Here in Rural Canada we have clusters of post boxes every few miles. Delivery is only 3 days per week. Small packages are left in the mailbox. For larger ones they leave a card that directs us to pick it up at the post office, 7 miles away. They even use laminated cards so they can slap another sticker on it, and reuse them until they get frayed at the edges.
In some places you have to present the card to get your package or jump through hoops to prove who you are. But our town is small enough that all the posties know everyone so they are reaching for the parcel as we come in the door.
An earlier poster commented that Canada has the same problem of 'surplus geography' that the U.S. does. Not so. Canada is overall more urban than the U.S. Generally the transportation networks out ran the settlers, so settlers were concentrated around the railways. So there are bunches of tiny towns (1000 people) and large cities, with the vast majority of people in the large cities. 90% of Canadians live within 100 miles of the US border.
There are very few midsize towns -- towns with 10 to 100 thousand people. What few there are, are pretty much on direct routes between larger centres.
The postal systems have an opportunity: Lower price parcel delivery that isn't in as much hurry as courier. In Canada we have Purolator which is run by the post office, and we have UPS which works pretty well. DHL is also pretty much nation wide, but they don't have the same access network, so they are mostly limited to commercial accounts.
I use greyhound a lot for the 20 to 80 lb size of shipment. It's basically about $30-50 bucks anywhere west of the Lakehead, and will get there in 48 hours.
It seems that with rather trivial work on DNS lookup software you could set them to check the 2nd level domain name component in set order:
E.g. When looking up gerber.xxx DNS first of all looks up gerber.com, gerber.org, gerber.net gerber.edu gerber.mil and gerber.gov
If there is such a listing, it redirects to that instead. Thus you could register gerber.xxx but you would have to go out of your way to actually see it.
Indeed, I can see a case being made to require 2nd level domains to be globally unique. E.g. if you register foo.com, you cannot register foo.org or foo.net.
You are correct. Typical motor to track efficiency is on the order of 85%. So you put in 60 MJ, the train now has a kinetic energy of 0.85 * 60 = 51 MJ. It takes a fairly small amount of power to keep it running at that speed. When you stop it, your 85% comes into play again, so you get 0.85 * 51 = 43 MJ back.
I may be wrong about that 85%. Large motors tend to be more efficient, so perhaps it's 90.
Subways and transit buses are appropriate for this tech, as they start and stop frequently. If it's many minutes between stops, then the start/stop energy is small compared to the running energy.
Another place where this is appropriate is on railways that go over mountain passes. Done right, the train running downhill provides most of the power for the train coming up.
In a previous life I was a school teacher. As of 3 years ago I was unable to find *any* decent Computer aided instruction (CAI) for math.
The best of a poor lot, presented a problem, which you had to either guess the answer or work out on scratch paper, then enter an answer.
A well written Math CAI:
A: At it's core is a Math Processor -- in an analogy to a Word Processor.
Just as a WP can help you create words, allowing you to type faster than you can write, having spelling and grammar correction, allowing you to insert details, or take out redundencies, allowing you to arrange the flow of your thoughts, so then a MP would need the following characteristics:
1. Entry: You have to be able to enter math faster than you can write it out by hand on pencil and paper. And just as it takes a few weeks to learn to touch type, so I'd expect that it would take a bit of time to learn how to touch math-type. Perhaps this is the place for touch screens.
2. Marking mistakes. If working an equation in several steps it should tell you when the current line is not equivalent to the previous line. This alone would go a long way to help kids learn math.
3. Good visualization. It should be easy to graph, to animate graphs e.g. show that the slope between two points on a function approaches a limit, as the points converge.
This program should NOT be mathematica, although I think that Wolfram would be someone to talk to about this. MMa does too much for you, and the last time I looked was missing #2, and clunky abouy #1 and #3.
B: Under the hood, a Math CAI looks at the mistakes the kid makes. Some years ago, an elementary school teacher/researcher discovered that kids had 'buggy software' in their math skills. E.g. a kid would see
184 - 62 and correctly write down 22
But when presented with 162 - 84 would also write down 22. The 'bug' in this case was that the kid subtracted the smaller digit from the larger.
He found that if you analyzed the mistakes looking for patterns, that most kids had a small number of bugs, and that fixing them was fairly quick.
C: Lot of empirical evidence that shows we learn best when just at the edge of our ability. Good CAI needs to be tuned to provide problems that are easy enough for a reasonable success rate, and hard enough to not be boring.
D. Good CAI would be developed by a team of teachers, with different methods of teaching. Instructional snippets would be recorded from various teachers. Ideally every concept is presented 6 different ways. The CAI program would track which approach worked with a given kid. If he practice showed that he didn't get it, try a different teacher's instruction to it. With time, the CAI would know which methods worked with a given kid.
This approach can work well for arithmetic and algebra. Geometry would be tougher. Higher math -- number theory, topology, group theory and their ilk would be a lot tougher.
Journalists are taught to write toward the 6th grade reader. (This may or may not be an advantage for dealing with PHBs) A friend of mine who ran a weekly regional news magazine wouldn't hire a journalism major. He preferred history and political science majors. He claimed, "I can teach any smart person how to write. But I can't teach them what to say. Journalism majors don't know enough history and politics to analyze breaking news in our cities and provinces."
If you want to learn how to write, any major, or indeed, any elective course with significant essay requirements will help you learn to write. Pick what interests you. History (with a technology emphasis) Political Science, whatever. Technical writing would be good.
I agree with previous poster that accounting would be useful. May want to take intro courses in marketing and advertising too, although it's not clear to me that this can be taught, given how awful most of it seems to be even in the nominally skilled.
If you are interested in computer control then see if any of the software engineering courses apply to you.
It also may be worth visiting every car factory, every supply factory you can and talk to the guys who are working there. Often with big plants there is a small number of bars where people hang out after shift. Good place to get them talking.
A final possibility if you have to pay for this on your own without loans: Start by picking up a millwright ticket at a trade school. The pay is good, it aligns with your interests. Here in Alberta the program alternates 8 weeks of class time with 1500 hours of work experience. If you can get into the oil/refinery patch it's easy to get 60 hours a week.
Keeping both gasses mixed would make a dandy bomb, but H2 is sufficiently more mobile that I suspect that separating it from O2 with a membrane would be fairly easy.
The ideal for this would be to keep both gasses handy and use them to operate a fuel cell at night. In practice, the energy to compress and store H2 in reasonable sized containers presents engineering difficulties.
