Wouldn't it be cheaper to put a single politician in the base of the tower to act as the source of hot air, rather than build 4 square miles of green house?
Remember: 80 to 90 percent of parents believe that their kid is above average. It is heresy to speak out loud the truth that 50% are below average ability.
(By definition of average)
Let me rephrase that subject:
Most current computer programs don't help students learn.
Exceptions:
Having a word processor makes it possible to help kids learn how to write better. As a teacher you scribble your comments on the printout and write "Redo" at the top. (So far there is no computer program as fast as my red pencil.)
Or you sit down and chat with them about what's wrong with the paper, and tell them to redo it. Redo's are easy with a WP.
There are chem and physics simulation programs that provide a reasonable alternative to the lab. While I still think that kids should get their hands dirty in the lab, having the ability to do some of the more expensive experiments in simulation is better than not doing them at all.
***
It should be possible to write a math teaching program that analyses kids mistakes, and doesn't let them game the system. However to do this well, you need a keyboard math entry system that is as fast as a pencil.
In general a good computer program *should* be able to handle 80% of the tutoring and practice sessions, keeping the kids from practicing the incorrect way, and getting the teacher to come over when it ran out of it's own limited repertoire.
Nobody writes that kind of software.
A computer should be able to handle everything that a book can. Kid's books can stay at school, and they can work with a ebook at home. AND the ebooks should be cheap like borscht.
Yes and no.
First the no.
Your farm kid can learn the basics of farming from his dad. But he also needs to learn about integrated pest management, different forms of crop rotation, and how to use a spread sheet well enough to plan and analyze a fertilizer trial.
Your bricklayer's kid will learn how to lay bricks, but if he wants to design and build a field stone fireplace, he needs to understand some basics about foundations, and strength of materials.
The welder's kid is fine for routine shop welding. And then he hits some weird allow and wonders why it doesn't work.
Here in Alberta we have an apprenticeship program for many professions. It's run out of our "Institutes of Technology" They are 'diploma' programs -- considered a cut below a 'college' degree. But increasingly college grads are going back to school an NAIT and SAIT to get a ticket that will allow them to make a living.
Now the yes.
Kids respond to enthusiasm. If I go in to my math class and am pumped up about some abstract concept, I can infect a good half my class with that enthusiasm. Teaching "The Tempest" is easy -- once you recognize that there has been a vocabulary shift in the last 400 years. It's not a "read Act II for tomorrow." It's acting it out, and discussing the meaning, and paraphrasing big chunks of it into modern English. I had a teacher who took us through one of the comedies, "As you Like it" and then we went to the matinee that the Winnipeg theatre was putting on. Our class knew all the dirty jokes, and understood the situtation behind the funny looks, and the expressions. We howled. ROTFL. Of course none of the other classes there had our good teacher.
Every kid should come out of school:
* Being able to compute and have good enough numeracy to recognize when the minus sign was hit instead of the plus sign in using a calculator.
* Being able to read, summarize and understand any reasonable text.
* Be able to express his thoughts and feelings in writing.
* Know enough history to realize that our present situations have parallels in the past -- but are still unique.
* Understand how society functions, and how to effect social changes using the mail box, the soap box and the ballot box.
* Have a few basic skills: How to buy food, cook, keep house, keep track of money, sew a button, fix a leak, plant a garden, and know how to find out about the ones s/he doesn't.
I question the "All the power we need can come from solar and wind."
According to wiki
http://en.wikipedia.org/wiki/World_energy_resources_and_consumption
we use an average of 1.5 Terawatts.
A square kilometer has a million square meters.
Sunlight runs about 250 W/m2 when averaged over 24 hours.
Assume 25% efficient solar cells, and a 50% placement rate. (Leaving aisles for service, etc)
So our million m2 is now 125,000 m2
Our power/km2 is 31 MW
So to get 1.5 TW requires just under half a million square kilometers.
Hmm.
Wind turbines currently come in 5 MW chunks, and have an average utilization of about 30%. So about 700 turbines per GW. 700,000 turbines per TW or about a million turbines.
This totally ignores the storage problem.
High altitude wind power may work, but we have no practice with 25 kilometer long kite strings.
I don't think there is any one-size-fits-all solution.
We have to explore all energy technologies, and get a lot better an energy use.
My limited understanding is that waste is a political + NIMBY problem, not a technology problem.
Set the rods in a cooling pond for 5-10 years for the short lifespan radiation to exhaust itself,
Melt the rest into glass, and cast as glass bricks.
Dip coat the outside of the bricks in another quarter inch of non-radioactive glass to prevent surface leaching.
