Here is a book that shows the field of market prediction mathematics will not directly make a person money by playing the market.
But on the other hand, it is an area of mathematical study that involves the most exciting developments in applied mathematics in the last 50 years, namely chaos and fractal theory.
"The (Mis)Behavior of Markets. A Fractal View of Risk, Ruin, and Reward" by Benoit Mandelbrot and Richard L. Hudson, Basic Books, New York, c. 2004.
This is a really good question and it points to the need for more utilities to check and fix bit level errors in files and memory images.
I wish there was a set of Linux utilities for generating and using error correction checksums.
The usual checksum tool simply reports if a file has an error.
What I would like to see is a checksum tool capable of fixing a multi-byte, single byte or single bit error.
These are checksums using Hammung codes I think?
I strongly suspect that huge hunks of don't care data like JPG photos and DVI movies develop bit level errors during storage, handling, copying, and editing.
A utility like gpg will spot errors. And diff spots character level errors. But they are both real clumsy at finding and precisely repairing individual bits. With diff you wind up unsure if your reference or original file changed.
This Linux system I am writing on had a bad RAN memory bit and the random number generator loaded over that location.
The random number generator didn't work but Linux mostly ran OK. I thought it was a software problem for several weeks. Finally after replacing the random number code and even checking bug reports, I finally looked at the RAM memory. Bing!
Check out the "avoid patents" development strategy and tactics that has been developed and tested by technical writer and engineer Don Lancaster.
Don wrote "CMOS Cookbook" around 1973 and he is still active selling some specialized products, books, EBay stuff and more.
His strategy is: avoid patents, build a kit or add on first, stay below the radar of the big players, treat the project as part of a process rather than a one time bonanza.
I recommend you get your school library to carry The World of Mathematics edited by Newman, it was first published in the '50's.
I discovered this 4 volume anthology at my local Los Angeles Public library branch in the 70's and I wound up hunting for years for a used copy of the anthology. In the mid-80's it was reprinted in paperback.
It is a 4 volume anthology containing selected essays and articles about every important field of mathematics.
Really enjoyable selections in this book include a terrific introduction to double entry accounting, life insurance, the Seven Bridges of Konigsberg, early papers on Turing machines, completeness theorems, codes and codebreaking and information theory.
Two more inexpensive and interesting books are:
Great Ideas in Operational Research, published by Dover. and Formal Knot Theory by Kauffman.
The painful part about having Microsoft computers in a special education classroom is the entire Windows interface is wrong for special ed kids.
On a Linux box, a web browser could switch on an age and ability appropriate interface model.
I work in Special Education classrooms with all kinds of disabled kids. The PC interface is awful.
The Linux desktops just shuffle some choices, but Linux could be radically better.
By interface, I mean the mouse, the left button, the right button, the roller, the keystrokes, the cursor, the active area on the screen. The interface is too chopped up, too many choices, and too easy to disrupt. The interface requires too many ideas linked to isolated fine motor actions. ---- The most successful interface for special ed kids is the teletubbies web games on pbs.org (where the youth can play peek-a-boo using a pushbutton wired to the left mouse button). The other success is the Stellaluna kids book on CD. It can be used where the adult holds the mouse and the youth pushes a button wired to the left mouse. It also models the success of musically modulated speech to enhance communication with kids who parse speech slowly. ------- But for these applications, the adult interface surrounding the application is still the problem. ------------ What I would like to see is research based interface protocols for specific levels of receptive language (like 1 year old), visual processing and motor ability (like can use hand but can't isolate finger).
The interface protocol would be like a style sheet, it masks and configures the user's mouse, display, speakers, and keyboard.
This would be great as a web app. You could search the web for apps that can be presented to your student with ability appropriate interface.
Applications for special ed kids are very few in number, the ones I know are 1990 vintage Windows/MacIntosh programs implemented with a slide show toolkit.
An example is "Teen Tunes". It is still a "CD must be in the drive" program with a '90's copyright date. It sells for over $100. Like other special ed programs, this one is a frozen experiment in interface simplification. The aide still has to hold the mouse and center the cursor while the student pushes a button.
Here are a few safety points that I leaned from using a generator during a hurricane and ice storm in Maryland. I did an emergency family power solution using a 3500 watt Craftsman generator.
First, buy and test your generator before you need it. The generator I got had a missing needle valve in the carburetor. Lucky I realized it was gasoline and not rain before I pulled the starter rope.
Second, the generator is a fire hazard, a poisonous gas generator and quite noisy. Find a place to position it in the yard away from burnable structures. For physical security, get a 6 foot long security chain with 5/16" thick links.
Third, I think we used 3 gallons of gasoline in 24 hours. I prefer to buy the gas just ahead of the storm and I pour it into the car if I don't use it. I keep the empty gas can in a galvanized trash can in the back of the yard.
Fourth, as pointed out earlier, oil burners are directly wired to a branch circuit in your house. The junction box I had was on the ceiling and in the dark of the basement. There are some wiring and safety problems you should figure out before you have an emergency. You need a hands free flashlight, a labeled circuit breaker, and a substitute power cord.
The wiring solution I used was as follows: A 100 foot 14 gage extension cord went from the generator to the house. Then I had a heavy duty "outlet tripler". One extension cord went down to the oil burner, one cord went to the kitchen, and one cord went to the TV and VCR. We had to unplug the refrigerator to run the oil burner.
The solution was noisy and I eventually put the generator in the basement stairwell with an 18" fan blowing fumes up the stairs. So while I had a minimal solution good enough for 2 winters, the system still presented risks.
I think by the third winter or third hurricane season I would have moved to a generator cut over switch and a safer and quieter generator.
Look at newspapers as essentially machines for adding value to pulped trees. As a previous poster said, news operations spend 1/3 of revenue buying the bulk paper.
About 20 years ago, as an "information systems inquiry" I looked at the total electricity consumption of people watching a television show in Los Angeles compared to the number of advertising dollars required to completely buy the same minutes of broadcast time.
It turns out, putting on a TV show is a great way to cause the consumption of a huge quantity of electricity.
The dollars spent by viewers on electricity to run their TVs exceeded the entire cost of operating the TV station by an integer multiple.
(The math is simple: get the ad rate from the TV station. They will tell you the size of the audience that station pulls at that time, from the ratings sweeps. Read your TV to see the watts it burns. Read electricity bill to find your price per KWH.)
Look at getting internet news: I am smoking up 176 watts on my PC plus I am paying Comcast 8 cents an hour to run their servers.
If the news organization could get just 10% of the money I spend receiving their "signal", or.8 cent per hour for my visit to their site, then 10,000 reader hours would be $17 per hour revenue.
That is the beginning of a reasonable business income stream iff the banks do not get in the way with "per transaction charges" that kill micro cash flows.
And for those who want a newspaper on paper, kick out the pdf on a printer that looks like a news or soda vending machine. Papers are now.75 each these days.
I'd extend the proposal: I'm dying to read Science Magazine (only two public libraries receive it in all of San Mateo County). But I would have to cough up $42 for what would at best amount to 12 hours a year of browsing.
What is the best blend of economic methods to deploy to bring about a low energy low carbon emission future?
Right now the low energy low carbon emission future mostly appears to cost too much or it works too slow:
Solar electricity costs somewhere above $.50/kwh. and high performance electric cars are priced like the Tesla.