A method to cheaply turn sunlight into an energy rich chemical (H2, CH4, CH3OH) expecially if the process could be ramped up and down at will would make alternative energy much more practical.
Of the 3 ethanol is a liquid at normal temp and pressure, and can be used with minimal conversion in gasoline engines.
Exactly because water is scarce on the moon. Some form of purification step is needed to turn pee into drinking water.
Here on Earth most of the water is unusable for non-biological reasons. Salt in sea water. Carbonates in well water. Various contaminants such as heavy metals, H2S.
Water is also a source of oxygen (breathing) and Hydrogen (rocket fuel) and is the easier way to store both. With a large energy source on the moon it will be a lot cheaper to ship liquid (or frozen) water than to ship LOX and LH2
***
Whether people live on the moon will depend on the economics. Bad idea to say 'never' to an economic notion. See G. Harry Stine's book "The Third Industrial Revolution" for details.
Consider right now the economics in the oil patch. Compare the cost of a mobile platform deep water rig to the costs of the crew. lf there is a sufficiently valuble product people will go there to produce it.
As an example, consider some of the Near Earth asteroids. What is the value of a cubic kilometer of a nickle iron asteroid moved into above synchronous (> 40,000 km) orbit?
1 cubic km = 10*9 cubic meters. = 5 E12 kg of what amounts to impure stainless steel. At $100 per pound that's worth 500 trillion dollars just because it is mass in orbit. (And right now no one can do $100/pound to orbit.)
Ni-Fe meteors assay out at significant amounts of gold and the platinum group of metals. Something like 0.1% Even a 1 km3 rock has more of these metals than we have ever mined.
***
During the age of colonization, colonies died. Not just people. Whole colonies. This was new technology. It was expensive. Compare the cost of sending exploration/colonization ships to the governmental budget.
I don't expect to go there next Tuesday. But after seeing the changes in the last 50 years, I am reluctant to say that it can never happen.
Even in small businesses the transition to a new person at the helm is fraught with peril. Indeed this is a leading cause of failure in companies over 5 years old.
There are companies that specialize in helping small businesses come up with a succession plan (often family...) that reduces the risks from this transition.
Parent commented about the 'money grubbing bean counters'
This is a problem in many sectors of the economy, most notaably in the the financial sector.
IMHO the answer to this is to make upper management remuneration dependent on the next 20 years success of the company. E.g. THIS year's pay is $100,000 plus the stock dividend on 100,000 shares for the next 20 years.
NEXT year you get $100,000 plus your first dividend cheque plus the stock dividend on another block of shares for the next 20 years.
Put in 20 years as CEO. You're getting dividends on 2,000,000 shares. And you better leave the company in good shape, because you're getting dividend cheques for 20 years after you leave.
Anyone who has any history at all of watching meteor showers knows the problem the moon can be.
Every reasonable almanac (and their online equivalents) has both full moon dates, and meteor shower peak dates.
Every full moon rises at approximately sunset. To be full, Sun - Earth - Moon have to make a 180 degree angle.
So where is the news?
****
Note to the owners of slashdot:
This reader is giving serious consideration to canceling his subscription.
1. Repetitious stories.
2. Summaries that are either sensationalistic, vague, or incorrect.
3. Inability to customize which stories I get. (I have no interest in games, little interest in the latest shenanigans of the RIAA, programming language wars..)
I would rather get the story a day later and have it right.
I would like summaries good enough that most of the time I don't need to go back to the article.
I would like a way to rate the story, instead of just modding the posts about the story. I would also like a way to rate the source, so that after a while you mark sources (gizmodo, popular science etc) with their average rating, as well as marking submitters with an average rating.
I would like a system that is smart enough that after I've rated 20 stories submitted by RedHotPickle as being of no interest that MY copy of/. no longer gets stories from RHP.
And yes. I completed the online survey. I ranted there too.
If the machines have the horsepower for this run XP under virtual box and use an immutable boot image. The VM mounts user directories either off the local machine, or in a lab, off a network server.
This makes maintenance far easier. Kids will do all sorts of things unless the machines are locked down, and this is one of the easier ways to do that.
Alternately, use XP, but lock them down with DeepFreeze
Why oh why is there not a data archive format at the file system level for optical media?
Why isn't there enough redundency and enough scatter to the information.
I should be able to:
1. Break a disk in half, glue back together with crazy glue, and read the disk.
2. Cover up to 45% of the disk with paint. ANY 45% and still recover my data.
Sure, reconstructing would be difficult. Capacity would be down. But I'd like to have an archive format that I can write to, and put in a box along with a reader and be reasonably confident that my grandkids can read it.
My main yard is about 3 acres. In addition I have 6 acres of tree farm with grassed aisles. I use 3 mowers: A self propelled walk behind to do the trim, and a Grasshopper ZTR to do the yard and aisles. In addition there are open grassed areas for future expansion of the trees, as well as 5 miles of 6 foot wide trail in the Christmas tree maze that I mow with a 6 foot bush hog behind my tractor. A full mowing takes 8 hours of grasshopper time, 3 hours of walk behind, and 5 hours of bushhog. Fortunately we have a climate that a full mowing is only required about 4 times per year.
So I consider myself experienced.
The math of the OP is flawed.
1. Every ZTR mower I've seen does not mow as a disk, but rather has 3 blades arranged in overlapping echelon.
2. Modeling the mowing as chain of vertexes on a triangular grid can miss parts of the interstitial spaces between the array. In addition there are serious edge issues.
3. Perhaps the OP is concerned about path length, but the primary concern for me is time. Cornering costs time. So the problem is more complex. A turn at greater than a certain radius can be taken at full speed. At tighter corners than this, you have to slow down. (On my grasshopper, I can go fast enough that I end up drifting through the corners. Which is a reflection that I need new tires.) On the bushhog, behind the deutz, cornering too fast will tilt the tractor (It runs 16 psi tires) a couple inches which causes the mower to scalp the lawn. (3 point hitch and 1 rear trailing wheel.)
If you doubt that cornering takes time, try modeling your path as a Peano (Hilbert?) curve
Note to original poster. Feel free to come and mow my yard at any time.
The article posits a system where the system tells you RIGHT NOW what the price is. So 200,000 sensors close relays and charge Chevy Volts when the price hits 5.035 c/kwh, that being the lowest announced price. This causes the the system to be unstable.
A better way to do it is that you have to program in your price schedule ahead of time. So you do something like this:
This on the basis that you needed a minimum 2 hour charge to get to an from work.