Park in a dry rock formation.
OR
Drop the glass bricks in a deep subduction zone.
We handle chemical materials that are far more toxic than radioactive materials on a routine basis. AND they don't have the courtesy to degrade. They are poisonous forever.
If everyone has the same desktop, keeps their files on a central server, and you can rebuild the desktop with a single command, then the correct answer is likely "one"
Reality strikes.
Not everyone will have the same desktop. Only a very few copies of AccPac accounting need to be installed.
Someone will need photoshop. Or InDesign. Someone will need Autocad.
If you can take the time to automate every install. Every computer is associated with a profile to install the right stuff. Then life is easy.
I've had 4 sysadmin jobs.
Space Physics -- 10 NeXTs, 2 RS6000, Stardent Titan, Dec Ultrix, Myrias SPS 3, couple PC's. Easy.
Math Dept -- 3 versions of SGI IRIX , 2 versions of Solaris, 2 dists of Linux, HP HPUX, RS 6000, 5 versions of Microsoft. 250 machines total. That kept me hopping. About half my time was spent on the 30 machiens that ran some form of windows.
YottaYotta. All linux. Had time to make automated reinstall systems for the developers to crash regularly. Easy.
High School. FreeBSD servers, windows 2000 clients. 60 machines total. Windows took up 90% of my time.
Generally:
Managing a hoard of anything isn't much harder than managing three of them.
If you have an OS where applications can be completely divorced from the OS (*ix) and can be run from a network drive, then almost all of your individual workstation customization is trivial. (symlink to the application for light weight stuff -- rsync nightly for heavy weight apps)
If you have an OS that is smart enough to recognize hardware changes, and not bork on you when a disk image is moved to a slightly different hardware platform (Linux, *BSD) then you are golden. You don't have to manage the combinatoric explosion of N different motherboards combined with M different desktops.
If you have a setup where users cannot install executable software, the crisis count goes way down.
In short: Managing a hoard of machines is easy if you don't have windows.
And that's why I said LOGO. It's visual enough to engage. The ability to interpret commands then automate them makes a nice bridge to the concept of abstraction.
My claim is that without a direct visual bridge is that programming is too abstract -- too much disconnect between what you write and when you get it -- for most 12 year olds.
Yes LOGO should be introduced to young kids. I've read it, I've seen it used. It's wonderful. In many ways it works better with young kids than it does with middle school.
Yes I took the same classes in getting my Ed degree that everyone else does. But after having dealt with junior high kids for twenty some years I would NOT want to teach them programming with a non-direct visual model.
Piaget is held up as the model. But real kids have a lot more variability. There is huge variation in both the timing and the sequence of the Piaget stages. In addition education is so ridden with fads that the research is suspect. Way too many 'researchers' get the results that a in fashion. If they don't they have a hard time getting published.
I found at one school I worked with that just getting the grade 10 kids to write a simple html page was pulling teeth. Sure, 50% of the kids could do it, but most of them didn't want to. The rest were incredibly frustrated by the need for matching tags, for proper syntax.
Mind you -- this was a school for the guys that failed in the public system -- over 3/4 of these kids didn't want to be in school at all.
Writing perl or C code to teach programming to a 12 year old?
With the 12 year old's I've worked with, I think I'd prefer to try teaching my border collie.
They don't think abstractly yet. There is a reason that Algebra isn't usually taught until grade 9.
All the success stories I've heard with teaching programming to 'little people' have been using an interactive graphical language such as Logo. Initially you use it as a keyboard controlled etch-a-sketch. When you show that you can define a subroute 'Box' that draws a square, then a subroutine 'window' that draws a 4 pane mullioned window either the eyes light up, or you get a 'so what, I want to play hockey' look.
Sure there are some kids who would dig programming right away at age 12 -- I suspect that a large fraction of/. readers are from that group. But most kids aren't.
If you don't work with something like LOGO, wait a couple years. Mean while, show him examples, and give him stories where the hero is a programmer. (Short list that...)
THIS might be a reasonable computer to use in the back country especially if you could use a stylus instead/in addition to a warm finger. (A limitation of the iPhone is that a gloved finger doesn't work. And an ungloved cold finger doesn't work.)
Consider:
* Every field guide with enough pix to give a chance at identification. Alternate keying systems. Seine Keys. (Too many keying systems depend on characteristics that aren't present all the time -- E.g. flowers. A seine keying system allows you to fill in what is known. (Leaves alternate/opposite -- margins entire or serate...) After each one the number of remaining plants are shown.