So is a loan to the big 3 or a loan to Tesla the best way to build an electric car?
All of these organizations have the same problem: No cheap electric car energy storage device.
The problem is not motors, magnets, electronics or chassis design. It is the energy storage device.
The automotive field previously was stalled for decades at a time waiting for innovations to trickle past the patent wall and other production barriers. Examples include: engine valve alloys and spring steel technology, hydraulic brakes, the ignition system, the deep draw steel oil pan, the front wheel drive CV joint, the automatic transmission.
Now who is holding what patents for what?
I don't know. On my wife's Honda I see lots and lots of really ingenious good ideas. I see ideas that were never around when I was working on Studebakers.
It seems to me all these "new battery" technologies involve making a roll of electrodes and spacer materials, dunking the thing in some glop and then sealing the top of the battery can.
Somebody has a patent for some parts of this process and basically that means a 10 year delay while we wait for the patent to expire.
Remember how Xerox held up the xerographic copy machine business for 20 or 30 years with patents? How about the pharmaceutical companies and their high price AIDS drugs?
What possible government assistance is available if foreign organizations hold the key patents needed for mass production of electric car batteries? There is no clear path for the government to clear away patent obstacles.
So let's explore the development of the syllogism:
If patents and undisclosed business secret technologies block the production of traction electrical storage devices what would be the most productive government intervention and assistance?
Should the solution be a few ivory tower Manhattan projects given to a few prestigious colleges or should we try a massively parallel research project done at the next 10,000 perfectly worthy colleges and junior colleges in the USA.
The bio tech field has had a pattern where once a scientist got a feeling for specific set of chemical processes and results. The scientist would either join a company and file patents or the scientist and the University would file patents. The result is the people get high priced proprietary Aids drugs even though public funds paid the initial research.
So let's restructure the government intervention in the "electric traction energy storage device" problem so the result is a low cost, generic, easily repairable energy storage device... with the right kind of government research and development and prototype trials: Open research with no patents and full publication of results and methods.
The Prius engine uses a very high compression ratio. It has very high thermal efficency. The high thermal efficency of the Prius engine is a main element in the car's high mileage.
I compared an early Prius and an early Echo (same body, same engine displacement, and the compression ratio on the Prius was (I think 10.5:1 and the Echo had 9.5:1).
From a Physics textbook, the fundamental Carnot Engine efficency of the Prius is a lot higher simply because of the higher compression.
Therefore, a really big technical accomplishment is Toyota managed to make an extremely high compression gasoline engine.
Toyota engineers disguised or managed away the nasty characteristics of extremely high compresson engines: they are hard to start and they ping on cheap gas.
I have this itch to build a alow autonomous electric vehicle.
The missing components I need are a series of data structures and layers so I can make a connection from points on a google map to a data structure describing the next 50 feet of roadway or highway shoulder that my vehicle is going to traverse.
The thing that is true for all the autonomous vehicle concepts is: most of the problem is finding a complete open source software stack.
Right now i have these components. These things just cry out for a "Do something interesting with this stuff."
-- I have a salvaged junk electric wheelchair that works. The signals from the joystick appear on a single 10 or 12 pin plug.
---I have a GPS that makes track data. I can convert the track data to google maps.kml overlays.
---I have an (old) laptop with a USB web cam, usb input for the GPS, and a high power wireless 802g.11 networking card. It runs Ubuntu beautifully.
--- The wheelchair pulls 2 to 5 kilowatts when running at 6.5 miles per hour top speed. (It is real hard to hold an ammeter over the battery lead while driving down the street!)
--- A 100 watt peak solar panel runs $800. So the charge -to- run ratio is 20:1 to 50:1.
--- The job for the wheelchair is to fetch a bag of groceries I ordered by email.
--- Of course, the same little heap of computer stuff can be dropped into any converted vehicle. If you stick to a low power and low speed performance model, then the main problem is simply coming up with an open source software stack and an open, extensible set of data structures.
I propose, web apps are a toolset for doing a better blog.
Web apps are pieces to use in establishing an Internet communication situation.
I have a blog, I plan to put a spreadsheet in it, make the spreadsheet do calculations based on visitor inputs. maybe draw colored circles on a map based on visitor clicks on a map.
Maybe do a narrated slideshow presentation online, again this is where apps can help.
I see that I am beginning to build structures that need another layer to enable description and auditing.
The interesting thing about ZoHo is it looks like they are beginning to assemble the components for building a complete business "system" as a bunch of online components. Way beyond just a blog or just a communication solution made by an individual or small team.
I am using Google Apps Hacks by Philipp Lenssen as an idea book for my communication project. As titled it is mostly about the Google family of applications. The methods for linking are not restricted to google only.
It looks to me that Google will accept links to documents created using ZoHo, EditGrid, Thinkdfree and PbWiki.
----------
Separately, yes Google spreadsheet did strange things to my data when I deleted columns. For primary data crunching I find Gnumeric has better input data handling. Then I copy the presentation level spreadsheet up to Google docs.
DARPA is short for "Defense Advanced Research Projects Administration." yes?
DARPA is a United States Military organization yes?
Any useful AI project coming to your attention will be used for military purposes, yes?
Does any US government non-military organization have a budget similar to yours? Is anybody else shopping for AI ideas to help with the equally hard work of building and working toward peace?
Well, I have done a number of these projects...
on
DIY Solar Resources?
·
· Score: 1
One starting point for books and project ideas is the Whole Earth Catalog. I don't think there is a current edition but it should be in a library.
I used Farrington Daniels' book Direct Use of the Sun's Energy as a starting point for several solar energy projects.
He lays out the physics in their elegant simplicity and shows Direct Uses of it as the title says.
Some writers suggest there are fire hazards, especially if you use batteries in your project. Those warnings also indicate that as an advanced amateur you really do need access to the design and installation and safety information used by the professionals.
For solar projects, the Code Check plastic guides at least cover solar electric inverter and wiring in outline.
It irritates me that Building Codes are copyright and building permit texts generally not available online.
It is too bad that example building permits and guides to drafting and writing a project are not available online. As an advanced amateur, what I need is a couple of well thought out example projects.
Even the Building Code used by San Mateo County is copyright and can not be reproduced on the Internet, which would be very helpful for people who want to build a solar energy project.
Staggeringly apt opportunity for Larry Lessig
on
Lessig For Congress?
·
· Score: 1
larry Lessig lives in the 12th Congressional district. Until he died on Feb 11th, the congressman was Tom Lantos. Mr. Lantos is an extraordinary figure. I would say this is an appropriate vacancy suitable for a man who has performed distinguished advocacy for human rights and the intellectual commons over the encroachment of purely commercial business interests.
I agree, Larry Lessig would be a good candidate for this seat in Congress. If Davey Crockett wasn't copyright Disney corporation I'd say, lets adapt the tune and lyrics from the Davey Crockett Goes to Congress story for Larry Lessig.
Here are some of the organizations and companies that are in this district, I guess:
The district covers roughly the western half of San Francisco and the inland bay half of the San Francisco peninsula down to Redwood City. The district doesn't reach Stanford but it contains San Francisco State and Notre Dame de Nameur. Companies within the district include Oracle and some players in the secure mail and security businesses including Strongmail and Qualys (they host a LUG I attend). There are also a bunch of south of Market street web enterprises in the district ( I see them advertising on Craigslist ) and Youtube is in the district I think.
