Last fall, Quirks & Quarks interviewed another guy about this.
He advocated aeroponics, and for shallow rooted plants, floors that were only 2 feet apart.
I wrote to Bob McDonald, the show's host:
Bob, you muffed it, on the story about vertical farming. You didn't ask the right questions. Enough that there may be merit in revisiting the idea on another show. That you didn't ask them makes me regard Q&Q with a lot less credibility.
Light: Where does the light come from? Dense forest floor, with several layers of leaves between the ground and the sky is often bare. Lodgepole pine and spruce in a dense stand only have needles on the top 20 feet or so of tree. One of the reasons greenhouse operators have gone to steel tube frames is the decrease in obstructed light, particularly in spring for growing bedding plants. That few percent difference between a 1.25 inch steel tube, and a 2x4 wooden truss is enough to make a difference. (The steel is easier to clean, and doesn't rot, but the growth difference is measurable. Sunlight runs about 800 watts/square meter, at right angles to the light.
Go into a multi level parking lot with open sides. Go to the north side. How bright is it? Go to the middle. Usually the sides are only half open, so you will get twice as much light as you see. Still not much.
Energy: Plants are not efficient at turning light into to stored energy -- 2-3% if I remember my botany correctly. I question your guest's statement that using the waste matter can provide 3/4 of the energy needed. If the needed energy is in any form but heat, you have to convert it. Thermal electric conversions are at best only about 40% working with plant matter.
Cooling: In a greenhouse operation right now, one of the biggest problems is cooling it. Large greenhouses have to run an air change every 90 seconds on a warm day. Our houses typically are considered leaky at an air change every 1.5 hours.
Weight: I heard the word aeroponics one in the show, but even if the entire operation was aeroponic, that means two spaces per square foot of plants: A space above filled with light, and a space below that is both dark, and has an awful lot of plumbing. This isn't light weight. Soil systems and hydroponic systems both have considerable mass. Your guest spoke of 5 layers of wheat in a single storey. If the equivalent mass for a layer of wheat was only that of 4" of water, you have then 20 inches of water per storey, or about 100 lbs per square foot. This is a lot larger than the live load typically designed into buildings. While well within engineering abilities, it would seriously limit retrofitting. Think about your carpark again. A car weighs, what, 4000 lbs. A parking space is 20' long by 8 feet wide. 160 square feet. So a parkade has only a 50 lb/square foot load even where the cars are. And that is not even half the building.
Film based greenhouses can be created for about $10/square foot, with economies of scale coming in at larger sizes. It takes an enormously valuable crop to make it worth running one in winter. I get some of my seedlings from Boreal Horticulture near Lloydminister. A 1 season spruce is about 50 cents, but they grow them at densities of 100 per square foot -- about 2 million bucks an acre. A company near Red Deer grows tomatoes and peppers in greenhouses. Their plants basically shut down for two months per year. They don't die, but even with supplemental lighting they don't grow. And it's expensive to keep the greenhouse warm. (Remember they still have to change the air every 5-6 minutes to provide CO2 for the plants.
Given the very low cost of a greenhouse compared to a multistorey building, if this is practical at all, it will be practical at the greenhouse level. And no one grows wheat in a greenhouse.
Disease: Greenhouse production has terrible problems with pests of all sorts. Doing it right requires both great care, and constant monitoring. Imag
I used to work for a school. In July it temporarily suspended operation because they couldn't get either staff or students.
As part of the effort to keep the school open, I set up a google group, set it up initially with all of the staff and the board of directors as members, and started bombarding the list with ideas. After a month we had a couple dozen alumni, exstaff, and parents of boys on the list.
Over the next 6 months there were some 800 posts to the list. Almost none came from the staff or the board.
In December they announced that the school was closing permanently. I tell you, I could spit nails.
I have found generally that people don't want new ideas. And the newer they are, the less they want them.
I would love to find a company that genuinely wanted my input,
(If any of you work for companies that need a technical generalist, "Jack of all trades, master of some" please contact sgbotsford@gmail.com. )
I'm really great at finding usability issues in software, new ideas for interface, and finding new ways to crash the program.)
Centrally managed. Can I remotely log into an encrypted drive laptop while it's running? This is a common support setup. How about someone wearing headgear with low albedo?
And if there is an administrator who can decrypt every hard drive, what is his price? A hundred thousand dollars? A millon? The continued well being of his son?
And if this administrator leaves, how do I reset the master password on every machine?
And how many people know this master password?
Is the same master password used all the time, or is it a function of the destination and the date?
Can a laptop be anywhere and be 'recovered'?
Suppose I wear a hat of singularly low reflectivity: I steal Smith's laptop and his wallet. I now know his name, his company. Do some internet sleuthing and find more info about him. Look up some public records. Now I phone the admin desk, and say, "This is Mike Smith. I thought I knew my password but the laptop says it doesn't recognize it."
With reasonable skill at social engineering, what are the chances of getting it reset?
I'm not a crypt nerd, but what reading I've done suggests that having the same plaintext encrypted with two different keys out of the same system can give you a lot of information about both keys. And that if you KNOW the plaintext, then you have even more information about the keys.
If you encrypt the entire disk, do you also automatically move, shuffle, and frag the OS files? That should be good for a 20% slowdown.
Roll out 1000 laptops. You use some form of imaging tool to do this. foo.dll starts at the same location on each disk. Steal two laptops, now I have foo.dll encrypted with two keys from the same system. And I now have the plaintext and encrypted version of some text. Further, I can get copies of the (several) forms of foo.dll from MicroSloth.
And if they do move the files around a bit, starting off with a million starting locations (assuming cluster boundaries) is a much smaller start set than brute force, yes? (Does truecrypt and it's ilk pad the front of a file with random cruft to discourage this form of attack?
And aren't the first few sectors of a disk in a pretty rigid format?
It seems to me that to develop a secure system you want to minimize having known standard files encrypted multiple times with the same system.
Which in turn means if you are working with winsnooze, you have to keep it from scrawling your data all over C:\.
I can see two ways to accomplish this with current technology. Doubtless there are more using techniques I'm not aware of.
1. Have an OS and a Data partition. Only the data partition is encrypted. Beat on the OS so that it uses the data partition for temp files, for the user's registry hive.
2. Run the secure environment in something like VirtualBox. The OS is booted from an immutable disk. Changes don't survive a reboot of the VM. Add a file shredder function tot he host system to remove the temporary files. The data disk is also a virtual disk, that is encrypted by the guest operating system.
The existence of the temporary files while the VM is running is still a hazard, but not as great as the temp files and swap files in windows. With a unix host file recovery is somewhat iffier to start with. But since VB is semi-open source adding a feature to generate a random key and encrypt the temporary file should be fairly easy. This key does not have to be known to the user: It changes with every boot of the virtual machine. Now if you can write a windows program that automatically logs the user out when the laptop is closed, or the face disappears from in front of the camera...
"And god said, 'Let there be a big bang' and on the second femtosecond he rested."
If you state that "All living things on this planet descended from a single organism" as the fundamental statement of darwinism, then you are right, it is non-falsifiable. There are too many gaps in the record. If I can see 20% of the branches of a tree, however, I can get a good idea of 'treeness' I don't have to show that every twig connects to the trunk. It's sufficient to show that twigs can connect. And given in that 20% we can find some branch points, this isn't a big jump to make.
Science doesn't require all the bits. Only enough to show that the theory is plausible. It's not all easy to repeat experiments like chemistry. Look at geology, cosmology, archeology, meteorology, large areas of any of the social sciences. Many of these sciences are in principle falsifiable, but require too much instrumentation or other data to actually do it. So we play with models that mimic some of the features. As the models successfully mimic the features, we claim to having understanding of what is going on, at least at some level.
If we can make predictions then the theory is falsifiable even if it isn't complete. To the degree that the predictions are verified, the theory is supported or rejected.
It is usually regarded as sufficient to show that a theory is: * Self-consistent * Consistent with the known data. * Offers a coherent explanation for the known data. * Offers elements that are falsifiable.