* Bird Shazaam. Point the microphone at a bird, wait for it to cheep, chirp, or squawk. Instant Identity. More significantly: Connect it to a high gain microphone, let it sit for an hour, and it does a survey: (27 different chickadees making terratory calls. Two juvenile whiskey jacks learning to sing, and the first robin in spring.)
* Take a picture. GPS location and compass orientation of picture is added to the metadata. Then you can make notes on the picture.
* GPS mapping that shows enough map to be useful.
Thing is: it's already too big to fit in any pocket I have. Make it 9" x 12" so that it's the size of a standard clip board.
Bigot:== Someone who's mind is made up, and is impervious to discussion.
Skeptic:== someone willing to examine new information.
A useful touchstone is to ask "What would you consider good evidence for X"
Thus, if I ask a global warming doubter what he would consider good evidence for global warming, if he replies "It's all a conspiracy" then there is little point in talking to him. On the other hand if says, "A good climate model should match with good accuracy the past -- including things like the Little Ice Age."
Or "I'd like to see the data sets, and on the processing see explanations why they adjusted the data the way they did"
Or "Good models should show large regional effects such as the El Nino / La Nina oscillation, the periodic drought pattern in the Palliser triangle on the Canadian prairies. While I don't expect the data to match year by year, I do expect it to show the patterns."
This is a person that may be open to discussion.
E.g. If you ask a creationist what evidence he would find convincing for evolution, and he answers "I want to see a whole new species with the intermediate steps produced by evolution" I then ask him how different they need to be: Are horses and donkeys sufficiently different? They can cross breed, but the progeny are sterile. How about Dachshunds and St. Bernards? They can't breed without assistance"
If I get answers to these without much foaming at the mouth, then perhaps we can have a discussion. If he asks me to breed dogs starting with cats, I'll have to ask him for more time.
e.g. I have friends who believe in Astrology.
I proposed the following experiment:
Pull a bunch of hospital records for a given hospital. Pick 12 days -- one from each sign. Pick 20 kids from each day.
20 years later. Get a bio from each individual describing the influences on their life, or alternately, provide a questionaire.
Now ask the astrologer to match up the birthdates to the people based on their life story or questionaire results.
IF astrology has as much insight as astrologers claim, they should be able to do a pretty good job of matching. I'd be convinced at levels better than 80% assuming that the data set hadn't been compromised.
(Check -- do it backward as a double blind. Give out the questionaire, let the astrologers make the prediction, THEN ask them for their birthday.)
(Alternate: The astrologer interviews the person, but certain questions aren't allowed -- e.g. what is your birthdate -- were you the oldest, youngest or in the middle of your grade regarding relative age? When the astrologer is fairly certain, they record the sign, and move on to the next interview.)
If he is uncooperative then everyone calls in sick for a week.
Alternate strategies:
Grab a conference room large enough to seat all the programmers and have a head hunter from come in to talk about the best way to transition. Make sure your boss knows about the purpose of the 'development meeting' after the meeting has started.
When I worked at YottaYotta one of the developers set up a spare box as music server. Bosses didn't care. They were mostly willing to do anything to keep the developers happy.
While I was there my supervisor said, "XXX (the company CEO) thinks we should all be working on the same computers. What would the developers do if we said they couldn't run linux desktops, but used winsooze instead?
I told him that it was easy: * Half of them would quit. * The company would have to pay several thousand dollars per seat to license development software equivalent to what they were using under linux for no fee. * Development would essentially come to a halt for weeks to months while people learned new tools
Fortunately with this, my supervisor accepted the advice, passed it up the chain, and we heard no more about this.
I'm the opposite. I can write and code with instrumental music on, with familiar songs on. I can't read, write or code with spoken words. It's as if I only have a single channel word processor in my head, and the ears have a higher priority interrupt.
If I'm struggling with a hunk of opaque perl, the last thing I want is someone talking to me. If I break state at this time, it can take 5-20 minutes to get back to work.
At my last computer job I had a sign on my door: "If this door is closed there better be smoke coming out of the server room. Do NOT interrupt. As the entire IT department, I was paid to be bothered, so I didn't close the door very often.
But what you don't get is a mature forest. Plantations will have much lower diversity.
Even in our limited northern forests in Canada compare:
* Plantation poplar will often be a single clone through the entire plantation. Natural poplar clones are typically only a few hundred feet across.
* Best results in plantation poplar are to keep bare earth underneath. Serious erosion issues. Second best is a shallow, low water usage grass such as sheep's fescue. A natural poplar grove here has an understory with wild rose, dogwood, hazelnut, chokecherry, lungwort, violets, bedstraw, thistles, nettles, bunchberry, false soloman's seal, wild clematis, harebell, dandelion...