To my great disappointment, I am outside of his district by about 1000 yards. Darn.
The other part of aptness is the times. Tom Lantos is the only Holocaust survivor to serve in Congress.
A different kind of appropriation of the spirit and ideas of our times and society is now taking place. Copyright has been warped into a theft of the commons. The forces for commercial copyright are aggressively removing from the public commons for a term of years longer than our lifetimes virtually all telecast sporting events, concert content, movies, textbooks, most published scientific journals, home videos made of concerts and sporting events, and everything published in a newspaper.
Mr. Lessig took a case to the Supreme Court over this misappropriation due to copyright. The answer of the Supreme Court was: It is exclusively Congress that establishes the term and law of copyright. So the point is: go to Congress and change it. Yes, Lessig for Congress. Right time, right guy.
A step away from compatibility between distros?
on
Ubuntu Picks Upstart, KVM
·
· Score: 3, Insightful
When I moved from Red Hat to Debian (remember Bruce Peren's UserLinux ?) and to Ubuntu the thing I missed was the clarity of SysV init and the simple tools to add and remove programs from a runlevel.
The article quoted shows examples of upstart scripts. I don't quite see if compatibility with SysV init is a goal of upstart.
It sure would be nice if upstart means easier application sharing between Red Hat and Ubuntu.
I had a job delivering auto parts and I looked at it as a kind of a traveling salesman problem.
Besides 'shortest route' I was trying for safest route, cheapest route, and smallest total miles in a workday.
The total dataset of customers was 20 and a full load was at most 5 customers.
As mentioned above, the loop always worked best if traveled it in a clockwise path due to the difficulty of making left hand turns where traffic drives on the right side of the street.
Eventually, I worked out the shortest loop path but then business needs kept messing up the ideal route.
When I get back to (my spare time project of) studying topology, I think I will revisit the traveling salesman problem, and see if there are ways in which a Jordan curve (a line connecting nodes with no crossing over) might yield some interesting theorms tied to the combinational puzzles of the TSP.
Great askslashdot question. Great bunch of posts eh?
I have been fascinated for many years about knots.
From graph theory there is the Utility Pole problem. You have three houses A,B and C. You have three utility service points Water, Gas and Electric. The problem is: Can you connect all three utilities to all three houses with no wire crossing? The answer is no. Computer cabling can also be drawn as a planar graph. At some time some connection appears to require a crossing of a wire.
The idea I got out of the graph theory was: Hold the ends when working with cables. Coiling things tends to create twists. If you coil or fold the paired wires (not letting go of the ends) the structure you create does not form knots, and usually makes at worst easy to resolve tangles. Holding on to the ends forces folding and coiling to continuously have a kind of symmetry. Knots need the asymmetry of a loose end.
Here is Lee's super duper no tangle cable folding technique:
Hold the ends. Fold the cable, half, quarter etc. Wrap the folded cable with a scrap paper belly band. Tape the band. It is an easy to handle bundle: ends together tells type, easy to guess or calculate length.
Topology is a really interesting subject to study. My most recent enjoyable book:
Formal Knot Theory by Louis H. Kauffman, published by Dover, $14.95 paperback.
It is interesting that the "voting machine problem" is at the vertex of a lot of ideas and assertions that have the status of myth.
Myth - a governent can't design and build a robust, accurate and verifyable voting system. They will encumber the process with too much documentation or they wlll allow a design with design terms omitted.
Myth a private company (like Diebold) can't build a robust, accurate and auditable voting machine. They will knock out some chicken poop menu and script gadget that can be tweaked before the election and boguesd after the vote.
Now if you take these two myths and hold them in your mind together, there begins to emerge some huge propositions and problems.
One proposition I see is: Why in the past two presidential elections did it seem like the more momemtous the difference between the two candidates the margin of popular vote between the candidates got smaller?
Check me on this: The margin of victory narrowed between the Democratic and Republican candidates as the days to election day dwindled.
The narrowness of the election victory in many places created many opportunities where small scale vote fraud and deployment of technicalities flipped the electoral vote.
The discussion of election mechanics taking place here is an attempt to mechanically resolve a really basic problem with democracy, the guy you didn't vote for gets elected.
What could we do to the election process itself to improve the assessment and judgments preceding the vote? Presently we need a voting system that is auditable to 1 vote in a billion.
Peter Drucker has written books on how to select a qualified chief executive. Why can't we implement some of those good ideas in selecting a President? Why do we still have "Swift boating" as a major component of the presidential selection process?
How might we better understand the founding fathers rather ambivalent approach to voting itself? They took pains to slow down the feedback loops with a long term for the presidency, staggered terms for senators, and no national vote at all on referendum items.
One cause of quality problems in the voting system is we don't do it often enough with full auditing and quality control. The founding fathers put that time delay into the voting system design.
From cybernetics, feedback theory and 200 years of development as a nation how might we guardedly implement more and more frequent elections to bring about "more democracy'?
On the concern about accuracy in elections, note that voters of the United States reelected the presently sitting president. What would it mean if we had a vote with 1:1billion count accuracy now? What it means is the 48% of losing voters have to credit the winner with the opportunities of the office. So the winners want a mathematically accurate count so they can claim the entire electorate's approbation.
Looking at elections involving the current president, his second victory came to him because during wars voters choose continuity. Roosevelt's second and third victories were wartime continuity decisions. Hmm, how about Lincoln and Woodrow Wilson? Were those also wartime continuity presidents?
So, is the election accuracy problem a technical manifestation of the anxiety we are feeling as a nation? We have the biggest military in the world, projecting force every day on every ocean and on most continents. There is this deep uncertainty that is repeatedly being invoked to keep repeatedly choosing "the slightly more conservative - slightly more aggressive - slightly more nationalistic" leadership. As we slide down a slope to the swamp of empire.
The paradox of the democratic election process is we keep wedging ourselves gradually into a set of policies favoring empire. Gee, like Athens in the Peloponesian wars?
Autonomous vehicle tech is needed for ultra low carbon emission vehicles. Slow vehicles need autonomous abilities because simple labor cost economics require it. --- This is another take on why the DARPA autonomous vehicle development is important.
It is relatively easy and feasible to build a 2 to 10 mile per hour freight vehicle that is "green" or ultra low carbon emissions, (compared to building a vehicle with 50 to 65 miles per hour speed ability).
But when the average speed of a vehicle drops below 35 miles per hour, then the labor cost of the driver operating the vehicle exceeds the non-labor operating cost of the vehicle.
I can envision existing trucks refitted with a 1500 watt solar roof (that's only 1.5 horsepower now), with some batteries, maybe electric braking and boost technology, with the existing engine and tranny left in place, about 100 watts of computer power. It could haul a load maybe at 1 to 10 miles per hour with maybe 1/5th or 1/3 the carbon emissions. There is no money to pay a driver to sit around and play cards in the cab while this puppy ambles down interstate 5.
Autonomous vehicle technology also could serve for "meals on wheels" and grocery delivery and rural mail delivery solutions. This entire class of solutions is dominated by labor costs that easily exceed the delivery vehicle operating costs.
Remember the Erie Canal? That was a slow and low energy transportation solution, autonomous vehicles can change the equilibrium and allow some of those solutions to start up again.
Here are two weaknesses and two strengths of this book:
It seems to me the Official Ubuntu book is weak on helping a user figure out what she is going to do with the installed system.