Evolution offers this with far less ad-hoc explanation than any of the pseudo-science alternatives.
Consider the oak tree. Oak trees of a given species can crossbreed with the same species in their valley, and often with oak trees within a few miles, but often not with the same 'species' 20 miles away, although they appear identical.
The evolutionists claim this is a case of speciation. Acorns don't generally move far, so gene flow is low. So each valley is effectively isolated, so mutations accumulate. Eventually we may end up with a bunch of different oak species from this.
In this sense evolution is falsifiable. We can watch an experiment with oak trees right now. Of course the results won't be in for a hundred thousand years or so. Be patient.
Consider a study done with yeast. Through a mutation, a particular gene that created enzymes for the digestion of galactose (a sugar) and also regulated the production of this enzyme doupled. -- there were two copies of the gene. While the research scientist bred a LOT of generations of yeast in environments that both contained and did not contain galactose, one of the genes mutated to build a better galactase enzyme, while the other one mutated to become a better regulator. This in a simple way shows evolution in action.
My brother working with yeast found one that would take alcohol a couple percent higher than normal. In essence the alcohol produced by the sugar and yeast was acting as a selection agent. Because of a badly designed experiment he didn't transfer from one solution to the next on a fast enough basis, so selected for alcohol tolerant yeast.
Numerous models with computers using plausible starting conditions show evolutionary behaviour. And many automina show behaviour that is far more complex than expected. See the complexity possible with Conway's "Life" Two states, three rules, two dimensions, utter determinism, yet it has a very rich 'universe' Why should we think that with 100+ atoms, three dimensions, quantum mechanics, and billions of years that we would get simplicity without a Designer's intervention.
Some of this depends on scale. Most of my career I've been the entire IT department, working in places where I had under 10 servers and under 300 machines.
I like whimsical names. I ended up having a bunch of naming schemes. The SGI irix lab had machines named after birds. All the secretaries windows machines were named after ancient goddesses. Servers were named after first magnitude stars. X-stations after fifth magnitude stars.
You are limited as to how much info you can code into a name. In my mind the information lost by using whimsical names is more than made up by having lower error rates typing.
Since these were small shops, most machines had to serve multiple functions. I used the CNAME facility in DNS a lot.
So Vega, the largest server we had, in addition to being the primary file server was also web server, mail server, and dns server. But they were cnamed as sambaprime, nfsprime, www, smtp, pop, and dns respectively.
As we grew large enough to need a separate mail server the new box was made ready, and the the cnames changed.
The big problem with functional names: 1. If a machine has multiple functions, you have to untangle the functions when it outgrows the original list. If a machine is repurposed, you have to either change its name, or risk getting confused. changing it's name makes hardware bookkeeping more interesting. 2. In many cases with lots of machines, you end up with alphabet soup for names. The chances of making a typo when you have a name you can't pronounce, or is 20 characters long is much higher than somYou are limited as to how much info you can code into a name. In my mind the information lost by using whimsical names is more than made up by having lower error rates typing.ething that is short and pronounceable. 3. In a small organization you can associate names with locations, functions, and so on. In a large organization you can eitehr give them functional names (web1,web2) or give them class names, and know that the class associates with a function. (E.g. name all your web servers after spiders)
In any case once you get to thousands of names you aren't going to remember all the associations. (Is Webserver1 the box that runs the shopping cart code) This is what databases are for, and, if you wish what DNS hinfo records are for.
Imagine a screen like a crazy carpet. (Kids snow sled made from 1/16" polyethylene, 2' x 5') Stand it on edge on a desk in an circular arc.
Imagine a screen for trade shows that can be set up like a projection screen -- and you never have to worry about casting a shadow on the screen or blinding yourself glancing at the projector.
The lower res version becomes a large screen TV without the $200 wall hanging hardware, and disappears into a valance at the ceiling.
Imagine high res screens with very low bandwidth that allow you to subscribe to the poster of the week club. (s/poster/artwork/)
Imagine the possibility of having animated logos on flags.
Imagine a laptop computer the size of whatever keyboard your fingers are in love with. The screen rolls up in to the upper edge of the keyboard, with the top inch of the screen showing above the keyboard when it's furled, so you can use it for small screen applications without unfurling.
Imagine using them as window coverings.
If you have a window that looks onto a factory roof or an air shaft, you can put a 'scenedow' that carries a live feed from a mountain meadow, a rocky beach.
If they work passively (they don't emit light, but just change reflectivity) then they can be used to change the albedo of your windows. Silver in summer to reflect the heat, black in winter to absorb it.
If they are really flexible, (E.g. like cloth) there are interesting art possibilities having animated imagery or abstract art on draped screens. Flags with animated images. Or clothing that truly makes a statement. (Imagine running X11 on your shirt. Now imagine Xroaches...
I recently spent two DAYS looking for software to manage information for a startup school. There are several packages out there, some free, that are targeted at non-profits, but most are limited to communication and fund raising.
The closest fit is a package called to Metrix (city of new york) but that will operate with a non-windows back end. (Who wants to expose windows to the internet...)
Partial list of requirements.
Multi-user. Record level locking. MS Access or similarly flexible front end. (Ideally an OSS front end, but so far I've not seen anything that allows such quick layout of forms and reports as Access. I'm open, no begging for correction on this issue. But it can run access inside windows inside virtualbox for linux users.
Full contact info. Preferred contact mode. Contact history. Subscription preferences. (Quarterly Printed newsletter, monthly email newsletter, monthly summary with links to web page, visions discussion list...) Skills both checkable and free form.
Pledges -- monetary. Pledges -- service -- list of tasks and roles that people are willing to do. E.g. "Will bring truck and trailer for bottle drives. Saturday only." Pledge history. -- do they come through. Link to notes.
Web interface for users to edit their own relevant information -- e.g. withdraw a pledge, change their contact info.
White Notes field. "Good resource for plumbing parts" Gray notes field "Triple the amount of time he says it will take." Black notes field. "Do not trust with money"
Notes fields have settable permissions. E.g. White notes are visible to all database users, gray notes only to project managers who have that person on their project, black notes only to administrators.
Transaction history on certain records. E.g. Keep old records but mark them as superceded, and who changed them. Keep view history of gray and black notes.
Definable roles: * Administrator -- can do anything * Program manager -- E.g. Bake Sale Coordinator, Outdoor Program manager. Can work with records involving their program , may have editing privleges in contact info * PR manager -- has read access to nearly everything. * Data entry person. Can be given a temporary role to enter data for a project.
The system should also be able to handle contacts for non-volunteers -- The local building inspectors both friently and un. The local school board people. The person at the provincial text-book bureau. The reporter that has the education beat at the metro paper. Editor at every small town paper in 100 miles. Our member of the legislative assembly. The education minister. The county snow clearing number. Local service people we deal with on non-volunteer basis. (It's nice to have a volunteer plumber, but it's two in the morning, and you have a broken pipe in the basement, you don't always have time.)
This is off the cuff. It will almost certainly change.
Or maybe they want the protocol done in a way that NSA CAN subvert any router detouring it's packets through their own computers, sniffing and injecting (cocaine & herion?) to their hearts content.
Just because I'm paranoid doesn't mean they aren't out to get you.
(He says, from his satellite connected hide-away in rural Alberta, 500 km from the nearest chunk of American soil)
To answer your question specifically: I have a 2 core linux box with 8 G ram. But Mac+iTUnes is the best way to work with my iPod.
I run a copy of WinXP in VirtualBox just so that I can run two applications: Whizfolders -- an outliner tool that has no real equivalent in unix -- and IMatch a photo management tool that has features that I can't find in a unix tool.
Anyone who is writing software for multiple OS's needs an easy way to test that code on those same multiple OS's
We will assume the dirtiest possible generation of electicity: Coal.
Coal runs about 6150 kW-hr thermal per ton. = 6785/tonne (metric ton) = 6.7 kW-hr per kilogram.
We'll assume 40% plant efficiency. So now we are down to 2.8.