In actual fact as electric cars join the smart grid, I would expect the difference in prices to vanish or become microscopic
***
A second way to fix the problem: High demand smart outlets would ramp up slowly. Or there is a 2 minute random delay on their start up.
***
Obligate power:
Good engineering can get around some of these. Example: Refrigerator: The box gets the freezer down to -40 when power is cheap, and uses that as a could source, to keep the refrigerator cool. The refrigerator part itself is brought down to just above freezing when power is cheap, and allowed to go up to 37 F when power is expensive.
Air conditioners instead of cooling air, chill water which is used later. (It takes about 300 gallons of water to store a day's worth of coolth. With well insulated tanks, the air conditioner could also be your hot water heater. This can be made really efficient if you have a variable volume of hot water so that the AC can work most of the time with ground temp water.
Suppose that this material had a phase change specific heat comparable to the liquid/solid change of water. 330 kJ/kg.
By comparison the specific heat of water is only 4 kJ/kg
Let's consider a house 20x40 feet with R30 walls and an R40 ceiling.
It has 120 linear feet of wall x 8 feet high so it has roughly 1000 square feet of wall. At R30 that's 33 btu//hr/F. The ceiling is 800 square feet at R40 so it's 20 btu/hr/F. So our cute windowless box takes 53 btu/hr/F
In 12 hours it will use 640 btu/F. I chose 12, figuring that the temperature would spend half the day being too hot, and half being too cold. Hmm. Two kilograms per degree F. It would take 160 kg of water to do the same thing with a 2 degree F fluctuation in temperature
If we assume a desert climate that averages nice, but has a 20F on either side of nice, now it takes 40 Kg of this wonder material.
Now that specific heat is too high by a factor of 3 or so.
Now we are up to 120 Kg.
And the heat/cooling load of a typical building is evenly split by the insulated walls and the windows. (Windows are normally 10% of floor area, but have 10-15 times the heat transmission unless you go high tech.)
So that adds a factor of 2. 240 kg.
In a heating environment air changes are ususallly responsible for another factor of 2 in heating. This can be reduced considerably by using a heat recovery ventilator. But lets be passive.
500 kg of wonder stuff.
I started out this note to show that it was unreasonable to do this. It looks quite feasible.
A sheet of gypock is, what, a kg/square foot. A house typically is 2 rooms thick, so it has 2000 square feet of wall gyprock, and 800 square feet of ceiling gyprock. So if roughly 20% of the gyprock mix were wonder stuff, it could handle a 20 degree excusrion on either side of nice.
However, it would make for an awfully thick layer of paint.
There are two types of DGPS. In one, you have a receiver at a known location. It basically tracks the atmospheric errors in the signal propagation, and broadcasts a set of corrections. WAIS does this at a more broad level. WAIS brings the intrinsic error down to about 2-3 meters. This first form of DGPS brings it down to a half meter or so.
The next level of DPGS, which has another fancy name that escapes me at the moment, depends on tracking the phase of the satellite signal. Lose track of the signal even momentarily, and you have to go back to a known good point, and reset it. It's good to sub centimeter accuracy.
In some of the forums the levels are referred to as "Navigator grade", "Map maker grade", and "Survey Grade" GPS
Inch level would be cool. If the units are inexpensive enough, this could mean real time positioning for cars.
3D location could be improved by locating one of these boxes at the top of every cell phone tower.
I am skeptical about coverage. If they use enough power to cover several square miles per transmitter, I think there will be a lot of unhappy wifi users near the transmitters.
One of the largest problems with the whole climate sideshow is that you have a very small signal (1 to 5 degrees C per century) buried in a mass of temperatures fluctuating 15-30 C per day.
It doesn't help that large numbers of your instruments have changed during the measurements, and another large number have become embedded in urban spaces.
(Recently they changed the specification for the paint used for the Stefanson (sic) boxes -- the white louver sided instrument hutches in the met office back yard. They were still painted white, but the pigment didn't have the same IR absorption band. Made some fraction of a degree difference. So now, to tease the signal form the noise, you have to track down when the boxes were painted, and what they were painted with.)
It doesn't help that global warming doesn't mean everywhere warming. Some places cool. But other places warm more than the some cool.
People aren't very good at perceiving very slow baseline change in a chaotic and rapidly varying series. (Look how many people do not see the gradual depletion of their wealth while gambling...)
Add to that: People are set in their ways. The changes required to stop CO2 buildup are non-trivial, and are buried deep in our infra-structure. For first world countries, the easiest (and not cheap) change is to re-insulate our houses. Doing this in a way that is both effective and healthy is not trivial. (Too often insulating and air sealing a house results in mold and mildew growing in the walls. The house has to have active ventilation. The detailing becomes important. Hole in the vapour barrier in the top of a wall, and exiting warm moist air condenses inside the wall. Newer standards for building are better, but a house has an average lifespan between 0.5 and several centuries.
Changing our love affair with the car will be harder, and I suspect that the only way will be to keep increasing the cost of fuel. to the point that people look for alternatives.
If fuel went to 20 bucks a gallon what would change in your life? Would you move to a house closer to your work? Would living quarters be part of the benefits package with large employers?
Houses right now have really narrow side yards. At what point does it become effective to merge single units into row houses, using the between space as semi-heated storage, just to stop the heat loss. (Is this cheaper than re-insulating that exterior wall?
At what point does it make sense to abandon the concept of single detached dwelling?
I'm a farmer. I live 75 km from the Big City. At this point if I take my pickup to town, it's about $30 for gas. Even at this price, I plan my day with some care and try to pick up or drop off a load to amke the pickup's use worth while. At $150 per trip, I'd likely own a smaller pickup in addition to this one, as well as a small fleet of utility trailers to use with the car.
I'm not denying climate change. Living on the land I see the change in new weeds, new bugs. I don't think we can stop it, and that our efforts to put Pandora's troubles back in the box are futile. Rather, we need to get off our butts and learn how to adapt. Ecologies are changing. We need to become ecological engineers to manage this change.
It doesn't have to be given escape velocity. It doesn't even need to be in geosync orbit. All it needs is a higher orbit. It cost 10,000 bucks a kg to get that mass up there. A permanent space presence is going to require raw materials. A permanent presence is going to require living quarters. A place to put spare socks and oxygen bottles.
At one point there was serious discussion about carrying the shuttle external fuel tanks to orbit. Turns out you gain very little by dropping them. Had we taken them to orbit we would have some prime real estate for LEO manufacturing.