For instancce, how would a new Ubuntu user backup her Windows files to CD ? Like transporting all your email and browser bookmarks from Windows to Ubuntu? Not covered. The focus of this book fades at the boundaries of Ubuntu itself.
Another weakness is, this book talks about Ubuntu but doesn't show a simple task like how to backup a/home partition to CD. Simple backups are not covered, as far as I can tell from the index.
The strength of The Official Ubuntu Book is it describes the Ubuntu philosophy and commitment to creating highly usable free software with multiple growing communities of users contributing back to build a usable set of tools.
The Ubuntu accomplishment is social: Ubuntu is a suite of free software deliberately giving computer power to newcomers, students and non-technical users. Ubuntu should have it's own book.
If you are a techie, you can look at this book as something you would use socially the same as you might give a Ubuntu CD or DVD to a potential Ubuntu user.
I have written with some anguish about weaknesses with Ubuntu (see two articles in my Slashdot journal) but the balancing factor is: Ubuntu is free in all four senses defined by Richard Stallman.
Here is a review of Ubuntu Hacks by a Peninsula Linux Computer Group member, this is another Ubuntu book with a more technical focus:
Post your solutions to pre-empt patent grabs
on
Trans-Atlantic Robots
·
· Score: 0, Offtopic
While you do your project, I would urge you to post your solutions in great detail to push lots of the patentable material into the realm of prior art. --- My perception is all societies will need zero carbon emitting low energy consumption autonomous vehicles. Your sailboat is exactly the vehicle needed for the low carbon footprint future. ---- I have been thinking about autonomous vehicles operating in the 1 to 4 mile per hour speed range. These vehicles will need a vehicle to vehicle communication process.
Someone recommended I review the one-laptop-per-child project. The OLPC laptop has ad-hoc peer to peer wireless networking built in. The laptop has a waterproof keyboard too.
Your sailboat should have the ability to communicate with nearby boats, right? Avoid collision, exchange wind and wave data, describe location and direction of travel, receive collision warnings, dodge oncoming tankers, heave to when being boarded at the end of the trip. For initial development, you will want a SSH link to a nearby boat and you will want to see what your boat is using as navigation inputs and rudder and sail angle control outputs.
Slow moving trucks going nose-to-tail down a highway will need a similar vocabulary to pass trip and safety information from vehicle to vehicle.
All this top level stuff should be available in a free software format. I mean should in the sense that the message language must be open. And second should in the sense that part of the problem has been addressed already. Perhaps reviewing kugle.com source code research engine might help you find source code. Perhaps there is a DARPA autonomous vehicle team that has written a vehicle to vehicle message exchange language.
How about a mast-top high brightness LED navigation light that broadcasts high speed data to other ships nearby? Sending out UDP packets in light flashes? Route the OLPC ad-hoc networking data stream up to an led and photo-transistor detector device.
The political strategist Thomas P.M. Barnett suggested that "side effects" of American policies with respect to the rest of the world in many cases may dwarf the explicitly observed and worried over effects of an explicitly defined problem like Al Quaeda or the resolution of the conflict in Iraq.
It looks to me like the foreign graduate student influx into the US graduate study system is one of those "side effect" processes. Question: how many foreign graduate students become permanent US Residents or immigrants every year? What is the net flow of people into the US?
The slashdot postings here have brought out an anecdotal explanation of why the grad student influx exists and what are the effects it causes within the US.
The recently proposed and rejected Immigration Bill in the US Congress explicitly favored providing immigration preference for graduates holding advanced degrees.
As reported here: We are pulling bright well educated people from around the world into American employment.
I have been recently trying to understand: Why does the US continue to be "wealthy"? How does the influx of bright well educated people contribute to the "wealth" of the USA?
There are factors like the US trade deficit and US military spending that suggest "wealth" (as cash flow to overseas destinations) is flowing out of the country.
Any suggestions for further reading on this "wealth" and "influx of grad students" issue?
You know, this case reveals a glaring crack in the foundations of enlightened self interest capitalism.
Another society (other than the USA) that could figure out how to rapidly, reasonably and cheaply license patents an copyrighted works could economically outperform the USA.
The multi million dollar amount of the patent settlement is relatively small compared to the economic diversion and damage done as the patent holder blocks the development of voip telephony. Essentially, the patent system is operating to block social change.
"blocking social change" needs to be appreciated. In it's positive manifestation it is the comfort and stability of the American System. There actually was a politicial platform called the "American System" described and advanced in the ~1880's. That system is a precursor to our current system.
After all, we all want a job to pay our mortgage and save up for our retirement and our kid's college tuition (partially, the loans begin in the Junior year). The stability of our American System is maintained by many interlocking mercantile compromises.
The problem has been patent and copyright lawyers have continuously blocked the writing of law to force reasonably priced licenses.
How about we require all patent an license fees to not exceed 5% of the dollar amount paid for the energy reequired to operate the patented device for a year. Only for registered, fee paid patents and copyrights for only 16 years from the date of issue of the patent.
Put the patent system and the vaunted value of copyrightable creative work in it's place: conceiving of something and building the first working prototype is just a tiny fraction of the lifetime costs of any invention or innovation. Lots and lots of independent re-invention happens. Let the lucky first inventor get a sliver of the follow on revenue if he is willing to get out there, charge consulting fees, and help out other people instead of sitting in a coffee shop with a lawyer scheming on how much money they can extract from other businesses.
For further thoughts, see Don Lancaster's web site and anti-patent comments. Janis Ian's website has a penetrating analysis of the music business. The big thinker of the 70's on this issue was Peter Drucker, a popularizer of knowledge based exports as a business advantage for highly educated technically advanced societies. (Planners in China and India read his books.). And of course there is Richard Stallman's free software. Essentially he has created a reasonably priced copyright license. He has opened a window into the next economic universe.
Thinking in terms of what social system will be economically superior to the American System, that next generation social framework will clearly surge ahead of us if they can design a clean, effective and cheap patent and copyright process.
The problems in management are often rubbery, mysterious misshapen things. The actions you take in response to these problems will reflect your unconscious and conscious assumptions about what your work is.
What Peter Drucker did during his life is crystalize some of these problems and find positions of strength and humanity that a manager can actually express at work.
Until another thinker appears, I would urge you to get every Drucker book you can from the library. Scan them and you will see the evolution of business thinking since WWII. Give at least one of the last of his books a good reading.
His studies of how to hire, how to fire and what counts in employee performance are still great. His recognition and exploration of the problem of performance in a non-profit organization is important.
I hope we will be fortunate enough to hear another management thinker of such scope.
What is the business outcome and advantage for Sun with this open hardware move?
It looks to me like Sun has figured out that they are in a knowledge business. They are operating on an information theory algorithm. They are creating a much larger pyramid of customers for their particular computer knowledge. There is a new enormous bottom layer of people contemplating using this fascinating powerful chunk of information. They are emitting information, not hardware.
The thing from information theory is: The more high quality information a source emits, the more valuable the source becomes. Sun still has the stable of PhD researchers from U.C. Berkeley and some more wags from Stanford. So the company will continue in the business of emitting information.
If you want the Macintosh of mainframes, they will sell them to you. If you want to boot Solaris or wire up Sparc chips you are still their customer. Sun will be your first publisher, web site and consultant.
It seems to me that this is a business innovation. It has been 60 years since Shannon's information theory paper suggested that the source that emits information increases in entropy. Sun is doing that by making available a uniquely sophisticated design - not hiding it in file cabinets in the basement.