7 grams of carbon dioxide is produced from 1.9 grams of carbon.
1.9 grams of carbon = 12.8 watt hours.
This seems to me to be an unreasonably large amount of energy for a single search.
If we take the search I just ran for the kilowatt hours per ton of coal as typical, it took.3 seconds. At this rate searching requires a power level of 150,000 watts.
There are all kinds of arguments that this time shouldn't be used: Much of that delay is waiting for spinning disks. And who knows how many simultaneous searches are done by that one server.
You can also make the distinction between the differential cost of searching versus the aggregate cost. The first is the carbon cost of an additional search. You could calculate this by dividing the carbon cost of putting a new search server on line and running it by the nubmer of searches per hour it could handle. The second is done by taking Google's total power consumption and dividing it by the total number of searches.
When you compress air it gets hot. A lot of the work you do goes to heat. As it cools, the pressure drops. The net effect is a huge loss of efficiency.
If you can pressurize it and keep it from cooling off, you can get more of the energy back.
Anybody got the figures for compressor efficiency?
Yes. It all works smoothly at 20 to 40 miles per hour. Most of the track isn't flat enough to run at 70 mph.
I think the width of the track dates back to the habit of wagon makers in England. Early rail had no particular need for a size, so they made them the same width as wagon makers. Meant the wagon makers didn't have to retool as much to make rail cars.
Wider tracks make a train more stable, but make it harder to make corners. Currently train wheels are welded to the axle. Both wheels turn at the same speed. The wheel itself is somewhat cone shaped. As you go around the corner, the axle moves to the outside of the corner, so the outside wheel is riding on the fattest part of the cone, while the inside wheel is riding on the thinnest. If the wheel doesn't move the right distance, there is slippage between the wheel and the rail, resulting in a screech.
But this means there is an optimum speed for each corner, depending on radius of the corner, how heavily loaded the train is, and probably six other things previously only thought to affect the flavour of pizza.
Most of the mainline track in Canada is twinned. With twinned track you can actually have a fast train pass a slow train, but it requires tricky scheduling, and long spaces between the slow trains. If there is enough demand for freight to keep the slow train track slots filled, then running fast trains gets very hard.
Branch lines are largely single track. Which means that one train has to get off on a siding for a train going the other way. or to pass. There is a lot of momentum is a 100 car freight train. They are not nimble, not sports cars. Branch lines are designed to have a few trains per day. I suspect that many branch lines seldom have more than a single train on them at any given time.
Steel on steel is by far the least energy intensive way to move stuff on land -- but you need to have the tracks. Rail transport is expensive to set up and requires heavy use to be economically viable.
I live in Canada: The population of California spread out mostly in a strip 100 miles wide by 4000 miles long. Here, even as a freight transport rail is having trouble. Oh,the mainlines are still functional, but more and more of the branch lines are being cut off. Farmers used to truck their grain to the nearest elevator, which was the nearest town, usually under 12 miles away. Now they are having to drive many times further.
We still have passenger service on our mainline rails, but it costs 2-3 times as much as flying, and is promoted as a tourist thing. Freight traffic generally moves much more slowly than passenger traffic. On dual track lines, you can set it up for fast trains to pass slow trains, but it makes the scheduling much trickier. During the height of the grain shipping season, the CN mainline has a train roughly every 20 minutes. Typically 100-120 cars long. At 40 mph these trains are 15 miles apart. Since one of the rules the conservation of wheels, for every train going one way, there is one going the other way. Weaving an 80 mph passenger train between the freight trains is tricky.
Light rail is making some progress in cities. Edmonton started a network 30 years ago. There may be 20 miles of track now.
The Toronto metro area has pretty good commuter trains running parallel to the lake. If your destination is close to the rails, you can cut your commute time in 1/2. But Toronto metro has over 10% of our population.
The other extreme is the terratories north of latitude 60. An area sized on the order of the continental U.S. (depending on how you count water spaces between islands...) that has under 100,000 people. Of that 100,000 80% are in/around 5 towns. An awful lot of Canada has population densities that make Montana and Wyoming look crowded.
The OP asked how to monitor. Most of the above is on prevention.
Encryption is part of it. Part too can be some form of chain of custody. Each document has some form of who did what. Version control with change logs.
This way you know who had the document.
But consider that Office apps have a bunch of hidden data in them. So mark each document with multiple flags so that each copy checked out is different. This can be done either in the hidden part of the document, or by setting subtle style flags, (Right margin on page 3 is 1.495" instead of 1.500 inches Font for Heading 2 is 15.95 points, not 16 ) You have to have enough flags so that a bunch can be changed and there are still enough to uniquely identify the version.
This way at least, if your customer can get the 'revealed' document, you have a chance of finding out what part of your organization is insecure.
Another possibility is to only allow editing of documents via a virtual machine located in the server room. There is no copy on the local machine. Connections to the virtual box are through the company VPN. If the document is printed locally it is imprinted with the version, who it was printed for, and where it was printed.
Indesign is fabulous for layout. It's not appropriate for composing.
Part of this depends on how you work. One of the best programs I've found is a winsnooze program called WhizFolders. This is really an outline processor. You build a tree of ideas/points, then write a chunk about each idea as the leaf of the tree. Re-arrange the tree as you wish.
If you are semi-scatterbrained, like me, this has a lot of appeal, as you just slot new ideas as they occur to the more-0r-less right place, then carry on with what you were doing. Of all the outline/composing programs I've used, this one does the least to get in my way when doing something that is longer than 10 pages.
Alas it's not free, but it doesn't cost a lot.
It does not do formatting beyond something very primitive, but you can export a document as RTF then take it into FrameMaker, InDesign or (shudder) MS Snerd, and do the final tweaks there.
I'm not a programmer -- ok not quite true. I'm a sysadmin, so I spent a lot of time writing perl, but most of my life was fixing hardware and configuring things to work together. In most of my jobs I was the entire IT department: network, servers, clients, software, hardware, specs, purchase orders which meant that I learned do a lot of different jobs badly.
For 15 years I worked at a university. And yes, I put in some long days. But the work week was nominally 35 hours, not forty; the U gave us all a week without pay between Christmas and New years. (At our option it was averaged into our monthly cheques) We started with 3 weeks holiday per year. And my bosses would let me record my hours, and take time off in lieu.
Later I took a job with a hardware development company, with about a 25% raise. Biggest mistake I ever made. 50-60 hours was expected of you, initially for 'the last push' to get a product ready to go out the door, but these pushes breed like rabbits.
The company was fun, small and exciting the first year. Within a year it had become totally dilbertized, cubicals and all. I saw the writing on the wall when it got to the point that I needed three signatures to buy a $110 dollar ethernet card, one being at head office in Seattle.
For the last 5 years I've been working half time for a local school at 40% of the pay I was getting in industry. However as part of that job: * 5 weeks of the year was canoeing, hiking, snowshoeing and orienteering, which I love. * I could bank hours in bad weather and take it off in good weather. * My commute reduced from 75 minutes each way to 12 minutes each way. In good weather it was reasonable to bike to work.
This gave me time to work on my tree farm.
The school has closed. So I'm working on the tree farm full time, and doing house renos that I've postponed for years.
Whenever I've flown over the landscape it seems to me that forests are much darker than either grasslands or most crops.
If you plant forest, it has a one time 50-100 year impact soaking up CO2 before the biomass reaches an equilibrium between growth and decay.
The only biosystem I'm aware of that sequesters carbon indefinitely are peat bogs. The little beasties create enough acidity that the previous generations don't rot. Of course they ahve to be in a situation where there continues to be standing water as the layers get thicker.
This is one of the reasons that climate warming affects arctic regions to a greater degree than tropical regions. Warm up the arctic, and the treeline moves north. Spruce (predominate tree line species) has a very low albedo. This in turn means that the net reflectivity of the region drops, which further warms the region. Positive feed back loop.