Given the shape, bringing it down where you want is going to be tricky. Not exactly an easy shape to model for drag. Raising it may be cheaper than Razing it.
This is not a new idea. I remember it on Popular Science in the 70's.
I've also seen it proposed running the other way. The tower is white to not absorb light, and you inject water at the top. The water evaporates as it falls, cooling the air in the tower. The heavier air, being more massive, creates a down draft. If you use the right sized droplets, you can use sea water, the water at the bottom is concentrated brine, and you have cold moist air flowing out from the tower. THIS can be used to effectively grow crops. One proposal suggested that the extremely heavy dew could water pasture for miles around.
Or you use the desert to heat water. Inject the water into the updraft. This reduces the lapse rate as the column of air rises, so you get more lift out of a given temperature. Hot water can be stored for night use. This requires a fresh water source. If you use sea water you are creating a plume of salt crystals down wind.
Pizza oven: Cooking produces significant evaporation. If the oven wasn't vented it would fill with steam to the point that non-soggy crusts would be impossible.
More generally: The source doesn't match the supply. E.g. Server farms produce heat at a more or less constant basis. But the heating needs of any attached building vary by season. So you either size the building so that the servers can heat it at the worst time of year, or you have to have dual heat sources for the building.
Most of heat scavengers can best be used for heating domestic hot water. The demand is much closer to constant if you use tanks that are large enough to hold a day's water. Even if you use this as a pre-heat source, it could be helpful.
Air conditioning is truely stupid. It would make far more sense to chill water at night, both to shift power use, and to gain efficiency. However given that ACs are rated in 'tons of ice per hour' it would take substantial amounts of tank.
Example.You have an AC that moves 20,000 BTU per hour. Suppose on a mixed duty cycle it runs 5 hours a day. So that 100,000 BTU. So you chill water to 35 F, and in use you warm it up to 75 F. 40 degree delta means 2500 lbs of water -- about 300 gallons. Multiply appropriately for larger AC or longer duty cycles. Note that you can get some tank advantage by using brine (cheap, but corrosive) or antifreeze (expensive, poisonous) as the coolant.
I'm not much of a gamer. I bought games from Infocomm (Zork...) I certainly didn't expect to get through them in 2 hours or even ten. I think it took several months to get through all 3 zork titles. Maybe I'm exceptionally stupid. (I was probably playing with them a few hours a week)
MS Flight simulator on a 386.
I figure a good game should come in at under $1/hour.
1. If I use multiple computers, how do I log in? E.g. I'm at a friends house, and log in in their web browser. THAT computer doesn't have my private key. Or if I regularly use public computers at a school.
2. If my computer is hacked, what happens to my key collection?
3. If my drive crashes, how do I recover my key collection?
4. If I regularly use linux, mac and windows real and virtual machines, how do I keep my keys sync'ed when Mozilla can't even do this with my bookmarks.
My property taxes last year were $1700. My brother in law in Edmonton, with a house with a value 2.5 times higher pays $1500.
Services: The county grades the gravel road by my house, and sends a bus by to pick up the kids for school. I provide my own water. I provide my own sewage disposal. I take garbage to the dump -- 15 miles away.
The ONLY form of broadband I can get is satellite. The phone company does not provide ISDN or ASDL (We are about 12 km from the exchange) There are 3 different WLAN providers in the area, and I'm out of Line of Sight for all 3 -- unless I build a 70 foot tower.
On the other hand: A busy day has 30 cars go past my driveway.
I never lock my door. I can leave my keys in the truck.
We don't have curtains in the windows. Windows are for letting light in, and vision out. Why block them?
I sometimes hire neighbor's kids to help on the farm, and their parents are casual about dropping them off -- or letting them ride over here on their bicycle or quad.
I can barely hear it when my neighbor mows his lawn.
I don't mind much, but I wish the bill better matched what I get. I really resent the fraction of my taxes that pay for the new county recreation center 60 kilometers away. Thank God I don't get all the government I pay for.
1. Face the outhouse south. It makes use in winter less unpleasant.
2. Make a seat oout of styrofoam. Again, winter use.
3. Anyplace that piss can hit should be coated with metal. Otherwise the salt builds up in the wood, and you get porcupines and mice gnawing the wood for the salt.
4. It's a good idea to have a funnel and pipe for what is effectively a urinal. This decreases the splashing on the seat, and dribbles on the floor. If you have a large enough group to have bathroom collisions, you may want to make a urinal stall on the side opposite the door.
5. All horizontal surfaces should be smooth.
6. All surfaces should be made as cleanable as possible. Good quality floor enamel on the floor. Everything else whitewashed yearly.
7. Separate vent pipe from the pit to several feet above the roof. If painted black, and capped properly it will draw air from the pit and make the room much less smelly.
8. Two part door so you have the option of watching the view while sitting.
9. Screened so well that flies have no access to the pit. Lid on the opening done so that it is fly proof, and it cannot be left open.
10. Sited so that it cannot contaminate your drinking water.
11. Far enough from the house to avoid smells, close enough to be convenient.
12. Built in a manner it can easily be moved to another hole, but also is resistant to being blown over in a strong wind.
He not only got insignificant revenge, he was stupid too.
Any BOFH worth his salt:
1. Would delay any action for long enough to have plausible deny-ability
2, Would make changes slowly and subtly so they were not immediately obvious.
Examples: A script that modfied all the percentages and total/subtotals in a spread sheet by a random small percentage.
Putting a bunch of errors in the financial report that was going tot he printers to be mailed out to everyone.
Replacing 1 file in a thousand with a copy from the backup system every week.
Once a month randomly overwrite some user's hard drive ideally making it look like a hardware failure.
Reconfiguring a few ports on the network switch to lock out a MAC address at random.
Auto emailing interesting files to the competition.
Scale this, starting with small things 6 weeks after you left, and gradually increasing so that the new IT department is scrambling to keep up.
You also have to do this in a way that is not obvious. You want systems where a machine inside the network establishes the connection, using a well known port so that it doesn't make much of a blip in network traffic analysis. And you actually want multiple such machines to avoid losing your access by a smart sysadmin. And it probably would help if you had a computer in a box that was attached to a spare cable and stuffed into the ceiling space.
Slashdot Mirror
Here in Rural Canada we have clusters of post boxes every few miles. Delivery is only 3 days per week.
Small packages are left in the mailbox. For larger ones they leave a card that directs us to pick it up at the post office, 7 miles away. They even use laminated cards so they can slap another sticker on it, and reuse them until they get frayed at the edges.