Slashdot Mirror
Here is a book that shows the field of market prediction mathematics will not directly make a person money by playing the market.
But on the other hand, it is an area of mathematical study that involves the most exciting developments in applied mathematics in the last 50 years, namely chaos and fractal theory.
"The (Mis)Behavior of Markets. A Fractal View of Risk, Ruin, and Reward" by Benoit Mandelbrot and Richard L. Hudson, Basic Books, New York, c. 2004.
This is a really good question and it points to the need for more utilities to check and fix bit level errors in files and memory images.
I wish there was a set of Linux utilities for generating and using error correction checksums.
The usual checksum tool simply reports if a file has an error.
What I would like to see is a checksum tool capable of fixing a multi-byte, single byte or single bit error.
These are checksums using Hammung codes I think?
I strongly suspect that huge hunks of don't care data like JPG photos and DVI movies develop bit level errors during storage, handling, copying, and editing.
A utility like gpg will spot errors. And diff spots character level errors. But they are both real clumsy at finding and precisely repairing individual bits. With diff you wind up unsure if your reference or original file changed.
This Linux system I am writing on had a bad RAN memory bit and the random number generator loaded over that location.
The random number generator didn't work but Linux mostly ran OK. I thought it was a software problem for several weeks. Finally after replacing the random number code and even checking bug reports, I finally looked at the RAM memory. Bing!
Check out the "avoid patents" development strategy and tactics that has been developed and tested by technical writer and engineer Don Lancaster.
Don wrote "CMOS Cookbook" around 1973 and he is still active selling some specialized products, books, EBay stuff and more.
His strategy is: avoid patents, build a kit or add on first, stay below the radar of the big players, treat the project as part of a process rather than a one time bonanza.
www.tinaja.com
en.wikipedia.org/wiki/Don_Lancaster
I recommend you get your school library to carry The World of Mathematics edited by Newman, it was first published in the '50's.
I discovered this 4 volume anthology at my local Los Angeles Public library branch in the 70's and I wound up hunting for years for a used copy of the anthology. In the mid-80's it was reprinted in paperback.
It is a 4 volume anthology containing selected essays and articles about every important field of mathematics.
Really enjoyable selections in this book include a terrific introduction to double entry accounting, life insurance, the Seven Bridges of Konigsberg, early papers on Turing machines, completeness theorems, codes and codebreaking and information theory.
Two more inexpensive and interesting books are:
Great Ideas in Operational Research, published by Dover.
and
Formal Knot Theory by Kauffman.
The painful part about having Microsoft computers in a special education classroom is the entire Windows interface is wrong for special ed kids.
On a Linux box, a web browser could switch on an age and ability appropriate interface model.
I work in Special Education classrooms with all kinds of disabled kids. The PC interface is awful.
The Linux desktops just shuffle some choices, but Linux could be radically better.
By interface, I mean the mouse, the left button, the right button, the roller, the keystrokes, the cursor, the active area on the screen. The interface is too chopped up, too many choices, and too easy to disrupt. The interface requires too many ideas linked to isolated fine motor actions.
----
The most successful interface for special ed kids is the teletubbies web games on pbs.org (where the youth can play peek-a-boo using a pushbutton wired to the left mouse button). The other success is the Stellaluna kids book on CD. It can be used where the adult holds the mouse and the youth pushes a button wired to the left mouse. It also models the success of musically modulated speech to enhance communication with kids who parse speech slowly.
-------
But for these applications, the adult interface surrounding the application is still the problem.
------------
What I would like to see is research based interface protocols for specific levels of receptive language (like 1 year old), visual processing and motor ability (like can use hand but can't isolate finger).
The interface protocol would be like a style sheet, it masks and configures the user's mouse, display, speakers, and keyboard.
This would be great as a web app. You could search the web for apps that can be presented to your student with ability appropriate interface.
Applications for special ed kids are very few in number, the ones I know are 1990 vintage Windows/MacIntosh programs implemented with a slide show toolkit.
An example is "Teen Tunes". It is still a "CD must be in the drive" program with a '90's copyright date. It sells for over $100. Like other special ed programs, this one is a frozen experiment in interface simplification. The aide still has to hold the mouse and center the cursor while the student pushes a button.
Here are a few safety points that I leaned from using a generator during a hurricane and ice storm in Maryland. I did an emergency family power solution using a 3500 watt Craftsman generator.
First, buy and test your generator before you need it. The generator I got had a missing needle valve in the carburetor. Lucky I realized it was gasoline and not rain before I pulled the starter rope.
Second, the generator is a fire hazard, a poisonous gas generator and quite noisy. Find a place to position it in the yard away from burnable structures. For physical security, get a 6 foot long security chain with 5/16" thick links.
Third, I think we used 3 gallons of gasoline in 24 hours. I prefer to buy the gas just ahead of the storm and I pour it into the car if I don't use it. I keep the empty gas can in a galvanized trash can in the back of the yard.
Fourth, as pointed out earlier, oil burners are directly wired to a branch circuit in your house. The junction box I had was on the ceiling and in the dark of the basement. There are some wiring and safety problems you should figure out before you have an emergency. You need a hands free flashlight, a labeled circuit breaker, and a substitute power cord.
The wiring solution I used was as follows: A 100 foot 14 gage extension cord went from the generator to the house. Then I had a heavy duty "outlet tripler". One extension cord went down to the oil burner, one cord went to the kitchen, and one cord went to the TV and VCR. We had to unplug the refrigerator to run the oil burner.
The solution was noisy and I eventually put the generator in the basement stairwell with an 18" fan blowing fumes up the stairs. So while I had a minimal solution good enough for 2 winters, the system still presented risks.
I think by the third winter or third hurricane season I would have moved to a generator cut over switch and a safer and quieter generator.
Look at newspapers as essentially machines for adding value to pulped trees. As a previous poster said, news operations spend 1/3 of revenue buying the bulk paper.
About 20 years ago, as an "information systems inquiry" I looked at the total electricity consumption of people watching a television show in Los Angeles compared to the number of advertising dollars required to completely buy the same minutes of broadcast time.
It turns out, putting on a TV show is a great way to cause the consumption of a huge quantity of electricity.
The dollars spent by viewers on electricity to run their TVs exceeded the entire cost of operating the TV station by an integer multiple.
(The math is simple: get the ad rate from the TV station. They will tell you the size of the audience that station pulls at that time, from the ratings sweeps. Read your TV to see the watts it burns. Read electricity bill to find your price per KWH.)
Look at getting internet news: I am smoking up 176 watts on my PC plus I am paying Comcast 8 cents an hour to run their servers.
If the news organization could get just 10% of the money I spend receiving their "signal", or .8 cent per hour for my visit to their site, then 10,000 reader hours would be $17 per hour revenue.
That is the beginning of a reasonable business income stream iff the banks do not get in the way with "per transaction charges" that kill micro cash flows.
And for those who want a newspaper on paper, kick out the pdf on a printer that looks like a news or soda vending machine. Papers are now .75 each these days.
I'd extend the proposal: I'm dying to read Science Magazine (only two public libraries receive it in all of San Mateo County). But I would have to cough up $42 for what would at best amount to 12 hours a year of browsing.
The big framework question is:
What is the best blend of economic methods to deploy to bring about a low energy low carbon emission future?