A better plan generally is to harvest biomass for energy, but instead of burning it all, use gassification systems,to make much of it into charcoal. Return the charcoal to the soil. Pure carbon has a very long residence time, and acts as a surface for nutrients to cling to. Some experiments with doing this in tropical rain forest slash and burn agriculture have found that the soil become quite productive for long periods.
What do you want!
GLOBAL WARMING When do you want it!
NOW
What do you want!
GLOBAL WARMING When do you want it!
NOW
As it write it's a day or two to the official start of winter, and the temperature on the front porch is -34 C. This is some 20 degrees C (36 F for the metricly disadvantaged) below normals for this time of year.
Let's take a look at this. To be viable you would have to evaporate an amount of water that is significant compared to the water that is already evaporated. Given that a largish thunderstorm cell has energy content on the order of an H bomb, methinks that the budget is going to be a bit stretched.
For houses this works well. For planets, no.
Also evaporating water may drop the temperature, but it doesn't drop the heat. You're just storing it as latent heat of water vapor. In that form it can still play with the rest of the climate engine.
Even if it were possible, there are good reasons to think that it MIGHT be counter productive.
1. Water is a potent greenhouse gas, blocking different IR bands than CO2. Increasing the average water content of the atmosphere may well cause the average temp to go up.
2. Enough water vapour and you increase the number of clouds. Increasing the number of clouds can push the climate either way: Thick clouds tend to cool the earth, thin clouds tend to warm it. Last time I checked (some years ago) cloud modeling was one of the sticky points in climate models.
3. If you used fresh water for this, it's going to put a major drain on our fresh water supplies. If you use salt water for this, you will put a huge amount of cute microscopic salt crystals in the air. These act as condensation nuclei for water droplets. The formation of rain is dependent on the number of nuclei. Too few and you get a few large drops of rain that result in light rainfall. Too many and you get masses of cloud with drops too small to fall, or that evaporate on the way down.
A possibly more viable form of climate control would be to use H bombs to turn mountain tops into stratospheric dust. We have significant data that large volcanic eruptions can cool the atmosphere for a few years. I don't know if anyone has figured out how much of this is due to dust, and how much to sulfates. Sure this method increases the background radiation. But I think most people would take a 1-2% increase of cancer in 20 years rather than become a refugee of rising ocean levels.
(Caveat: I've not done the math. How many bombs a year does it take to do a Krakatoa? What is the radiation release of an H bomb designed to pulverize the maximum amount of rock and inject it into the stratosphere? I submit that the math for this is less than trivial.)
As some one who managed a school's network, I've found that the kids are moderately clever at defeating what you do. You can make the off limit stuff harder or slower, but not impossible unless you lock the computer down to the point that it is impossible to use for much of anything.
A good part of this is education, both of kids and of parents. We had a set of rules -- a code of conduct or appropriate use, and fairly limited filtering. E.g. I 'enforced' a no porn rule. My definition was simple: If you are uncomfortable with the thought of your sister or mom (boys' school) then it's porn. Monitors in the lab faced hallway windows. First offense, I took the kid aside and explained why I considered inappropriate. Second offence he had no access until he wrote 200 times (by hand) "I will not use school computers to surf porn." At this point they either get clever enough to not get caught, or they get alternate sources for their pix.
Similarly part of the intro process is appropriate use of social networking sites. We talk about the risks, the damage that group gossip can do, the possibility of predators.
Your situation will be different. You're somewhat younger than my group. (I had 7-12) You have full time access by the kids. Your filtering will be more limited.
You need to educate the parents as well as the kids. Explain to parents what the issues are, and suggest to the parents that depending on their kid, they may want home rules such as, "You use the laptop at the kitchen table when your mom or I are in the room." Or "Please leave your bedroom door open and the screen facing the door."
You should have a system of sanctions for inappropriate use. These can range from "Your laptop needs to stay after school" to blocking internet access outside of school to forced rebuild of system and applications folders.
OpenDNS is your friend. It makes it easy to block the worst of the really raunchy sites.
As a second thread: Your ownership period is way too long. The lifespan of a laptop is short -- typically only a couple years. Then too, consider that even a state of the art laptop given to a 6th grader will, 6 years later, be totally lame.
I suggest that laptops stay in active use for 2 years, then you replace them all. Or if budget is a big constraint, then 3 years. If really a constraint, then 3 years in senior high, then move them to junior high so the kids have a bright and shiny laptop to look forward to making the transition from grade 9 to grade 10.
On lists I've been a member of that discussed laptop per child ownerership the consensus seemed to be that you needed 10% over for spares, and that had about 5% mortality (uneconomic to repair)
I know of one school that had a two year cycle. They would lease the laptops for 2 years, and replace them all simultaneously. This made imaging easier, as they all had identical hardware. Now that was winsnooze, which is a lot trickier in terms of maintaining multiple images.
Yes, II have a fair selection of free/cheap apps on my phone. But the latest one, Omnifocus, at $20 is worth every penny of it's price. I've bought the Mac version ($80) for it too and they sync flawlessly.
For any of you who are iPhone/iPod touch developers:
I would pay $100 for a program that allows me to replace my pencil and clipboard with an ipod touch for taking inventory at my tree farm. Contact me for specs if interested.
Some of this depends on what the previous experience of these gifted kids (sorry, I like the term...) is. I found in teaching that a lot of kids in grade 8 and 9 had a real struggle with the abstractions of algebra, and a year later had no trouble with it.
On the flip side, I see kids in my computer lab during their free time playing with exceedingly lame java script games, the best of which (while still lame) allowing construction of courses for their 2 dimensional glyph motorcycles.
I think there is a lot to be said for logo. You get a connection between text and graphical elements, a connection between abstract and "real"
As to senseless shapes: That can change very quickly. I saw one logo into class where the teacher spent most of the first hour teaching kids how to type the program in, and getting across to them that "almost right" wasn't good enough for the computer to understand you. By the end of the class each kid could draw a box. Hello World!
By the end of the second period he'd introduced procedures and they were drawing mullioned double hung sash windows.
The end of the next period saw them drawing a house, then the first assignment -- to draw their own house.
I was fascinated just watching this class and stayed after school to play myself.
With a combined graphic/command world you could implement all sorts of neat things that would keep kids fascinated:
Consider: By the time you are in geosync orbit, you are moving sideways at 3 km/s (orbital velocity) The carrier only lifts you up. That sideways velocity has come from the tether.
This is why most space elevator proposals I've seen attach a small asteroid to the upper end of the tether. A carrier arrives from down below. The asteroid slows down a bit, and drops behind the anchor point on the earth. Now it is no longer directly above the anchor point. It acts like a pendulum, and swings back.
When I'm sending more mass out than is coming back the net effect is to slow down the earth's spin.
It's gotten significant attention by enough peer reviewed journals that include a lot of people who have education in physics.
The usual conclusions are that it's a tough engineering problem.
The NS article is sparse on what assumptions they made, but the illustration shows the carrier as being large compared to the tether. What I've read is the other way around. The tether is massive compared to the weight of the carriers.
The NS article goes on to say that they may have to move slowly. I don't see how this helps. As postulated the tether has no real damping mechanism. If I go up slowly or fast force times time will be the same product for the trip.
The key is going to be the ratio of the tension to the mass of the cable. Suppose that the tether has a natural resonate period of 2 days. Traveling up it in 2 days will 'pluck' it with maximal force. The tether will continue to vibrate for a long time.
Now send a second carrier up 1 day out of phase. To first approximation, it will cancel out the vibration induced by the first carrier.
Consider the situation when there is a carrier every kilometer both ways on the tether. At this point the coriolis forces are balanced. (Or not quite: They are balanced when the carriers are passing each other.
There are going to be higher frequencies on the line. Doubtless it will take some very clever engineering. But it's an engineer problem, not a physics problem.
Do remember Clarke's first law: When an (elderly) scientist says something is impossible, he is almost certainly wrong.
Slashdot Mirror
Seems to be a fad for this idea.
Last fall, Quirks & Quarks interviewed another guy about this.
He advocated aeroponics, and for shallow rooted plants, floors that were only 2 feet apart.
