is big) to buy out the music industry (which is tiny) and just cache the RIAA's entire output on local servers. Just cacheing the top 100 releases or so might cut traffic in half.
(This won't scale to movies, though. Movies are bigger and more expensive to make.)
It's time for one of the major desktop manufacturers to cut a deal with Apple to make Mac desktop machines. It's time for Apple to exit desktops anyway; laptops are taking over in the personal market. But in business, where there are desks, desktops will be around for years to come. Since they're just x86 machines, there's no technical obstacle.
Psystar may be on to something.
The problem is that the Slashdot article has links to places that get Roland the Plogger ad revenue, but doesn't have a link to the original paper. This is typical Roland the Plogger behavior.
As is typical of a Roland the Plogger article, there's no link to the original article, but there's a link to his ad-laden blog. Here's the abstract:
Hierarchical structure and the prediction of missing links in networks
Nature 453, 98 (2008). doi:10.1038/nature06830
Authors: Aaron Clauset, Cristopher Moore & M. E. J. Newman
Networks have in recent years emerged as an invaluable tool for describing and quantifying complex systems in many branches of science. Recent studies suggest that networks often exhibit hierarchical organization, in which vertices divide into groups that further subdivide into groups of groups, and so forth over multiple scales. In many cases the groups are found to correspond to known functional units, such as ecological niches in food webs, modules in biochemical networks (protein interaction networks, metabolic networks or genetic regulatory networks) or communities in social networks. Here we present a general technique for inferring hierarchical structure from network data and show that the existence of hierarchy can simultaneously explain and quantitatively reproduce many commonly observed topological properties of networks, such as right-skewed degree distributions, high clustering coefficients and short path lengths. We further show that knowledge of hierarchical structure can be used to predict missing connections in partly known networks with high accuracy, and for more general network structures than competing techniques. Taken together, our results suggest that hierarchy is a central organizing principle of complex networks, capable of offering insight into many network phenomena.
So now, unlike Roland, we now have a clue what's being talked about. It's a scheme for finding some structure in networks and inferring what links might be missing.
It's a serviceable, but mediocre superhero movie. Worse than any of the Batman or Spiderman franchises. Worse than the second Fantastic 4 movie. On a par, maybe, with "Supergirl". It's better than the other comic book derivative with Howard Hughes, "The Rocketeer", but that's not saying much.
Marvel Studios isn't the next Pixar. More like the next DisneyToons, the "Crap Sequel Division" of Disney (Cinderella III, Bambi II, Brother Bear 2, The Lion King 1½, The Lion King 2, Lady and the Tramp 2) working to milk the last possible dollar out of each franchise.
Stay home and play through GTA IV. Wait for "Speed Racer" next week.
you might like to check out ICE
There are lots of little IPC packages. Too many, actually. OpenRPC is somewhat dated, but OK. The last time I had to do this, I used Python, CPickle and pipes. But that wasn't a IPC-intensive application. When we did a robot vehicle for the DARPA Grand Challenge, there were about 20 processes communicating via QNX MsgSend/MsgReceive, and that worked out well, including dealing with hard real-time constraints, heavy computation on the same CPU as low-level hard real time control loops, and video over IPC. That's why I'm impressed with QNX RPC.
A major problem in the Unix/Linux world is that there's no decent IPC system that's always there, always on, and used for almost everything. At least Windows has OLE. Gnome and OpenOffice both use CORBA, but have each has its own, incompatible, CORBA ORB. That kind of thing is part of the problem.