Right now the low energy low carbon emission future mostly appears to cost too much or it works too slow:
Solar electricity costs somewhere above $.50/kwh. and high performance electric cars are priced like the Tesla.
So is a loan to the big 3 or a loan to Tesla the best way to build an electric car?
All of these organizations have the same problem: No cheap electric car energy storage device.
The problem is not motors, magnets, electronics or chassis design. It is the energy storage device.
The automotive field previously was stalled for decades at a time waiting for innovations to trickle past the patent wall and other production barriers. Examples include: engine valve alloys and spring steel technology, hydraulic brakes, the ignition system, the deep draw steel oil pan, the front wheel drive CV joint, the automatic transmission.
Now who is holding what patents for what?
I don't know. On my wife's Honda I see lots and lots of really ingenious good ideas. I see ideas that were never around when I was working on Studebakers.
It seems to me all these "new battery" technologies involve making a roll of electrodes and spacer materials, dunking the thing in some glop and then sealing the top of the battery can.
Somebody has a patent for some parts of this process and basically that means a 10 year delay while we wait for the patent to expire.
Remember how Xerox held up the xerographic copy machine business for 20 or 30 years with patents? How about the pharmaceutical companies and their high price AIDS drugs?
What possible government assistance is available if foreign organizations hold the key patents needed for mass production of electric car batteries? There is no clear path for the government to clear away patent obstacles.
So let's explore the development of the syllogism:
If patents and undisclosed business secret technologies block the production of traction electrical storage devices what would be the most productive government intervention and assistance?
Should the solution be a few ivory tower Manhattan projects given to a few prestigious colleges or should we try a massively parallel research project done at the next 10,000 perfectly worthy colleges and junior colleges in the USA.
The bio tech field has had a pattern where once a scientist got a feeling for specific set of chemical processes and results. The scientist would either join a company and file patents or the scientist and the University would file patents. The result is the people get high priced proprietary Aids drugs even though public funds paid the initial research.
So let's restructure the government intervention in the "electric traction energy storage device" problem so the result is a low cost, generic, easily repairable energy storage device ... with the right kind of government research and development and prototype trials: Open research with no patents and full publication of results and methods.
I want to add to your comment about the Prius.
The Prius engine uses a very high compression ratio. It has very high thermal efficency. The high thermal efficency of the Prius engine is a main element in the car's high mileage.
I compared an early Prius and an early Echo (same body, same engine displacement, and the compression ratio on the Prius was (I think 10.5:1 and the Echo had 9.5:1).
From a Physics textbook, the fundamental Carnot Engine efficency of the Prius is a lot higher simply because of the higher compression.
Therefore, a really big technical accomplishment is Toyota managed to make an extremely high compression gasoline engine.
Toyota engineers disguised or managed away the nasty characteristics of extremely high compresson engines: they are hard to start and they ping on cheap gas.
I have this itch to build a alow autonomous electric vehicle.
The missing components I need are a series of data structures and layers so I can make a connection from points on a google map to a data structure describing the next 50 feet of roadway or highway shoulder that my vehicle is going to traverse.
The thing that is true for all the autonomous vehicle concepts is: most of the problem is finding a complete open source software stack.
Right now i have these components. These things just cry out for a "Do something interesting with this stuff."
-- I have a salvaged junk electric wheelchair that works. The signals from the joystick appear on a single 10 or 12 pin plug.
---I have a GPS that makes track data. I can convert the track data to google maps .kml overlays.
---I have an (old) laptop with a USB web cam, usb input for the GPS, and a high power wireless 802g.11 networking card. It runs Ubuntu beautifully.
--- The wheelchair pulls 2 to 5 kilowatts when running at 6.5 miles per hour top speed. (It is real hard to hold an ammeter over the battery lead while driving down the street!)
--- A 100 watt peak solar panel runs $800. So the charge -to- run ratio is 20:1 to 50:1.
--- The job for the wheelchair is to fetch a bag of groceries I ordered by email.
--- Of course, the same little heap of computer stuff can be dropped into any converted vehicle. If you stick to a low power and low speed performance model, then the main problem is simply coming up with an open source software stack and an open, extensible set of data structures.
I propose, web apps are a toolset for doing a better blog.
Web apps are pieces to use in establishing an Internet communication situation.
I have a blog, I plan to put a spreadsheet in it, make the spreadsheet do calculations based on visitor inputs. maybe draw colored circles on a map based on visitor clicks on a map.
Maybe do a narrated slideshow presentation online, again this is where apps can help.
I see that I am beginning to build structures that need another layer to enable description and auditing.
The interesting thing about ZoHo is it looks like they are beginning to assemble the components for building a complete business "system" as a bunch of online components. Way beyond just a blog or just a communication solution made by an individual or small team.
I am using Google Apps Hacks by Philipp Lenssen as an idea book for my communication project. As titled it is mostly about the Google family of applications. The methods for linking are not restricted to google only.
It looks to me that Google will accept links to documents created using ZoHo, EditGrid, Thinkdfree and PbWiki.
----------
Separately, yes Google spreadsheet did strange things to my data when I deleted columns. For primary data crunching I find Gnumeric has better input data handling. Then I copy the presentation level spreadsheet up to Google docs.
DARPA is a United States Military organization yes?
Any useful AI project coming to your attention will be used for military purposes, yes?
Does any US government non-military organization have a budget similar to yours? Is anybody else shopping for AI ideas to help with the equally hard work of building and working toward peace?
I used Farrington Daniels' book Direct Use of the Sun's Energy as a starting point for several solar energy projects.
He lays out the physics in their elegant simplicity and shows Direct Uses of it as the title says.
Some writers suggest there are fire hazards, especially if you use batteries in your project. Those warnings also indicate that as an advanced amateur you really do need access to the design and installation and safety information used by the professionals.
For solar projects, the Code Check plastic guides at least cover solar electric inverter and wiring in outline.
It irritates me that Building Codes are copyright and building permit texts generally not available online.
It is too bad that example building permits and guides to drafting and writing a project are not available online. As an advanced amateur, what I need is a couple of well thought out example projects.
Even the Building Code used by San Mateo County is copyright and can not be reproduced on the Internet, which would be very helpful for people who want to build a solar energy project.
larry Lessig lives in the 12th Congressional district. Until he died on Feb 11th, the congressman was Tom Lantos. Mr. Lantos is an extraordinary figure. I would say this is an appropriate vacancy suitable for a man who has performed distinguished advocacy for human rights and the intellectual commons over the encroachment of purely commercial business interests.
I agree, Larry Lessig would be a good candidate for this seat in Congress. If Davey Crockett wasn't copyright Disney corporation I'd say, lets adapt the tune and lyrics from the Davey Crockett Goes to Congress story for Larry Lessig.
Here are some of the organizations and companies that are in this district, I guess:
The district covers roughly the western half of San Francisco and the inland bay half of the San Francisco peninsula down to Redwood City. The district doesn't reach Stanford but it contains San Francisco State and Notre Dame de Nameur. Companies within the district include Oracle and some players in the secure mail and security businesses including Strongmail and Qualys (they host a LUG I attend). There are also a bunch of south of Market street web enterprises in the district ( I see them advertising on Craigslist ) and Youtube is in the district I think.
http://www.sen.ca.gov/ftp/sen/cngplan/PDF_CD_ATLAS/cd12_new.pdf
To my great disappointment, I am outside of his district by about 1000 yards. Darn.