I wrote to Bob McDonald, the show's host:
Bob, you muffed it, on the story about vertical farming. You didn't ask the right questions. Enough that there may be merit in revisiting the idea on another show. That you didn't ask them makes me regard Q&Q with a lot less credibility.
Light: Where does the light come from? Dense forest floor, with several layers of leaves between the ground and the sky is often bare. Lodgepole pine and spruce in a dense stand only have needles on the top 20 feet or so of tree. One of the reasons greenhouse operators have gone to steel tube frames is the decrease in obstructed light, particularly in spring for growing bedding plants. That few percent difference between a 1.25 inch steel tube, and a 2x4 wooden truss is enough to make a difference. (The steel is easier to clean, and doesn't rot, but the growth difference is measurable. Sunlight runs about 800 watts/square meter, at right angles to the light.
Go into a multi level parking lot with open sides. Go to the north side. How bright is it? Go to the middle. Usually the sides are only half open, so you will get twice as much light as you see. Still not much.
Energy: Plants are not efficient at turning light into to stored energy -- 2-3% if I remember my botany correctly. I question your guest's statement that using the waste matter can provide 3/4 of the energy needed. If the needed energy is in any form but heat, you have to convert it. Thermal electric conversions are at best only about 40% working with plant matter.
Cooling: In a greenhouse operation right now, one of the biggest problems is cooling it. Large greenhouses have to run an air change every 90 seconds on a warm day. Our houses typically are considered leaky at an air change every 1.5 hours.
Weight: I heard the word aeroponics one in the show, but even if the entire operation was aeroponic, that means two spaces per square foot of plants: A space above filled with light, and a space below that is both dark, and has an awful lot of plumbing. This isn't light weight. Soil systems and hydroponic systems both have considerable mass. Your guest spoke of 5 layers of wheat in a single storey. If the equivalent mass for a layer of wheat was only that of 4" of water, you have then 20 inches of water per storey, or about 100 lbs per square foot. This is a lot larger than the live load typically designed into buildings. While well within engineering abilities, it would seriously limit retrofitting. Think about your carpark again. A car weighs, what, 4000 lbs. A parking space is 20' long by 8 feet wide. 160 square feet. So a parkade has only a 50 lb/square foot load even where the cars are. And that is not even half the building.
Film based greenhouses can be created for about $10/square foot, with economies of scale coming in at larger sizes. It takes an enormously valuable crop to make it worth running one in winter. I get some of my seedlings from Boreal Horticulture near Lloydminister. A 1 season spruce is about 50 cents, but they grow them at densities of 100 per square foot -- about 2 million bucks an acre. A company near Red Deer grows tomatoes and peppers in greenhouses. Their plants basically shut down for two months per year. They don't die, but even with supplemental lighting they don't grow. And it's expensive to keep the greenhouse warm. (Remember they still have to change the air every 5-6 minutes to provide CO2 for the plants.
Given the very low cost of a greenhouse compared to a multistorey building, if this is practical at all, it will be practical at the greenhouse level. And no one grows wheat in a greenhouse.
Disease: Greenhouse production has terrible problems with pests of all sorts. Doing it right requires both great care, and constant monitoring. Imag
Hear hear.
I used to work for a school. In July it temporarily suspended operation because they couldn't get either staff or students.
As part of the effort to keep the school open, I set up a google group, set it up initially with all of the staff and the board of directors as members, and started bombarding the list with ideas. After a month we had a couple dozen alumni, exstaff, and parents of boys on the list.
Over the next 6 months there were some 800 posts to the list. Almost none came from the staff or the board.
In December they announced that the school was closing permanently. I tell you, I could spit nails.
I have found generally that people don't want new ideas. And the newer they are, the less they want them.
I would love to find a company that genuinely wanted my input,
(If any of you work for companies that need a technical generalist, "Jack of all trades, master of some" please contact sgbotsford@gmail.com. )
I'm really great at finding usability issues in software, new ideas for interface, and finding new ways to crash the program.)
Centrally managed. Can I remotely log into an encrypted drive laptop while it's running? This is a common support setup. How about someone wearing headgear with low albedo?
And if there is an administrator who can decrypt every hard drive, what is his price? A hundred thousand dollars? A millon? The continued well being of his son?
And if this administrator leaves, how do I reset the master password on every machine?
And how many people know this master password?
Is the same master password used all the time, or is it a function of the destination and the date?
Can a laptop be anywhere and be 'recovered'?
Suppose I wear a hat of singularly low reflectivity: I steal Smith's laptop and his wallet. I now know his name, his company. Do some internet sleuthing and find more info about him. Look up some public records. Now I phone the admin desk, and say, "This is Mike Smith. I thought I knew my password but the laptop says it doesn't recognize it."
With reasonable skill at social engineering, what are the chances of getting it reset?
I'm not a crypt nerd, but what reading I've done suggests that having the same plaintext encrypted with two different keys out of the same system can give you a lot of information about both keys. And that if you KNOW the plaintext, then you have even more information about the keys.
If you encrypt the entire disk, do you also automatically move, shuffle, and frag the OS files? That should be good for a 20% slowdown.
Roll out 1000 laptops. You use some form of imaging tool to do this. foo.dll starts at the same location on each disk.
Steal two laptops, now I have foo.dll encrypted with two keys from the same system. And I now have the plaintext and encrypted version of some text. Further, I can get copies of the (several) forms of foo.dll from MicroSloth.
And if they do move the files around a bit, starting off with a million starting locations (assuming cluster boundaries) is a much smaller start set than brute force, yes? (Does truecrypt and it's ilk pad the front of a file with random cruft to discourage this form of attack?
And aren't the first few sectors of a disk in a pretty rigid format?
It seems to me that to develop a secure system you want to minimize having known standard files encrypted multiple times with the same system.
Which in turn means if you are working with winsnooze, you have to keep it from scrawling your data all over C:\.
I can see two ways to accomplish this with current technology. Doubtless there are more using techniques I'm not aware of.
1. Have an OS and a Data partition. Only the data partition is encrypted. Beat on the OS so that it uses the data partition for temp files, for the user's registry hive.
2. Run the secure environment in something like VirtualBox. The OS is booted from an immutable disk. Changes don't survive a reboot of the VM. Add a file shredder function tot he host system to remove the temporary files. The data disk is also a virtual disk, that is encrypted by the guest operating system.
The existence of the temporary files while the VM is running is still a hazard, but not as great as the temp files and swap files in windows. With a unix host file recovery is somewhat iffier to start with. But since VB is semi-open source adding a feature to generate a random key and encrypt the temporary file should be fairly easy. This key does not have to be known to the user: It changes with every boot of the virtual machine. Now if you can write a windows program that automatically logs the user out when the laptop is closed, or the face disappears from in front of the camera...
Whew. I think someone has an axe to grind.
"And god said, 'Let there be a big bang' and on the second femtosecond he rested."
If you state that "All living things on this planet descended from a single organism" as the fundamental statement of darwinism, then you are right, it is non-falsifiable. There are too many gaps in the record. If I can see 20% of the branches of a tree, however, I can get a good idea of 'treeness' I don't have to show that every twig connects to the trunk. It's sufficient to show that twigs can connect. And given in that 20% we can find some branch points, this isn't a big jump to make.
Science doesn't require all the bits. Only enough to show that the theory is plausible. It's not all easy to repeat experiments like chemistry. Look at geology, cosmology, archeology, meteorology, large areas of any of the social sciences. Many of these sciences are in principle falsifiable, but require too much instrumentation or other data to actually do it. So we play with models that mimic some of the features. As the models successfully mimic the features, we claim to having understanding of what is going on, at least at some level.
If we can make predictions then the theory is falsifiable even if it isn't complete. To the degree that the predictions are verified, the theory is supported or rejected.
It is usually regarded as sufficient to show that a theory is:
* Self-consistent
* Consistent with the known data.
* Offers a coherent explanation for the known data.
* Offers elements that are falsifiable.