* Plantations are done at uniform density. Wild poplar groves have edges and holes in them. Even more diversity there.
* Plantations are even aged. Most natural groves will have a distribution of ages. (Not true for fire succession species such as jack pine, and lodgepole pine.)
* Poplar plantations here are typically cropped in 8-12 years. The entire grove is clearcut. Natural poplar groves have a life span of 50-120 years, with new trees starting at the edges, and spruce taking over in the interior.
This does not mean that plantations are evil. But they are not as diverse as natural forests. They aren't planned that way, and they don't have enough time.
You won't find as many kinds of birds, kinds of mammals, kinds of bugs.
Most low diversity ecologies are unstable: Populations fluctuate wildly. Even natural systems here have low enough diversity that some animals have wild swings in population. E.g. rabbits have a 7-9 year cycle. Predators that have strong preference for rabbit show a similar cycle. This year we have a huge spike in the meadow vole population locally. No idea why.
Overall I would rather see a mixed strategy of some uncut natural forest and intense plantation rather than cutting down all the forest on a longer cycle.
I suspect that using wood for paper is a passing fad. GM versions of crops like switch grass and wheat grass may have more potential for producing cellulose. Trees have way too much lignum in their make up.
So I downloaded the paper. It gives as the address:
patent_quality_comments@uspto.gov
So I fired up, sent them 6 ideas
A second or so later I get the dreaded:
Delivery to the following recipient failed permanently.
Sigh.
If any others care, this is what I suggested:
0. Establish a web site forum where this whole idea can be discussed.
1. Public Search for prior art.
When an application comes it, the claims and abstract are published and are publicly available. Further, there is an automated email system that users can set up with keyword searches using Google syntax to get weekly notifications in areas of interest.
Each patent is the start of a forum thread where people can point out prior art. An apprentice patent examiner acts as a moderator of the forum. The patent applicant also is a participant.
The moderator awards 'karma' points to participants based on the clarity and insight of their commentary. Writers with higher karma are given greater attention.
In some cases the questions brought up on the forum will cause the applicant to modify the language of his patent.
2. All patents must be backed up with working examples. It is no longer sufficient to patent an idea. However an application can start before the completion of the working example. Initially this is for a year, but perhaps should be extendible on payment of a reasonable fee.
Because of the public review process, however, a person who takes too long after making application may find that someone else beats them to it.
3. There are occasions when there is an idea whose time has come. There may be multiple submissions for what amounts to the same thing. In this case the patent office has the choice of:
* Granting the patent to the best implementation of the idea. * Granting the patent to the first completed application. * Granting a joint patent in multiple names. * Grant a patent to one individual, encumbered with a fee sharing agreement to the other.
This may result in situations where A proposes a patent. During the review period B sees it, and writes an application for the same idea, but demonstrates a better application of that idea. B gets the patent.
4. Elimination of overly broad claims. If a patent claims results "using a catalyst from the Palladium group of elements" he has to demonstrate that several different elements from that group are effective as catalysts. It's not sufficient just to show that Palladium works. If somone clones a sheep, they can't claim that the method works for all mammals. Once they have demonstrated that the method works for several different mammals, they can broaden their claims.
5. Clarity of application. The application should be clear enough that a third party equipped only with capability in the prior art can duplicate the results.
6. Pre-licensing Someone who wishes to license a prospective patent may enter into an agreement with the applicant. License fees are held in excrow until the application is accepted. In the event that an application is turned down, license fees revert. This needs to be carefully thought out to deal with multiple overlapping applications.
Our schools don't really teach science. They give lip service to the 'scientific method' and it's application in experimental sciences. (It's really hard to do experiments in geology, stellar evolution, cosmology, meteorology.)
Most of the science is presented as tablets presented from on high. (This is how you do stoichiometry problems...This is Charles law, Boyles law...)
We need to teach people:
1. Science is a process. Not a result.
2. If we teach science well, we teach people to doubt, to question, to be critical in their evaluation of facts, and authorities. Coupled with the ability to think numerically -- to estimate problems with 10-20% errors, this is a valuable life skill.
3. Science is a process done by human beings. Scientists have their foibles, biases, prejudices just like the rest of us.
My fantasy science class is to walk in on the first day, and talk about "Earth, air, fire, and water" and teach it as fact.
Then the next day, come in and contradict myself.
***
Show a picture of the night sky. Show a sequence. Point out the wanderers. Get the kids to figure out how to keep track. Get them to figure out how to keep track without using paper. Show the weird motions. Teach about the celestial crystal sphere. That the heavens are perfect, and the circle is the only perfect shape, therefore, planets must move in circular motion. Epicycles, Eccentrics, the whole Ptolomeic structure of the universe.