The other part of aptness is the times. Tom Lantos is the only Holocaust survivor to serve in Congress.
A different kind of appropriation of the spirit and ideas of our times and society is now taking place. Copyright has been warped into a theft of the commons. The forces for commercial copyright are aggressively removing from the public commons for a term of years longer than our lifetimes virtually all telecast sporting events, concert content, movies, textbooks, most published scientific journals, home videos made of concerts and sporting events, and everything published in a newspaper.
Mr. Lessig took a case to the Supreme Court over this misappropriation due to copyright. The answer of the Supreme Court was: It is exclusively Congress that establishes the term and law of copyright. So the point is: go to Congress and change it. Yes, Lessig for Congress. Right time, right guy.
When I moved from Red Hat to Debian (remember Bruce Peren's UserLinux ?) and to Ubuntu the thing I missed was the clarity of SysV init and the simple tools to add and remove programs from a runlevel.
The article quoted shows examples of upstart scripts. I don't quite see if compatibility with SysV init is a goal of upstart.
It sure would be nice if upstart means easier application sharing between Red Hat and Ubuntu.
I had a job delivering auto parts and I looked at it as a kind of a traveling salesman problem.
Besides 'shortest route' I was trying for safest route, cheapest route, and smallest total miles in a workday.
The total dataset of customers was 20 and a full load was at most 5 customers.
As mentioned above, the loop always worked best if traveled it in a clockwise path due to the difficulty of making left hand turns where traffic drives on the right side of the street.
Eventually, I worked out the shortest loop path but then business needs kept messing up the ideal route.
When I get back to (my spare time project of) studying topology, I think I will revisit the traveling salesman problem, and see if there are ways in which a Jordan curve (a line connecting nodes with no crossing over) might yield some interesting theorms tied to the combinational puzzles of the TSP.
Great askslashdot question. Great bunch of posts eh?
I have been fascinated for many years about knots.
From graph theory there is the Utility Pole problem. You have three houses A,B and C. You have three utility service points Water, Gas and Electric. The problem is: Can you connect all three utilities to all three houses with no wire crossing? The answer is no. Computer cabling can also be drawn as a planar graph. At some time some connection appears to require a crossing of a wire.
The idea I got out of the graph theory was: Hold the ends when working with cables. Coiling things tends to create twists. If you coil or fold the paired wires (not letting go of the ends) the structure you create does not form knots, and usually makes at worst easy to resolve tangles. Holding on to the ends forces folding and coiling to continuously have a kind of symmetry. Knots need the asymmetry of a loose end.
Here is Lee's super duper no tangle cable folding technique:
Hold the ends. Fold the cable, half, quarter etc. Wrap the folded cable with a scrap paper belly band. Tape the band. It is an easy to handle bundle: ends together tells type, easy to guess or calculate length.
Topology is a really interesting subject to study. My most recent enjoyable book:
Formal Knot Theory by Louis H. Kauffman, published by Dover, $14.95 paperback.
It is interesting that the "voting machine problem" is at the vertex of a lot of ideas and assertions that have the status of myth.
Myth - a governent can't design and build a robust, accurate and verifyable voting system. They will encumber the process with too much documentation or they wlll allow a design with design terms omitted.
Myth a private company (like Diebold) can't build a robust, accurate and auditable voting machine. They will knock out some chicken poop menu and script gadget that can be tweaked before the election and boguesd after the vote.
Now if you take these two myths and hold them in your mind together, there begins to emerge some huge propositions and problems.
One proposition I see is: Why in the past two presidential elections did it seem like the more momemtous the difference between the two candidates the margin of popular vote between the candidates got smaller?
Check me on this: The margin of victory narrowed between the Democratic and Republican candidates as the days to election day dwindled.
The narrowness of the election victory in many places created many opportunities where small scale vote fraud and deployment of technicalities flipped the electoral vote.
The discussion of election mechanics taking place here is an attempt to mechanically resolve a really basic problem with democracy, the guy you didn't vote for gets elected.
What could we do to the election process itself to improve the assessment and judgments preceding the vote? Presently we need a voting system that is auditable to 1 vote in a billion.
Peter Drucker has written books on how to select a qualified chief executive. Why can't we implement some of those good ideas in selecting a President? Why do we still have "Swift boating" as a major component of the presidential selection process?
How might we better understand the founding fathers rather ambivalent approach to voting itself? They took pains to slow down the feedback loops with a long term for the presidency, staggered terms for senators, and no national vote at all on referendum items.
One cause of quality problems in the voting system is we don't do it often enough with full auditing and quality control. The founding fathers put that time delay into the voting system design.
From cybernetics, feedback theory and 200 years of development as a nation how might we guardedly implement more and more frequent elections to bring about "more democracy'?
On the concern about accuracy in elections, note that voters of the United States reelected the presently sitting president. What would it mean if we had a vote with 1:1billion count accuracy now? What it means is the 48% of losing voters have to credit the winner with the opportunities of the office. So the winners want a mathematically accurate count so they can claim the entire electorate's approbation.
Looking at elections involving the current president, his second victory came to him because during wars voters choose continuity. Roosevelt's second and third victories were wartime continuity decisions. Hmm, how about Lincoln and Woodrow Wilson? Were those also wartime continuity presidents?
So, is the election accuracy problem a technical manifestation of the anxiety we are feeling as a nation? We have the biggest military in the world, projecting force every day on every ocean and on most continents. There is this deep uncertainty that is repeatedly being invoked to keep repeatedly choosing "the slightly more conservative - slightly more aggressive - slightly more nationalistic" leadership. As we slide down a slope to the swamp of empire.
The paradox of the democratic election process is we keep wedging ourselves gradually into a set of policies favoring empire. Gee, like Athens in the Peloponesian wars?
Autonomous vehicle tech is needed for ultra low carbon emission vehicles. Slow vehicles need autonomous abilities because simple labor cost economics require it.
---
This is another take on why the DARPA autonomous vehicle development is important.
It is relatively easy and feasible to build a 2 to 10 mile per hour freight vehicle that is "green" or ultra low carbon emissions, (compared to building a vehicle with 50 to 65 miles per hour speed ability).
But when the average speed of a vehicle drops below 35 miles per hour, then the labor cost of the driver operating the vehicle exceeds the non-labor operating cost of the vehicle.
I can envision existing trucks refitted with a 1500 watt solar roof (that's only 1.5 horsepower now), with some batteries, maybe electric braking and boost technology, with the existing engine and tranny left in place, about 100 watts of computer power. It could haul a load maybe at 1 to 10 miles per hour with maybe 1/5th or 1/3 the carbon emissions. There is no money to pay a driver to sit around and play cards in the cab while this puppy ambles down interstate 5.
Autonomous vehicle technology also could serve for "meals on wheels" and grocery delivery and rural mail delivery solutions. This entire class of solutions is dominated by labor costs that easily exceed the delivery vehicle operating costs.
Remember the Erie Canal? That was a slow and low energy transportation solution, autonomous vehicles can change the equilibrium and allow some of those solutions to start up again.
Here are two weaknesses and two strengths of this book:
/home partition to CD. Simple backups are not covered, as far as I can tell from the index.
It seems to me the Official Ubuntu book is weak on helping a user figure out what she is going to do with the installed system.