Evolution offers this with far less ad-hoc explanation than any of the pseudo-science alternatives.
Consider the oak tree. Oak trees of a given species can crossbreed with the same species in their valley, and often with oak trees within a few miles, but often not with the same 'species' 20 miles away, although they appear identical.
The evolutionists claim this is a case of speciation. Acorns don't generally move far, so gene flow is low. So each valley is effectively isolated, so mutations accumulate. Eventually we may end up with a bunch of different oak species from this.
In this sense evolution is falsifiable. We can watch an experiment with oak trees right now. Of course the results won't be in for a hundred thousand years or so. Be patient.
Consider a study done with yeast. Through a mutation, a particular gene that created enzymes for the digestion of galactose (a sugar) and also regulated the production of this enzyme doupled. -- there were two copies of the gene. While the research scientist bred a LOT of generations of yeast in environments that both contained and did not contain galactose, one of the genes mutated to build a better galactase enzyme, while the other one mutated to become a better regulator. This in a simple way shows evolution in action.
My brother working with yeast found one that would take alcohol a couple percent higher than normal. In essence the alcohol produced by the sugar and yeast was acting as a selection agent. Because of a badly designed experiment he didn't transfer from one solution to the next on a fast enough basis, so selected for alcohol tolerant yeast.
Numerous models with computers using plausible starting conditions show evolutionary behaviour. And many automina show behaviour that is far more complex than expected. See the complexity possible with Conway's "Life" Two states, three rules, two dimensions, utter determinism, yet it has a very rich 'universe' Why should we think that with 100+ atoms, three dimensions, quantum mechanics, and billions of years that we would get simplicity without a Designer's intervention.
Some of this depends on scale. Most of my career I've been the entire IT department, working in places where I had under 10 servers and under 300 machines.
I like whimsical names. I ended up having a bunch of naming schemes.
The SGI irix lab had machines named after birds.
All the secretaries windows machines were named after ancient goddesses.
Servers were named after first magnitude stars. X-stations after fifth magnitude stars.
You are limited as to how much info you can code into a name. In my mind the information lost by using whimsical names is more than made up by having lower error rates typing.
Since these were small shops, most machines had to serve multiple functions. I used the CNAME facility in DNS a lot.
So Vega, the largest server we had, in addition to being the primary file server was also web server, mail server, and dns server. But they were cnamed as sambaprime, nfsprime, www, smtp, pop, and dns respectively.
As we grew large enough to need a separate mail server the new box was made ready, and the the cnames changed.
The big problem with functional names:
1. If a machine has multiple functions, you have to untangle the functions when it outgrows the original list. If a machine is repurposed, you have to either change its name, or risk getting confused. changing it's name makes hardware bookkeeping more interesting.
2. In many cases with lots of machines, you end up with alphabet soup for names. The chances of making a typo when you have a name you can't pronounce, or is 20 characters long is much higher than somYou are limited as to how much info you can code into a name. In my mind the information lost by using whimsical names is more than made up by having lower error rates typing.ething that is short and pronounceable.
3. In a small organization you can associate names with locations, functions, and so on. In a large organization you can eitehr give them functional names (web1,web2) or give them class names, and know that the class associates with a function. (E.g. name all your web servers after spiders)
In any case once you get to thousands of names you aren't going to remember all the associations. (Is Webserver1 the box that runs the shopping cart code) This is what databases are for, and, if you wish what DNS hinfo records are for.
Imagine a screen like a crazy carpet. (Kids snow sled made from 1/16" polyethylene, 2' x 5') Stand it on edge on a desk in an circular arc.
Imagine a screen for trade shows that can be set up like a projection screen -- and you never have to worry about casting a shadow on the screen or blinding yourself glancing at the projector.
The lower res version becomes a large screen TV without the $200 wall hanging hardware, and disappears into a valance at the ceiling.
Imagine high res screens with very low bandwidth that allow you to subscribe to the poster of the week club. (s/poster/artwork/)
Imagine the possibility of having animated logos on flags.
Imagine a laptop computer the size of whatever keyboard your fingers are in love with. The screen rolls up in to the upper edge of the keyboard, with the top inch of the screen showing above the keyboard when it's furled, so you can use it for small screen applications without unfurling.
Imagine using them as window coverings.
If you have a window that looks onto a factory roof or an air shaft, you can put a 'scenedow' that carries a live feed from a mountain meadow, a rocky beach.
If they work passively (they don't emit light, but just change reflectivity) then they can be used to change the albedo of your windows. Silver in summer to reflect the heat, black in winter to absorb it.
If they are really flexible, (E.g. like cloth) there are interesting art possibilities having animated imagery or abstract art on draped screens. Flags with animated images. Or clothing that truly makes a statement. (Imagine running X11 on your shirt. Now imagine Xroaches...
I recently spent two DAYS looking for software to manage information for a startup school. There are several packages out there, some free, that are targeted at non-profits, but most are limited to communication and fund raising.
The closest fit is a package called to Metrix (city of new york) but that will operate with a non-windows back end. (Who wants to expose windows to the internet...)
Partial list of requirements.
Multi-user.
Record level locking.
MS Access or similarly flexible front end. (Ideally an OSS front end, but so far I've not seen anything that allows such quick layout of forms and reports as Access. I'm open, no begging for correction on this issue. But it can run access inside windows inside virtualbox for linux users.
Full contact info.
Preferred contact mode.
Contact history.
Subscription preferences. (Quarterly Printed newsletter, monthly email newsletter, monthly summary with links to web page, visions discussion list...)
Skills both checkable and free form.
Pledges -- monetary.
Pledges -- service -- list of tasks and roles that people are willing to do. E.g. "Will bring truck and trailer for bottle drives. Saturday only."
Pledge history. -- do they come through. Link to notes.
Web interface for users to edit their own relevant information -- e.g. withdraw a pledge, change their contact info.
White Notes field. "Good resource for plumbing parts"
Gray notes field "Triple the amount of time he says it will take."
Black notes field. "Do not trust with money"
Notes fields have settable permissions. E.g. White notes are visible to all database users, gray notes only to project managers who have that person on their project, black notes only to administrators.
Transaction history on certain records. E.g. Keep old records but mark them as superceded, and who changed them.
Keep view history of gray and black notes.
Definable roles:
* Administrator -- can do anything
* Program manager -- E.g. Bake Sale Coordinator, Outdoor
Program manager. Can work with records involving their program , may have editing privleges in contact info
* PR manager -- has read access to nearly everything.
* Data entry person. Can be given a temporary role to enter data for a project.
The system should also be able to handle contacts for non-volunteers -- The local building inspectors both friently and un. The local school board people. The person at the provincial text-book bureau. The reporter that has the education beat at the metro paper. Editor at every small town paper in 100 miles. Our member of the legislative assembly. The education minister. The county snow clearing number. Local service people we deal with on non-volunteer basis. (It's nice to have a volunteer plumber, but it's two in the morning, and you have a broken pipe in the basement, you don't always have time.)
This is off the cuff. It will almost certainly change.
Or maybe they want the protocol done in a way that NSA CAN subvert any router detouring it's packets through their own computers, sniffing and injecting (cocaine & herion?) to their hearts content.
Just because I'm paranoid doesn't mean they aren't out to get you.
(He says, from his satellite connected hide-away in rural Alberta, 500 km from the nearest chunk of American soil)
To answer your question specifically: I have a 2 core linux box with 8 G ram. But Mac+iTUnes is the best way to work with my iPod.
I run a copy of WinXP in VirtualBox just so that I can run two applications: Whizfolders -- an outliner tool that has no real equivalent in unix -- and IMatch a photo management tool that has features that I can't find in a unix tool.
Anyone who is writing software for multiple OS's needs an easy way to test that code on those same multiple OS's
These numbers don't seem reasonable.
We will assume the dirtiest possible generation of electicity: Coal.
Coal runs about 6150 kW-hr thermal per ton. = 6785/tonne (metric ton)
= 6.7 kW-hr per kilogram.