Flash forward to the Kepler and Copernicus crowd.
Have detour to Galileo. Consternation from extra things -- moons of Jupiter. Venus with phases. Mountains on the moon. Sunspots.
***
Throughout the course start out with folk wisdom -- common sense, go through the Aristotealian view, then more modern views. In all cases present each layer as fact. Stress that 'current view is that..." when we get to modern times.
Talk about proof, and the nature of proof.
Talk about "What would you consider to be evidence" for odd ball things such as UFO's re-incarnation, Mu and Atlantis, astrology, sasquatch,
$8/hour is grunt work. Even telephone help desks start at $13 and there isn't that much difference between the U.S. and Canada.
I suspect you won't learn much. I would certainly go and look very carefully at both the working conditions and at the job requirements before starting in.
Intern positions here in technical fields get 2-3 times minimum wage. Engineering interns make enough during the semester to pay for their next two semesters education.
Do you want to be a manager? Some people take to this, some don't. That you are asking the question here makes me think that you either don't want it, or aren't ready for it.
What are the duties? Do you want to do that?
One possible answer: "I don't feel ready for an administrative role at this time. I would like to remain in my present position for 2-3 more years."
I have seen people get sucked into a job they hate because of the pay, and then get trapped there because they think they are addicted tot he pay. I have cleverly negotiated a 66% pay cut over the last 4 years, but now I'm my own boss. I sleep to 7:30. We don't miss the money.
To me 5 day week and on call would be major losses. Be sure you get it nailed down what the compensation is for both being on call, and for responding to on call. If my wife and I are in town on an on-call day, we either have to travel in separate vehicles, or she has to be ready to either wait for me or come in and pick me up when the problem is fixed. It means I can't go hunting,hiking, canoeing -- anything that takes me more than a few minutes further from the city.
I currently live an hours drive from the city of Edmonton. If I were on call:
1. two hours wages at overtime rates for each on-call shift. 2. 1.5 hour response time. 3. Company picks up fee for cell phone. 4. Connectivity to solve IT problems remotely -- they pick up the charge. 5. If have to go in, they pay for my travel time + 25 cents per kilometer. 6. At least 50% of my weekends have no on-call shift. 7. As a bargaining chip: 1 more day paid holiday for each three on call days. 8. I can specify 10 days per year that I cannot be on call: Wife's/kids birthday, aniversary. 9. Certain family time holidays the remuneration is at 3 times standard wage. (Christmas...) 10. If I am in charge of uptime, I get a yearly bonus based on how few times someone had to come in to fix something. And I get a reasonable equipment budget to inplement that failsafe system.
One my peaves is that I cannot use the keyboard with the phone in my ear. I would like the device to come with a separate ear piece/mike, or alternately, a handset about the size of a soup spoon. Either would have a single button for pickup, hangup. As long as I'm in bluetooth range of the netbook, I can use my phone.
I also want voice recognition. Netbook -- Grocery list -- onions, oranges, crackers. End. Netbook -- phone Laura's cell. Retry every 10 minutes until answer. Don't leave message End. Netbook -- Put short order on Apple. Sell 2000 shares at 185 End.
I've attended only one conference, and left two days early.
I've looked at attending other conferences but came up against the following stumbling blocks:
1. Expense: A 5 day conference will typically cost me or my boss $400-500 per day to attend. Add another $100-200/day for hotel and meals, and $500-1000 to fly there and back. As someone who had a $8000 budget for servers, network, internet connectivity and repairs I could not justify this to my boss.
2. Subject matter: There needs to be enough material that isn't just a rehash of an O'Reilly book that interests me. Going to a conference and finding only one session per day that has any interest for me is a loss.
3. Time: Most conferences are non-local. A three day conference takes a full week with transport.
The one advantage of these conferences is they give me leverage: I put together a proposal to my boss one year to go to SANS. He had no budget for it, but at the end of the talk I asked for an open ticket for any computer books I wanted. He ended up agreeing to up to $1000/year in books for my 'professional development'.
The conference mechanism right now has a single tier: SANS is a one shot event in a really big city with several thousand attendees. And for the top couple thousand people in the trade, this is a good thing. However there is a market for a lower level tier of conferences.
In this tier a better model would be to cooperate with local learning institutions to provide venues and provide 'micro schools' These would be local: You drive there in the morning, and go home at night. These would be small: Typical attendance would be 300-600 people. They would be cheap: Attendance at a session would be at a price comparable in per hour costs to going to a movie or university extension course costs.