In some places you have to present the card to get your package or jump through hoops to prove who you are. But our town is small enough that all the posties know everyone so they are reaching for the parcel as we come in the door.
An earlier poster commented that Canada has the same problem of 'surplus geography' that the U.S. does. Not so. Canada is overall more urban than the U.S. Generally the transportation networks out ran the settlers, so settlers were concentrated around the railways. So there are bunches of tiny towns (1000 people) and large cities, with the vast majority of people in the large cities. 90% of Canadians live within 100 miles of the US border.
There are very few midsize towns -- towns with 10 to 100 thousand people. What few there are, are pretty much on direct routes between larger centres.
The postal systems have an opportunity: Lower price parcel delivery that isn't in as much hurry as courier. In Canada we have Purolator which is run by the post office, and we have UPS which works pretty well. DHL is also pretty much nation wide, but they don't have the same access network, so they are mostly limited to commercial accounts.
I use greyhound a lot for the 20 to 80 lb size of shipment. It's basically about $30-50 bucks anywhere west of the Lakehead, and will get there in 48 hours.
It seems that with rather trivial work on DNS lookup software you could set them to check the 2nd level domain name component in set order:
E.g. When looking up gerber.xxx DNS first of all looks up gerber.com, gerber.org, gerber.net gerber.edu gerber.mil and gerber.gov
If there is such a listing, it redirects to that instead. Thus you could register gerber.xxx but you would have to go out of your way to actually see it.
Indeed, I can see a case being made to require 2nd level domains to be globally unique. E.g. if you register foo.com, you cannot register foo.org or foo.net.
You are correct. Typical motor to track efficiency is on the order of 85%. So you put in 60 MJ, the train now has a kinetic energy of 0.85 * 60 = 51 MJ. It takes a fairly small amount of power to keep it running at that speed. When you stop it, your 85% comes into play again, so you get 0.85 * 51 = 43 MJ back.
I may be wrong about that 85%. Large motors tend to be more efficient, so perhaps it's 90.
Subways and transit buses are appropriate for this tech, as they start and stop frequently. If it's many minutes between stops, then the start/stop energy is small compared to the running energy.
Another place where this is appropriate is on railways that go over mountain passes. Done right, the train running downhill provides most of the power for the train coming up.
In a previous life I was a school teacher. As of 3 years ago I was unable to find *any* decent Computer aided instruction (CAI) for math.
The best of a poor lot, presented a problem, which you had to either guess the answer or work out on scratch paper, then enter an answer.
A well written Math CAI:
A: At it's core is a Math Processor -- in an analogy to a Word Processor.
Just as a WP can help you create words, allowing you to type faster than you can write, having spelling and grammar correction, allowing you to insert details, or take out redundencies, allowing you to arrange the flow of your thoughts, so then a MP would need the following characteristics:
1. Entry: You have to be able to enter math faster than you can write it out by hand on pencil and paper. And just as it takes a few weeks to learn to touch type, so I'd expect that it would take a bit of time to learn how to touch math-type. Perhaps this is the place for touch screens.
2. Marking mistakes. If working an equation in several steps it should tell you when the current line is not equivalent to the previous line. This alone would go a long way to help kids learn math.
3. Good visualization. It should be easy to graph, to animate graphs e.g. show that the slope between two points on a function approaches a limit, as the points converge.
This program should NOT be mathematica, although I think that Wolfram would be someone to talk to about this. MMa does too much for you, and the last time I looked was missing #2, and clunky abouy #1 and #3.
B: Under the hood, a Math CAI looks at the mistakes the kid makes. Some years ago, an elementary school teacher/researcher discovered that kids had 'buggy software' in their math skills. E.g. a kid would see
184
- 62
and correctly write down
22
But when presented with 162 - 84 would also write down 22. The 'bug' in this case was that the kid subtracted the smaller digit from the larger.
He found that if you analyzed the mistakes looking for patterns, that most kids had a small number of bugs, and that fixing them was fairly quick.
C: Lot of empirical evidence that shows we learn best when just at the edge of our ability. Good CAI needs to be tuned to provide problems that are easy enough for a reasonable success rate, and hard enough to not be boring.
D. Good CAI would be developed by a team of teachers, with different methods of teaching. Instructional snippets would be recorded from various teachers. Ideally every concept is presented 6 different ways. The CAI program would track which approach worked with a given kid. If he practice showed that he didn't get it, try a different teacher's instruction to it. With time, the CAI would know which methods worked with a given kid.
This approach can work well for arithmetic and algebra. Geometry would be tougher. Higher math -- number theory, topology, group theory and their ilk would be a lot tougher.
Journalists are taught to write toward the 6th grade reader. (This may or may not be an advantage for dealing with PHBs) A friend of mine who ran a weekly regional news magazine wouldn't hire a journalism major. He preferred history and political science majors. He claimed, "I can teach any smart person how to write. But I can't teach them what to say. Journalism majors don't know enough history and politics to analyze breaking news in our cities and provinces."
If you want to learn how to write, any major, or indeed, any elective course with significant essay requirements will help you learn to write. Pick what interests you. History (with a technology emphasis) Political Science, whatever. Technical writing would be good.
I agree with previous poster that accounting would be useful. May want to take intro courses in marketing and advertising too, although it's not clear to me that this can be taught, given how awful most of it seems to be even in the nominally skilled.
If you are interested in computer control then see if any of the software engineering courses apply to you.
It also may be worth visiting every car factory, every supply factory you can and talk to the guys who are working there. Often with big plants there is a small number of bars where people hang out after shift. Good place to get them talking.
A final possibility if you have to pay for this on your own without loans: Start by picking up a millwright ticket at a trade school. The pay is good, it aligns with your interests. Here in Alberta the program alternates 8 weeks of class time with 1500 hours of work experience. If you can get into the oil/refinery patch it's easy to get 60 hours a week.
Keeping both gasses mixed would make a dandy bomb, but H2 is sufficiently more mobile that I suspect that separating it from O2 with a membrane would be fairly easy.
The ideal for this would be to keep both gasses handy and use them to operate a fuel cell at night. In practice, the energy to compress and store H2 in reasonable sized containers presents engineering difficulties.
A method to cheaply turn sunlight into an energy rich chemical (H2, CH4, CH3OH) expecially if the process could be ramped up and down at will would make alternative energy much more practical.
Of the 3 ethanol is a liquid at normal temp and pressure, and can be used with minimal conversion in gasoline engines.
Why is this important:
Exactly because water is scarce on the moon. Some form of purification step is needed to turn pee into drinking water.