We'll assume 40% plant efficiency. So now we are down to 2.8.
7 grams of carbon dioxide is produced from 1.9 grams of carbon.
1.9 grams of carbon = 12.8 watt hours.
This seems to me to be an unreasonably large amount of energy for a single search.
If we take the search I just ran for the kilowatt hours per ton of coal as typical, it took .3 seconds. At this rate searching requires a power level of 150,000 watts.
There are all kinds of arguments that this time shouldn't be used: Much of that delay is waiting for spinning disks. And who knows how many simultaneous searches are done by that one server.
You can also make the distinction between the differential cost of searching versus the aggregate cost. The first is the carbon cost of an additional search. You could calculate this by dividing the carbon cost of putting a new search server on line and running it by the nubmer of searches per hour it could handle. The second is done by taking Google's total power consumption and dividing it by the total number of searches.
When you compress air it gets hot. A lot of the work you do goes to heat. As it cools, the pressure drops. The net effect is a huge loss of efficiency.
If you can pressurize it and keep it from cooling off, you can get more of the energy back.
Anybody got the figures for compressor efficiency?
Yes. It all works smoothly at 20 to 40 miles per hour. Most of the track isn't flat enough to run at 70 mph.
I think the width of the track dates back to the habit of wagon makers in England. Early rail had no particular need for a size, so they made them the same width as wagon makers. Meant the wagon makers didn't have to retool as much to make rail cars.
Wider tracks make a train more stable, but make it harder to make corners. Currently train wheels are welded to the axle. Both wheels turn at the same speed. The wheel itself is somewhat cone shaped. As you go around the corner, the axle moves to the outside of the corner, so the outside wheel is riding on the fattest part of the cone, while the inside wheel is riding on the thinnest. If the wheel doesn't move the right distance, there is slippage between the wheel and the rail, resulting in a screech.
But this means there is an optimum speed for each corner, depending on radius of the corner, how heavily loaded the train is, and probably six other things previously only thought to affect the flavour of pizza.
Most of the mainline track in Canada is twinned. With twinned track you can actually have a fast train pass a slow train, but it requires tricky scheduling, and long spaces between the slow trains. If there is enough demand for freight to keep the slow train track slots filled, then running fast trains gets very hard.
Branch lines are largely single track. Which means that one train has to get off on a siding for a train going the other way. or to pass. There is a lot of momentum is a 100 car freight train. They are not nimble, not sports cars. Branch lines are designed to have a few trains per day. I suspect that many branch lines seldom have more than a single train on them at any given time.
Steel on steel is by far the least energy intensive way to move stuff on land -- but you need to have the tracks. Rail transport is expensive to set up and requires heavy use to be economically viable.
I live in Canada: The population of California spread out mostly in a strip 100 miles wide by 4000 miles long. Here, even as a freight transport rail is having trouble. Oh,the mainlines are still functional, but more and more of the branch lines are being cut off. Farmers used to truck their grain to the nearest elevator, which was the nearest town, usually under 12 miles away. Now they are having to drive many times further.
We still have passenger service on our mainline rails, but it costs 2-3 times as much as flying, and is promoted as a tourist thing. Freight traffic generally moves much more slowly than passenger traffic. On dual track lines, you can set it up for fast trains to pass slow trains, but it makes the scheduling much trickier. During the height of the grain shipping season, the CN mainline has a train roughly every 20 minutes. Typically 100-120 cars long. At 40 mph these trains are 15 miles apart. Since one of the rules the conservation of wheels, for every train going one way, there is one going the other way. Weaving an 80 mph passenger train between the freight trains is tricky.
Light rail is making some progress in cities. Edmonton started a network 30 years ago. There may be 20 miles of track now.
The Toronto metro area has pretty good commuter trains running parallel to the lake. If your destination is close to the rails, you can cut your commute time in 1/2. But Toronto metro has over 10% of our population.
The other extreme is the terratories north of latitude 60. An area sized on the order of the continental U.S. (depending on how you count water spaces between islands...) that has under 100,000 people. Of that 100,000 80% are in/around 5 towns. An awful lot of Canada has population densities that make Montana and Wyoming look crowded.
The OP asked how to monitor. Most of the above is on prevention.
Encryption is part of it.
Part too can be some form of chain of custody.
Each document has some form of who did what. Version control with change logs.
This way you know who had the document.
But consider that Office apps have a bunch of hidden data in them. So mark each document with multiple flags so that each copy checked out is different. This can be done either in the hidden part of the document, or by setting subtle style flags, (Right margin on page 3 is 1.495" instead of 1.500 inches Font for Heading 2 is 15.95 points, not 16 ) You have to have enough flags so that a bunch can be changed and there are still enough to uniquely identify the version.
This way at least, if your customer can get the 'revealed' document, you have a chance of finding out what part of your organization is insecure.
Another possibility is to only allow editing of documents via a virtual machine located in the server room. There is no copy on the local machine. Connections to the virtual box are through the company VPN. If the document is printed locally it is imprinted with the version, who it was printed for, and where it was printed.
Indesign is fabulous for layout. It's not appropriate for composing.
Part of this depends on how you work. One of the best programs I've found is a winsnooze program called WhizFolders. This is really an outline processor. You build a tree of ideas/points, then write a chunk about each idea as the leaf of the tree. Re-arrange the tree as you wish.
If you are semi-scatterbrained, like me, this has a lot of appeal, as you just slot new ideas as they occur to the more-0r-less right place, then carry on with what you were doing. Of all the outline/composing programs I've used, this one does the least to get in my way when doing something that is longer than 10 pages.
Alas it's not free, but it doesn't cost a lot.
It does not do formatting beyond something very primitive, but you can export a document as RTF then take it into FrameMaker, InDesign or (shudder) MS Snerd, and do the final tweaks there.
I'm not a programmer -- ok not quite true. I'm a sysadmin, so I spent a lot of time writing perl, but most of my life was fixing hardware and configuring things to work together. In most of my jobs I was the entire IT department: network, servers, clients, software, hardware, specs, purchase orders which meant that I learned do a lot of different jobs badly.
For 15 years I worked at a university. And yes, I put in some long days. But the work week was nominally 35 hours, not forty; the U gave us all a week without pay between Christmas and New years. (At our option it was averaged into our monthly cheques) We started with 3 weeks holiday per year. And my bosses would let me record my hours, and take time off in lieu.
Later I took a job with a hardware development company, with about a 25% raise. Biggest mistake I ever made. 50-60 hours was expected of you, initially for 'the last push' to get a product ready to go out the door, but these pushes breed like rabbits.
The company was fun, small and exciting the first year. Within a year it had become totally dilbertized, cubicals and all. I saw the writing on the wall when it got to the point that I needed three signatures to buy a $110 dollar ethernet card, one being at head office in Seattle.
For the last 5 years I've been working half time for a local school at 40% of the pay I was getting in industry. However as part of that job:
* 5 weeks of the year was canoeing, hiking, snowshoeing and orienteering, which I love.
* I could bank hours in bad weather and take it off in good weather.
* My commute reduced from 75 minutes each way to 12 minutes each way. In good weather it was reasonable to bike to work.
This gave me time to work on my tree farm.
The school has closed. So I'm working on the tree farm full time, and doing house renos that I've postponed for years.
Whenever I've flown over the landscape it seems to me that forests are much darker than either grasslands or most crops.
If you plant forest, it has a one time 50-100 year impact soaking up CO2 before the biomass reaches an equilibrium between growth and decay.
The only biosystem I'm aware of that sequesters carbon indefinitely are peat bogs. The little beasties create enough acidity that the previous generations don't rot. Of course they ahve to be in a situation where there continues to be standing water as the layers get thicker.
This is one of the reasons that climate warming affects arctic regions to a greater degree than tropical regions. Warm up the arctic, and the treeline moves north. Spruce (predominate tree line species) has a very low albedo. This in turn means that the net reflectivity of the region drops, which further warms the region. Positive feed back loop.