At such a venue some of the costs would be defrayed by local reps of major companies putting on infomercial sessions on their product. May or may not have an associated trade show.
So, for example, it may make sense after SANS to have SANS road trip. Or take vids of all the sessions, and for a much cheaper fee, you can subscribe to the vids. Or get a printed or PDF copy of all the papers presented at the conference.
We have an annual green conference here in Edmonton every fall. It's primarily for horticulture, turf, arborists and the associated trades. They have a public trade show. They also have 3 streams of sessions over the 2.5 day course. The price for attending the show is $350 per person. This year there was one session that would have held interest for me. I didn't attend. $160/day would have been worth it had there been 3-4 sessions per day I wanted to go to. If I could have bought a ticket to the one session I wanted to go to for $20 I probably would have gone.
So I played with the dictionary. Not bad. I like the multiple definitions, and possible links to chase down.
But what I *really* want is a 'distinctive thesaurus' -- a dictionary that distinguishes between synonyms so that you can get closer to the perfect word.
As an example,
Consider the differences between
Irony
Sarcasm
Sardony (Ok sardonic)
Facetiousness
All of them involve some degree of humour by stating things as they aren't.
If I look up sarcasm on thesaurus.com I get a longer list, yielding words that range from near to distant in their connotations.
Yes creating my own distinctions is possible. So is writing my own definition possible. But trying to define a word from my own experience with a word is hard, and frought with potential pitfalls where my mental model of the word world is defective, so even harder is it to define the differences between closely allied words.
Anybody know of an online thesaurus that distinguishes between synonyms?
My own crack at the above four.
Irony applies to both statements and description. In events has a perverseness to it, poetic justice. In statements it has has less connotation of derision and mocking.
Sardony has a bitter, derisive quality to it. The object of sardony is most often the speaker, less often the world generally. Self-deprecating on steroids.
Sarcasm is a contrary statement intended to hurt someone else, to express contempt.
Facetiousness is similar to sarcasm, but humour is it's main goal. There is no intent to hurt.
Slashdot Mirror
Wouldn't it be cheaper to put a single politician in the base of the tower to act as the source of hot air, rather than build 4 square miles of green house?
Remember: 80 to 90 percent of parents believe that their kid is above average. It is heresy to speak out loud the truth that 50% are below average ability.
(By definition of average)
Let me rephrase that subject: Most current computer programs don't help students learn. Exceptions: Having a word processor makes it possible to help kids learn how to write better. As a teacher you scribble your comments on the printout and write "Redo" at the top. (So far there is no computer program as fast as my red pencil.) Or you sit down and chat with them about what's wrong with the paper, and tell them to redo it. Redo's are easy with a WP. There are chem and physics simulation programs that provide a reasonable alternative to the lab. While I still think that kids should get their hands dirty in the lab, having the ability to do some of the more expensive experiments in simulation is better than not doing them at all. *** It should be possible to write a math teaching program that analyses kids mistakes, and doesn't let them game the system. However to do this well, you need a keyboard math entry system that is as fast as a pencil. In general a good computer program *should* be able to handle 80% of the tutoring and practice sessions, keeping the kids from practicing the incorrect way, and getting the teacher to come over when it ran out of it's own limited repertoire. Nobody writes that kind of software. A computer should be able to handle everything that a book can. Kid's books can stay at school, and they can work with a ebook at home. AND the ebooks should be cheap like borscht.
Yes and no. First the no. Your farm kid can learn the basics of farming from his dad. But he also needs to learn about integrated pest management, different forms of crop rotation, and how to use a spread sheet well enough to plan and analyze a fertilizer trial. Your bricklayer's kid will learn how to lay bricks, but if he wants to design and build a field stone fireplace, he needs to understand some basics about foundations, and strength of materials. The welder's kid is fine for routine shop welding. And then he hits some weird allow and wonders why it doesn't work. Here in Alberta we have an apprenticeship program for many professions. It's run out of our "Institutes of Technology" They are 'diploma' programs -- considered a cut below a 'college' degree. But increasingly college grads are going back to school an NAIT and SAIT to get a ticket that will allow them to make a living. Now the yes. Kids respond to enthusiasm. If I go in to my math class and am pumped up about some abstract concept, I can infect a good half my class with that enthusiasm. Teaching "The Tempest" is easy -- once you recognize that there has been a vocabulary shift in the last 400 years. It's not a "read Act II for tomorrow." It's acting it out, and discussing the meaning, and paraphrasing big chunks of it into modern English. I had a teacher who took us through one of the comedies, "As you Like it" and then we went to the matinee that the Winnipeg theatre was putting on. Our class knew all the dirty jokes, and understood the situtation behind the funny looks, and the expressions. We howled. ROTFL. Of course none of the other classes there had our good teacher. Every kid should come out of school: * Being able to compute and have good enough numeracy to recognize when the minus sign was hit instead of the plus sign in using a calculator. * Being able to read, summarize and understand any reasonable text. * Be able to express his thoughts and feelings in writing. * Know enough history to realize that our present situations have parallels in the past -- but are still unique. * Understand how society functions, and how to effect social changes using the mail box, the soap box and the ballot box. * Have a few basic skills: How to buy food, cook, keep house, keep track of money, sew a button, fix a leak, plant a garden, and know how to find out about the ones s/he doesn't.