Here on Earth most of the water is unusable for non-biological reasons. Salt in sea water. Carbonates in well water. Various contaminants such as heavy metals, H2S.
Water is also a source of oxygen (breathing) and Hydrogen (rocket fuel) and is the easier way to store both. With a large energy source on the moon it will be a lot cheaper to ship liquid (or frozen) water than to ship LOX and LH2
***
Whether people live on the moon will depend on the economics. Bad idea to say 'never' to an economic notion. See G. Harry Stine's book "The Third Industrial Revolution" for details.
Consider right now the economics in the oil patch. Compare the cost of a mobile platform deep water rig to the costs of the crew. lf there is a sufficiently valuble product people will go there to produce it.
As an example, consider some of the Near Earth asteroids. What is the value of a cubic kilometer of a nickle iron asteroid moved into above synchronous (> 40,000 km) orbit?
1 cubic km = 10*9 cubic meters. = 5 E12 kg of what amounts to impure stainless steel. At $100 per pound that's worth 500 trillion dollars just because it is mass in orbit. (And right now no one can do $100/pound to orbit.)
Ni-Fe meteors assay out at significant amounts of gold and the platinum group of metals. Something like 0.1% Even a 1 km3 rock has more of these metals than we have ever mined.
***
During the age of colonization, colonies died. Not just people. Whole colonies. This was new technology. It was expensive. Compare the cost of sending exploration/colonization ships to the governmental budget.
I don't expect to go there next Tuesday. But after seeing the changes in the last 50 years, I am reluctant to say that it can never happen.
Even in small businesses the transition to a new person at the helm is fraught with peril. Indeed this is a leading cause of failure in companies over 5 years old.
There are companies that specialize in helping small businesses come up with a succession plan (often family...) that reduces the risks from this transition.
Parent commented about the 'money grubbing bean counters'
This is a problem in many sectors of the economy, most notaably in the the financial sector.
IMHO the answer to this is to make upper management remuneration dependent on the next 20 years success of the company. E.g. THIS year's pay is $100,000 plus the stock dividend on 100,000 shares for the next 20 years.
NEXT year you get $100,000 plus your first dividend cheque plus the stock dividend on another block of shares for the next 20 years.
Put in 20 years as CEO. You're getting dividends on 2,000,000 shares. And you better leave the company in good shape, because you're getting dividend cheques for 20 years after you leave.
Why is this news?
Anyone who has any history at all of watching meteor showers knows the problem the moon can be.
Every reasonable almanac (and their online equivalents) has both full moon dates, and meteor shower peak dates.
Every full moon rises at approximately sunset. To be full, Sun - Earth - Moon have to make a 180 degree angle.
So where is the news?
****
Note to the owners of slashdot:
This reader is giving serious consideration to canceling his subscription.
1. Repetitious stories.
2. Summaries that are either sensationalistic, vague, or incorrect.
3. Inability to customize which stories I get. (I have no interest in games, little interest in the latest shenanigans of the RIAA, programming language wars..)
I would rather get the story a day later and have it right.
I would like summaries good enough that most of the time I don't need to go back to the article.
I would like a way to rate the story, instead of just modding the posts about the story. I would also like a way to rate the source, so that after a while you mark sources (gizmodo, popular science etc) with their average rating, as well as marking submitters with an average rating.
I would like a system that is smart enough that after I've rated 20 stories submitted by RedHotPickle as being of no interest that MY copy of /. no longer gets stories from RHP.
And yes. I completed the online survey. I ranted there too.
If the machines have the horsepower for this run XP under virtual box and use an immutable boot image. The VM mounts user directories either off the local machine, or in a lab, off a network server.
This makes maintenance far easier. Kids will do all sorts of things unless the machines are locked down, and this is one of the easier ways to do that.
Alternately, use XP, but lock them down with DeepFreeze
Why oh why is there not a data archive format at the file system level for optical media?
Why isn't there enough redundency and enough scatter to the information.
I should be able to:
1. Break a disk in half, glue back together with crazy glue, and read the disk.
2. Cover up to 45% of the disk with paint. ANY 45% and still recover my data.
Sure, reconstructing would be difficult. Capacity would be down. But I'd like to have an archive format that I can write to, and put in a box along with a reader and be reasonably confident that my grandkids can read it.
My main yard is about 3 acres. In addition I have 6 acres of tree farm with grassed aisles. I use 3 mowers: A self propelled walk behind to do the trim, and a Grasshopper ZTR to do the yard and aisles. In addition there are open grassed areas for future expansion of the trees, as well as 5 miles of 6 foot wide trail in the Christmas tree maze that I mow with a 6 foot bush hog behind my tractor. A full mowing takes 8 hours of grasshopper time, 3 hours of walk behind, and 5 hours of bushhog. Fortunately we have a climate that a full mowing is only required about 4 times per year.
So I consider myself experienced.
The math of the OP is flawed.
1. Every ZTR mower I've seen does not mow as a disk, but rather has 3 blades arranged in overlapping echelon.
2. Modeling the mowing as chain of vertexes on a triangular grid can miss parts of the interstitial spaces between the array. In addition there are serious edge issues.
3. Perhaps the OP is concerned about path length, but the primary concern for me is time. Cornering costs time. So the problem is more complex. A turn at greater than a certain radius can be taken at full speed. At tighter corners than this, you have to slow down. (On my grasshopper, I can go fast enough that I end up drifting through the corners. Which is a reflection that I need new tires.) On the bushhog, behind the deutz, cornering too fast will tilt the tractor (It runs 16 psi tires) a couple inches which causes the mower to scalp the lawn. (3 point hitch and 1 rear trailing wheel.)
If you doubt that cornering takes time, try modeling your path as a Peano (Hilbert?) curve
Note to original poster. Feel free to come and mow my yard at any time.
The article posits a system where the system tells you RIGHT NOW what the price is. So 200,000 sensors close relays and charge Chevy Volts when the price hits 5.035 c/kwh, that being the lowest announced price. This causes the the system to be unstable.
A better way to do it is that you have to program in your price schedule ahead of time. So you do something like this:
10 p.m. - 6 a.m. 5.035 c/kwh Turn on Chevy Outlet
12 p.m. - 6 a.m. 5.555 c/kwh Turn on Chevy Outlet
2 a.m. - 6 a.m 6.211 c/kwh Turn on Chevy Outlet
4 a.m. - 6 a.m. 8 c/kwh Turn on Chevy Outlet.
This on the basis that you needed a minimum 2 hour charge to get to an from work.