A better plan generally is to harvest biomass for energy, but instead of burning it all, use gassification systems,to make much of it into charcoal. Return the charcoal to the soil. Pure carbon has a very long residence time, and acts as a surface for nutrients to cling to. Some experiments with doing this in tropical rain forest slash and burn agriculture have found that the soil become quite productive for long periods.
What do you want!
GLOBAL WARMING
When do you want it!
NOW
What do you want!
GLOBAL WARMING
When do you want it!
NOW
As it write it's a day or two to the official start of winter, and the temperature on the front porch is -34 C.
This is some 20 degrees C (36 F for the metricly disadvantaged) below normals for this time of year.
Let's take a look at this. To be viable you would have to evaporate an amount of water that is significant compared to the water that is already evaporated. Given that a largish thunderstorm cell has energy content on the order of an H bomb, methinks that the budget is going to be a bit stretched.
For houses this works well. For planets, no.
Also evaporating water may drop the temperature, but it doesn't drop the heat. You're just storing it as latent heat of water vapor. In that form it can still play with the rest of the climate engine.
Even if it were possible, there are good reasons to think that it MIGHT be counter productive.
1. Water is a potent greenhouse gas, blocking different IR bands than CO2. Increasing the average water content of the atmosphere may well cause the average temp to go up.
2. Enough water vapour and you increase the number of clouds. Increasing the number of clouds can push the climate either way: Thick clouds tend to cool the earth, thin clouds tend to warm it. Last time I checked (some years ago) cloud modeling was one of the sticky points in climate models.
3. If you used fresh water for this, it's going to put a major drain on our fresh water supplies. If you use salt water for this, you will put a huge amount of cute microscopic salt crystals in the air. These act as condensation nuclei for water droplets. The formation of rain is dependent on the number of nuclei. Too few and you get a few large drops of rain that result in light rainfall. Too many and you get masses of cloud with drops too small to fall, or that evaporate on the way down.
A possibly more viable form of climate control would be to use H bombs to turn mountain tops into stratospheric dust. We have significant data that large volcanic eruptions can cool the atmosphere for a few years. I don't know if anyone has figured out how much of this is due to dust, and how much to sulfates. Sure this method increases the background radiation. But I think most people would take a 1-2% increase of cancer in 20 years rather than become a refugee of rising ocean levels.
(Caveat: I've not done the math. How many bombs a year does it take to do a Krakatoa? What is the radiation release of an H bomb designed to pulverize the maximum amount of rock and inject it into the stratosphere? I submit that the math for this is less than trivial.)
As some one who managed a school's network, I've found that the kids are moderately clever at defeating what you do. You can make the off limit stuff harder or slower, but not impossible unless you lock the computer down to the point that it is impossible to use for much of anything.
A good part of this is education, both of kids and of parents. We had a set of rules -- a code of conduct or appropriate use, and fairly limited filtering. E.g. I 'enforced' a no porn rule. My definition was simple: If you are uncomfortable with the thought of your sister or mom (boys' school) then it's porn. Monitors in the lab faced hallway windows. First offense, I took the kid aside and explained why I considered inappropriate. Second offence he had no access until he wrote 200 times (by hand) "I will not use school computers to surf porn." At this point they either get clever enough to not get caught, or they get alternate sources for their pix.
Similarly part of the intro process is appropriate use of social networking sites. We talk about the risks, the damage that group gossip can do, the possibility of predators.
Your situation will be different. You're somewhat younger than my group. (I had 7-12) You have full time access by the kids. Your filtering will be more limited.
You need to educate the parents as well as the kids. Explain to parents what the issues are, and suggest to the parents that depending on their kid, they may want home rules such as, "You use the laptop at the kitchen table when your mom or I are in the room." Or "Please leave your bedroom door open and the screen facing the door."
You should have a system of sanctions for inappropriate use. These can range from "Your laptop needs to stay after school" to blocking internet access outside of school to forced rebuild of system and applications folders.
OpenDNS is your friend. It makes it easy to block the worst of the really raunchy sites.
As a second thread:
Your ownership period is way too long. The lifespan of a laptop is short -- typically only a couple years. Then too, consider that even a state of the art laptop given to a 6th grader will, 6 years later, be totally lame.
I suggest that laptops stay in active use for 2 years, then you replace them all. Or if budget is a big constraint, then 3 years. If really a constraint, then 3 years in senior high, then move them to junior high so the kids have a bright and shiny laptop to look forward to making the transition from grade 9 to grade 10.
On lists I've been a member of that discussed laptop per child ownerership the consensus seemed to be that you needed 10% over for spares, and that had about 5% mortality (uneconomic to repair)
I know of one school that had a two year cycle. They would lease the laptops for 2 years, and replace them all simultaneously. This made imaging easier, as they all had identical hardware. Now that was winsnooze, which is a lot trickier in terms of maintaining multiple images.
Provide the functionality, I will buy.
Yes, II have a fair selection of free/cheap apps on my phone. But the latest one, Omnifocus, at $20 is worth every penny of it's price. I've bought the Mac version ($80) for it too and they sync flawlessly.
For any of you who are iPhone/iPod touch developers:
I would pay $100 for a program that allows me to replace my pencil and clipboard with an ipod touch for taking inventory at my tree farm. Contact me for specs if interested.
Some of this depends on what the previous experience of these gifted kids (sorry, I like the term...) is. I found in teaching that a lot of kids in grade 8 and 9 had a real struggle with the abstractions of algebra, and a year later had no trouble with it.
On the flip side, I see kids in my computer lab during their free time playing with exceedingly lame java script games, the best of which (while still lame) allowing construction of courses for their 2 dimensional glyph motorcycles.
I think there is a lot to be said for logo. You get a connection between text and graphical elements, a connection between abstract and "real"
As to senseless shapes: That can change very quickly. I saw one logo into class where the teacher spent most of the first hour teaching kids how to type the program in, and getting across to them that "almost right" wasn't good enough for the computer to understand you. By the end of the class each kid could draw a box. Hello World!
By the end of the second period he'd introduced procedures and they were drawing mullioned double hung sash windows.
The end of the next period saw them drawing a house, then the first assignment -- to draw their own house.
I was fascinated just watching this class and stayed after school to play myself.
With a combined graphic/command world you could implement all sorts of neat things that would keep kids fascinated:
Swarm behaviour, automata, fractals, simulations.
Actually it will be quite large.
Consider: By the time you are in geosync orbit, you are moving sideways at 3 km/s (orbital velocity) The carrier only lifts you up. That sideways velocity has come from the tether.
This is why most space elevator proposals I've seen attach a small asteroid to the upper end of the tether. A carrier arrives from down below. The asteroid slows down a bit, and drops behind the anchor point on the earth. Now it is no longer directly above the anchor point. It acts like a pendulum, and swings back.
When I'm sending more mass out than is coming back the net effect is to slow down the earth's spin.
It's gotten significant attention by enough peer reviewed journals that include a lot of people who have education in physics.
The usual conclusions are that it's a tough engineering problem.
The NS article is sparse on what assumptions they made, but the illustration shows the carrier as being large compared to the tether. What I've read is the other way around. The tether is massive compared to the weight of the carriers.
The NS article goes on to say that they may have to move slowly. I don't see how this helps. As postulated the tether has no real damping mechanism. If I go up slowly or fast force times time will be the same product for the trip.
The key is going to be the ratio of the tension to the mass of the cable. Suppose that the tether has a natural resonate period of 2 days. Traveling up it in 2 days will 'pluck' it with maximal force. The tether will continue to vibrate for a long time.
Now send a second carrier up 1 day out of phase. To first approximation, it will cancel out the vibration induced by the first carrier.
Consider the situation when there is a carrier every kilometer both ways on the tether. At this point the coriolis forces are balanced. (Or not quite: They are balanced when the carriers are passing each other.
There are going to be higher frequencies on the line. Doubtless it will take some very clever engineering. But it's an engineer problem, not a physics problem.
Do remember Clarke's first law: When an (elderly) scientist says something is impossible, he is almost certainly wrong.