For instancce, how would a new Ubuntu user backup her Windows files to CD ? Like transporting all your email and browser bookmarks from Windows to Ubuntu? Not covered. The focus of this book fades at the boundaries of Ubuntu itself.
Another weakness is, this book talks about Ubuntu but doesn't show a simple task like how to backup a
The strength of The Official Ubuntu Book is it describes the Ubuntu philosophy and commitment to creating highly usable free software with multiple growing communities of users contributing back to build a usable set of tools.
The Ubuntu accomplishment is social: Ubuntu is a suite of free software deliberately giving computer power to newcomers, students and non-technical users. Ubuntu should have it's own book.
If you are a techie, you can look at this book as something you would use socially the same as you might give a Ubuntu CD or DVD to a potential Ubuntu user.
I have written with some anguish about weaknesses with Ubuntu (see two articles in my Slashdot journal) but the balancing factor is: Ubuntu is free in all four senses defined by Richard Stallman.
Here is a review of Ubuntu Hacks by a Peninsula Linux Computer Group member, this is another Ubuntu book with a more technical focus:
http://www.penlug.org/twiki/bin/view/Main/LinuxBookReviewsUbuntuHack
While you do your project, I would urge you to post your solutions in great detail to push lots of the patentable material into the realm of prior art.
---
My perception is all societies will need zero carbon emitting low energy consumption autonomous vehicles. Your sailboat is exactly the vehicle needed for the low carbon footprint future.
----
I have been thinking about autonomous vehicles operating in the 1 to 4 mile per hour speed range. These vehicles will need a vehicle to vehicle communication process.
Someone recommended I review the one-laptop-per-child project. The OLPC laptop has ad-hoc peer to peer wireless networking built in. The laptop has a waterproof keyboard too.
Your sailboat should have the ability to communicate with nearby boats, right? Avoid collision, exchange wind and wave data, describe location and direction of travel, receive collision warnings, dodge oncoming tankers, heave to when being boarded at the end of the trip. For initial development, you will want a SSH link to a nearby boat and you will want to see what your boat is using as navigation inputs and rudder and sail angle control outputs.
Slow moving trucks going nose-to-tail down a highway will need a similar vocabulary to pass trip and safety information from vehicle to vehicle.
All this top level stuff should be available in a free software format. I mean should in the sense that the message language must be open. And second should in the sense that part of the problem has been addressed already. Perhaps reviewing kugle.com source code research engine might help you find source code. Perhaps there is a DARPA autonomous vehicle team that has written a vehicle to vehicle message exchange language.
How about a mast-top high brightness LED navigation light that broadcasts high speed data to other ships nearby? Sending out UDP packets in light flashes? Route the OLPC ad-hoc networking data stream up to an led and photo-transistor detector device.
The political strategist Thomas P.M. Barnett suggested that "side effects" of American policies with respect to the rest of the world in many cases may dwarf the explicitly observed and worried over effects of an explicitly defined problem like Al Quaeda or the resolution of the conflict in Iraq.
It looks to me like the foreign graduate student influx into the US graduate study system is one of those "side effect" processes. Question: how many foreign graduate students become permanent US Residents or immigrants every year? What is the net flow of people into the US?
The slashdot postings here have brought out an anecdotal explanation of why the grad student influx exists and what are the effects it causes within the US.
The recently proposed and rejected Immigration Bill in the US Congress explicitly favored providing immigration preference for graduates holding advanced degrees.
As reported here: We are pulling bright well educated people from around the world into American employment.
I have been recently trying to understand: Why does the US continue to be "wealthy"? How does the influx of bright well educated people contribute to the "wealth" of the USA?
There are factors like the US trade deficit and US military spending that suggest "wealth" (as cash flow to overseas destinations) is flowing out of the country.
Any suggestions for further reading on this "wealth" and "influx of grad students" issue?
You know, this case reveals a glaring crack in the foundations of enlightened self interest capitalism.
Another society (other than the USA) that could figure out how to rapidly, reasonably and cheaply license patents an copyrighted works could economically outperform the USA.
The multi million dollar amount of the patent settlement is relatively small compared to the economic diversion and damage done as the patent holder blocks the development of voip telephony. Essentially, the patent system is operating to block social change.
"blocking social change" needs to be appreciated. In it's positive manifestation it is the comfort and stability of the American System. There actually was a politicial platform called the "American System" described and advanced in the ~1880's. That system is a precursor to our current system.
After all, we all want a job to pay our mortgage and save up for our retirement and our kid's college tuition (partially, the loans begin in the Junior year). The stability of our American System is maintained by many interlocking mercantile compromises.
The problem has been patent and copyright lawyers have continuously blocked the writing of law to force reasonably priced licenses.
How about we require all patent an license fees to not exceed 5% of the dollar amount paid for the energy reequired to operate the patented device for a year. Only for registered, fee paid patents and copyrights for only 16 years from the date of issue of the patent.
Put the patent system and the vaunted value of copyrightable creative work in it's place: conceiving of something and building the first working prototype is just a tiny fraction of the lifetime costs of any invention or innovation. Lots and lots of independent re-invention happens. Let the lucky first inventor get a sliver of the follow on revenue if he is willing to get out there, charge consulting fees, and help out other people instead of sitting in a coffee shop with a lawyer scheming on how much money they can extract from other businesses.
For further thoughts, see Don Lancaster's web site and anti-patent comments. Janis Ian's website has a penetrating analysis of the music business. The big thinker of the 70's on this issue was Peter Drucker, a popularizer of knowledge based exports as a business advantage for highly educated technically advanced societies. (Planners in China and India read his books.). And of course there is Richard Stallman's free software. Essentially he has created a reasonably priced copyright license. He has opened a window into the next economic universe.
Thinking in terms of what social system will be economically superior to the American System, that next generation social framework will clearly surge ahead of us if they can design a clean, effective and cheap patent and copyright process.
The problems in management are often rubbery, mysterious misshapen things. The actions you take in response to these problems will reflect your unconscious and conscious assumptions about what your work is.
What Peter Drucker did during his life is crystalize some of these problems and find positions of strength and humanity that a manager can actually express at work.
Until another thinker appears, I would urge you to get every Drucker book you can from the library. Scan them and you will see the evolution of business thinking since WWII. Give at least one of the last of his books a good reading.
His studies of how to hire, how to fire and what counts in employee performance are still great. His recognition and exploration of the problem of performance in a non-profit organization is important.
I hope we will be fortunate enough to hear another management thinker of such scope.
What is the business outcome and advantage for Sun with this open hardware move?
It looks to me like Sun has figured out that they are in a knowledge business. They are operating on an information theory algorithm. They are creating a much larger pyramid of customers for their particular computer knowledge. There is a new enormous bottom layer of people contemplating using this fascinating powerful chunk of information. They are emitting information, not hardware.
The thing from information theory is: The more high quality information a source emits, the more valuable the source becomes. Sun still has the stable of PhD researchers from U.C. Berkeley and some more wags from Stanford. So the company will continue in the business of emitting information.
If you want the Macintosh of mainframes, they will sell them to you. If you want to boot Solaris or wire up Sparc chips you are still their customer. Sun will be your first publisher, web site and consultant.
It seems to me that this is a business innovation. It has been 60 years since Shannon's information theory paper suggested that the source that emits information increases in entropy. Sun is doing that by making available a uniquely sophisticated design - not hiding it in file cabinets in the basement.