... was inflicted on us by a foreign power, it would be considered an act of war!
I question the "All the power we need can come from solar and wind." According to wiki http://en.wikipedia.org/wiki/World_energy_resources_and_consumption we use an average of 1.5 Terawatts. A square kilometer has a million square meters. Sunlight runs about 250 W/m2 when averaged over 24 hours. Assume 25% efficient solar cells, and a 50% placement rate. (Leaving aisles for service, etc) So our million m2 is now 125,000 m2 Our power/km2 is 31 MW So to get 1.5 TW requires just under half a million square kilometers. Hmm. Wind turbines currently come in 5 MW chunks, and have an average utilization of about 30%. So about 700 turbines per GW. 700,000 turbines per TW or about a million turbines. This totally ignores the storage problem. High altitude wind power may work, but we have no practice with 25 kilometer long kite strings. I don't think there is any one-size-fits-all solution. We have to explore all energy technologies, and get a lot better an energy use.
My limited understanding is that waste is a political + NIMBY problem, not a technology problem. Set the rods in a cooling pond for 5-10 years for the short lifespan radiation to exhaust itself, Melt the rest into glass, and cast as glass bricks. Dip coat the outside of the bricks in another quarter inch of non-radioactive glass to prevent surface leaching. Park in a dry rock formation. OR Drop the glass bricks in a deep subduction zone. We handle chemical materials that are far more toxic than radioactive materials on a routine basis. AND they don't have the courtesy to degrade. They are poisonous forever.
If everyone has the same desktop, keeps their files on a central server, and you can rebuild the desktop with a single command, then the correct answer is likely "one"
Reality strikes.
Not everyone will have the same desktop. Only a very few copies of AccPac accounting need to be installed.
Someone will need photoshop. Or InDesign. Someone will need Autocad.
If you can take the time to automate every install. Every computer is associated with a profile to install the right stuff. Then life is easy.
I've had 4 sysadmin jobs.
Space Physics -- 10 NeXTs, 2 RS6000, Stardent Titan, Dec Ultrix, Myrias SPS 3, couple PC's. Easy.
Math Dept -- 3 versions of SGI IRIX , 2 versions of Solaris, 2 dists of Linux, HP HPUX, RS 6000, 5 versions of Microsoft. 250 machines total. That kept me hopping. About half my time was spent on the 30 machiens that ran some form of windows.
YottaYotta. All linux. Had time to make automated reinstall systems for the developers to crash regularly. Easy.
High School. FreeBSD servers, windows 2000 clients. 60 machines total. Windows took up 90% of my time.
Generally:
Managing a hoard of anything isn't much harder than managing three of them.
If you have an OS where applications can be completely divorced from the OS (*ix) and can be run from a network drive, then almost all of your individual workstation customization is trivial. (symlink to the application for light weight stuff -- rsync nightly for heavy weight apps)
If you have an OS that is smart enough to recognize hardware changes, and not bork on you when a disk image is moved to a slightly different hardware platform (Linux, *BSD) then you are golden. You don't have to manage the combinatoric explosion of N different motherboards combined with M different desktops.
If you have a setup where users cannot install executable software, the crisis count goes way down.
In short: Managing a hoard of machines is easy if you don't have windows.
And that's why I said LOGO. It's visual enough to engage. The ability to interpret commands then automate them makes a nice bridge to the concept of abstraction.
My claim is that without a direct visual bridge is that programming is too abstract -- too much disconnect between what you write and when you get it -- for most 12 year olds.
Yes LOGO should be introduced to young kids. I've read it, I've seen it used. It's wonderful. In many ways it works better with young kids than it does with middle school.
Yes I took the same classes in getting my Ed degree that everyone else does. But after having dealt with junior high kids for twenty some years I would NOT want to teach them programming with a non-direct visual model.
Piaget is held up as the model. But real kids have a lot more variability. There is huge variation in both the timing and the sequence of the Piaget stages. In addition education is so ridden with fads that the research is suspect. Way too many 'researchers' get the results that a in fashion. If they don't they have a hard time getting published.