In actual fact as electric cars join the smart grid, I would expect the difference in prices to vanish or become microscopic
***
A second way to fix the problem: High demand smart outlets would ramp up slowly. Or there is a 2 minute random delay on their start up.
***
Obligate power:
Good engineering can get around some of these. Example: Refrigerator: The box gets the freezer down to -40 when power is cheap, and uses that as a could source, to keep the refrigerator cool. The refrigerator part itself is brought down to just above freezing when power is cheap, and allowed to go up to 37 F when power is expensive.
Air conditioners instead of cooling air, chill water which is used later. (It takes about 300 gallons of water to store a day's worth of coolth. With well insulated tanks, the air conditioner could also be your hot water heater. This can be made really efficient if you have a variable volume of hot water so that the AC can work most of the time with ground temp water.
Suppose that this material had a phase change specific heat comparable to the liquid/solid change of water. 330 kJ/kg.
By comparison the specific heat of water is only 4 kJ/kg
Let's consider a house 20x40 feet with R30 walls and an R40 ceiling.
It has 120 linear feet of wall x 8 feet high so it has roughly 1000 square feet of wall. At R30 that's 33 btu//hr/F. The ceiling is 800 square feet at R40 so it's 20 btu/hr/F. So our cute windowless box takes 53 btu/hr/F
In 12 hours it will use 640 btu/F. I chose 12, figuring that the temperature would spend half the day being too hot, and half being too cold. Hmm. Two kilograms per degree F. It would take 160 kg of water to do the same thing with a 2 degree F fluctuation in temperature
If we assume a desert climate that averages nice, but has a 20F on either side of nice, now it takes 40 Kg of this wonder material.
Now that specific heat is too high by a factor of 3 or so.
Now we are up to 120 Kg.
And the heat/cooling load of a typical building is evenly split by the insulated walls and the windows. (Windows are normally 10% of floor area, but have 10-15 times the heat transmission unless you go high tech.)
So that adds a factor of 2. 240 kg.
In a heating environment air changes are ususallly responsible for another factor of 2 in heating. This can be reduced considerably by using a heat recovery ventilator. But lets be passive.
500 kg of wonder stuff.
I started out this note to show that it was unreasonable to do this. It looks quite feasible.
A sheet of gypock is, what, a kg/square foot. A house typically is 2 rooms thick, so it has 2000 square feet of wall gyprock, and 800 square feet of ceiling gyprock. So if roughly 20% of the gyprock mix were wonder stuff, it could handle a 20 degree excusrion on either side of nice.
However, it would make for an awfully thick layer of paint.
There are two types of DGPS. In one, you have a receiver at a known location. It basically tracks the atmospheric errors in the signal propagation, and broadcasts a set of corrections. WAIS does this at a more broad level. WAIS brings the intrinsic error down to about 2-3 meters. This first form of DGPS brings it down to a half meter or so.
The next level of DPGS, which has another fancy name that escapes me at the moment, depends on tracking the phase of the satellite signal. Lose track of the signal even momentarily, and you have to go back to a known good point, and reset it. It's good to sub centimeter accuracy.
In some of the forums the levels are referred to as "Navigator grade", "Map maker grade", and "Survey Grade" GPS
Inch level would be cool. If the units are inexpensive enough, this could mean real time positioning for cars.
3D location could be improved by locating one of these boxes at the top of every cell phone tower.
I am skeptical about coverage. If they use enough power to cover several square miles per transmitter, I think there will be a lot of unhappy wifi users near the transmitters.
One of the largest problems with the whole climate sideshow is that you have a very small signal (1 to 5 degrees C per century) buried in a mass of temperatures fluctuating 15-30 C per day.
It doesn't help that large numbers of your instruments have changed during the measurements, and another large number have become embedded in urban spaces.
(Recently they changed the specification for the paint used for the Stefanson (sic) boxes -- the white louver sided instrument hutches in the met office back yard. They were still painted white, but the pigment didn't have the same IR absorption band. Made some fraction of a degree difference. So now, to tease the signal form the noise, you have to track down when the boxes were painted, and what they were painted with.)
It doesn't help that global warming doesn't mean everywhere warming. Some places cool. But other places warm more than the some cool.
People aren't very good at perceiving very slow baseline change in a chaotic and rapidly varying series. (Look how many people do not see the gradual depletion of their wealth while gambling...)
Add to that: People are set in their ways. The changes required to stop CO2 buildup are non-trivial, and are buried deep in our infra-structure. For first world countries, the easiest (and not cheap) change is to re-insulate our houses. Doing this in a way that is both effective and healthy is not trivial. (Too often insulating and air sealing a house results in mold and mildew growing in the walls. The house has to have active ventilation. The detailing becomes important. Hole in the vapour barrier in the top of a wall, and exiting warm moist air condenses inside the wall. Newer standards for building are better, but a house has an average lifespan between 0.5 and several centuries.
Changing our love affair with the car will be harder, and I suspect that the only way will be to keep increasing the cost of fuel. to the point that people look for alternatives.
If fuel went to 20 bucks a gallon what would change in your life? Would you move to a house closer to your work? Would living quarters be part of the benefits package with large employers?
Houses right now have really narrow side yards. At what point does it become effective to merge single units into row houses, using the between space as semi-heated storage, just to stop the heat loss. (Is this cheaper than re-insulating that exterior wall?
At what point does it make sense to abandon the concept of single detached dwelling?
I'm a farmer. I live 75 km from the Big City. At this point if I take my pickup to town, it's about $30 for gas. Even at this price, I plan my day with some care and try to pick up or drop off a load to amke the pickup's use worth while. At $150 per trip, I'd likely own a smaller pickup in addition to this one, as well as a small fleet of utility trailers to use with the car.
I'm not denying climate change. Living on the land I see the change in new weeds, new bugs. I don't think we can stop it, and that our efforts to put Pandora's troubles back in the box are futile. Rather, we need to get off our butts and learn how to adapt. Ecologies are changing. We need to become ecological engineers to manage this change.
It doesn't have to be given escape velocity. It doesn't even need to be in geosync orbit. All it needs is a higher orbit. It cost 10,000 bucks a kg to get that mass up there. A permanent space presence is going to require raw materials. A permanent presence is going to require living quarters. A place to put spare socks and oxygen bottles.
At one point there was serious discussion about carrying the shuttle external fuel tanks to orbit. Turns out you gain very little by dropping them. Had we taken them to orbit we would have some prime real estate for LEO manufacturing.
Given the shape, bringing it down where you want is going to be tricky. Not exactly an easy shape to model for drag. Raising it may be cheaper than Razing it.
This is not a new idea. I remember it on Popular Science in the 70's.
I've also seen it proposed running the other way. The tower is white to not absorb light, and you inject water at the top. The water evaporates as it falls, cooling the air in the tower. The heavier air, being more massive, creates a down draft. If you use the right sized droplets, you can use sea water, the water at the bottom is concentrated brine, and you have cold moist air flowing out from the tower. THIS can be used to effectively grow crops. One proposal suggested that the extremely heavy dew could water pasture for miles around.
Or you use the desert to heat water. Inject the water into the updraft. This reduces the lapse rate as the column of air rises, so you get more lift out of a given temperature. Hot water can be stored for night use. This requires a fresh water source. If you use sea water you are creating a plume of salt crystals down wind.
Not clear to me which mode makes more power.
Good answers. I'll add to them:
Pizza oven: Cooking produces significant evaporation. If the oven wasn't vented it would fill with steam to the point that non-soggy crusts would be impossible.
More generally: The source doesn't match the supply. E.g. Server farms produce heat at a more or less constant basis. But the heating needs of any attached building vary by season. So you either size the building so that the servers can heat it at the worst time of year, or you have to have dual heat sources for the building.
Most of heat scavengers can best be used for heating domestic hot water. The demand is much closer to constant if you use tanks that are large enough to hold a day's water. Even if you use this as a pre-heat source, it could be helpful.
Air conditioning is truely stupid. It would make far more sense to chill water at night, both to shift power use, and to gain efficiency. However given that ACs are rated in 'tons of ice per hour' it would take substantial amounts of tank.
Example.You have an AC that moves 20,000 BTU per hour. Suppose on a mixed duty cycle it runs 5 hours a day. So that 100,000 BTU. So you chill water to 35 F, and in use you warm it up to 75 F. 40 degree delta means 2500 lbs of water -- about 300 gallons. Multiply appropriately for larger AC or longer duty cycles. Note that you can get some tank advantage by using brine (cheap, but corrosive) or antifreeze (expensive, poisonous) as the coolant.
I'm not much of a gamer. I bought games from Infocomm (Zork...) I certainly didn't expect to get through them in 2 hours or even ten. I think it took several months to get through all 3 zork titles. Maybe I'm exceptionally stupid. (I was probably playing with them a few hours a week)
MS Flight simulator on a 386.
I figure a good game should come in at under $1/hour.
1. If I use multiple computers, how do I log in? E.g. I'm at a friends house, and log in in their web browser. THAT computer doesn't have my private key. Or if I regularly use public computers at a school.
2. If my computer is hacked, what happens to my key collection?
3. If my drive crashes, how do I recover my key collection?
4. If I regularly use linux, mac and windows real and virtual machines, how do I keep my keys sync'ed when Mozilla can't even do this with my bookmarks.
My property taxes last year were $1700. My brother in law in Edmonton, with a house with a value 2.5 times higher pays $1500.
Services: The county grades the gravel road by my house, and sends a bus by to pick up the kids for school. I provide my own water. I provide my own sewage disposal. I take garbage to the dump -- 15 miles away.
The ONLY form of broadband I can get is satellite. The phone company does not provide ISDN or ASDL (We are about 12 km from the exchange) There are 3 different WLAN providers in the area, and I'm out of Line of Sight for all 3 -- unless I build a 70 foot tower.
On the other hand:
A busy day has 30 cars go past my driveway.
I never lock my door. I can leave my keys in the truck.
We don't have curtains in the windows. Windows are for letting light in, and vision out. Why block them?
I sometimes hire neighbor's kids to help on the farm, and their parents are casual about dropping them off -- or letting them ride over here on their bicycle or quad.
I can barely hear it when my neighbor mows his lawn.
I don't mind much, but I wish the bill better matched what I get. I really resent the fraction of my taxes that pay for the new county recreation center 60 kilometers away. Thank God I don't get all the government I pay for.
You've got the basics right.
1. Face the outhouse south. It makes use in winter less unpleasant.
2. Make a seat oout of styrofoam. Again, winter use.
3. Anyplace that piss can hit should be coated with metal. Otherwise the salt builds up in the wood, and you get porcupines and mice gnawing the wood for the salt.
4. It's a good idea to have a funnel and pipe for what is effectively a urinal. This decreases the splashing on the seat, and dribbles on the floor. If you have a large enough group to have bathroom collisions, you may want to make a urinal stall on the side opposite the door.
5. All horizontal surfaces should be smooth.
6. All surfaces should be made as cleanable as possible. Good quality floor enamel on the floor. Everything else whitewashed yearly.
7. Separate vent pipe from the pit to several feet above the roof. If painted black, and capped properly it will draw air from the pit and make the room much less smelly.
8. Two part door so you have the option of watching the view while sitting.
9. Screened so well that flies have no access to the pit. Lid on the opening done so that it is fly proof, and it cannot be left open.
10. Sited so that it cannot contaminate your drinking water.
11. Far enough from the house to avoid smells, close enough to be convenient.
12. Built in a manner it can easily be moved to another hole, but also is resistant to being blown over in a strong wind.
13. Mouse proof storage for toilet paper.
Insane tides only while the moon tidally locks to the primary.
Also depends on the distance to the primary. While gravity goes down with the square, tidal forces go down with the cube of distance.
He not only got insignificant revenge, he was stupid too.
Any BOFH worth his salt:
1. Would delay any action for long enough to have plausible deny-ability
2, Would make changes slowly and subtly so they were not immediately obvious.
Examples:
A script that modfied all the percentages and total/subtotals in a spread sheet by a random small percentage.
Putting a bunch of errors in the financial report that was going tot he printers to be mailed out to everyone.
Replacing 1 file in a thousand with a copy from the backup system every week.
Once a month randomly overwrite some user's hard drive ideally making it look like a hardware failure.
Reconfiguring a few ports on the network switch to lock out a MAC address at random.
Auto emailing interesting files to the competition.
Scale this, starting with small things 6 weeks after you left, and gradually increasing so that the new IT department is scrambling to keep up.
You also have to do this in a way that is not obvious. You want systems where a machine inside the network establishes the connection, using a well known port so that it doesn't make much of a blip in network traffic analysis. And you actually want multiple such machines to avoid losing your access by a smart sysadmin. And it probably would help if you had a computer in a box that was attached to a spare cable and stuffed into the ceiling space.