I found at one school I worked with that just getting the grade 10 kids to write a simple html page was pulling teeth. Sure, 50% of the kids could do it, but most of them didn't want to. The rest were incredibly frustrated by the need for matching tags, for proper syntax.
Mind you -- this was a school for the guys that failed in the public system -- over 3/4 of these kids didn't want to be in school at all.
Writing perl or C code to teach programming to a 12 year old?
With the 12 year old's I've worked with, I think I'd prefer to try teaching my border collie.
They don't think abstractly yet. There is a reason that Algebra isn't usually taught until grade 9.
All the success stories I've heard with teaching programming to 'little people' have been using an interactive graphical language such as Logo. Initially you use it as a keyboard controlled etch-a-sketch. When you show that you can define a subroute 'Box' that draws a square, then a subroutine 'window' that draws a 4 pane mullioned window either the eyes light up, or you get a 'so what, I want to play hockey' look.
Sure there are some kids who would dig programming right away at age 12 -- I suspect that a large fraction of /. readers are from that group. But most kids aren't.
If you don't work with something like LOGO, wait a couple years. Mean while, show him examples, and give him stories where the hero is a programmer. (Short list that...)
THIS might be a reasonable computer to use in the back country especially if you could use a stylus instead/in addition to a warm finger. (A limitation of the iPhone is that a gloved finger doesn't work. And an ungloved cold finger doesn't work.)
Consider:
* Every field guide with enough pix to give a chance at identification. Alternate keying systems. Seine Keys. (Too many keying systems depend on characteristics that aren't present all the time -- E.g. flowers. A seine keying system allows you to fill in what is known. (Leaves alternate/opposite -- margins entire or serate...) After each one the number of remaining plants are shown.
* Bird Shazaam. Point the microphone at a bird, wait for it to cheep, chirp, or squawk. Instant Identity. More significantly: Connect it to a high gain microphone, let it sit for an hour, and it does a survey: (27 different chickadees making terratory calls. Two juvenile whiskey jacks learning to sing, and the first robin in spring.)
* Take a picture. GPS location and compass orientation of picture is added to the metadata. Then you can make notes on the picture.
* GPS mapping that shows enough map to be useful.
Thing is: it's already too big to fit in any pocket I have. Make it 9" x 12" so that it's the size of a standard clip board.
I've been a subscriber to several email alerts from the Economist for three years now. I usually read about 1/3 to 1/2 the articles referenced.
In general I have found that the Economist:
1. Reports stories in a very shallow manner. They are short, lack details, lack references.
2. Have a consistent bias..
3. Don't do a good job of separating opinion from fact.
4. Often don't talk about anything related to economy.
Bigot :== Someone who's mind is made up, and is impervious to discussion.
Skeptic :== someone willing to examine new information.
A useful touchstone is to ask "What would you consider good evidence for X"
Thus, if I ask a global warming doubter what he would consider good evidence for global warming, if he replies "It's all a conspiracy" then there is little point in talking to him. On the other hand if says, "A good climate model should match with good accuracy the past -- including things like the Little Ice Age."
Or "I'd like to see the data sets, and on the processing see explanations why they adjusted the data the way they did"
Or "Good models should show large regional effects such as the El Nino / La Nina oscillation, the periodic drought pattern in the Palliser triangle on the Canadian prairies. While I don't expect the data to match year by year, I do expect it to show the patterns."
This is a person that may be open to discussion.
E.g.
If you ask a creationist what evidence he would find convincing for evolution, and he answers "I want to see a whole new species with the intermediate steps produced by evolution" I then ask him how different they need to be: Are horses and donkeys sufficiently different? They can cross breed, but the progeny are sterile. How about Dachshunds and St. Bernards? They can't breed without assistance"
If I get answers to these without much foaming at the mouth, then perhaps we can have a discussion. If he asks me to breed dogs starting with cats, I'll have to ask him for more time.
e.g. I have friends who believe in Astrology.
I proposed the following experiment:
Pull a bunch of hospital records for a given hospital. Pick 12 days -- one from each sign. Pick 20 kids from each day.
20 years later. Get a bio from each individual describing the influences on their life, or alternately, provide a questionaire.
Now ask the astrologer to match up the birthdates to the people based on their life story or questionaire results.
IF astrology has as much insight as astrologers claim, they should be able to do a pretty good job of matching. I'd be convinced at levels better than 80% assuming that the data set hadn't been compromised.
(Check -- do it backward as a double blind. Give out the questionaire, let the astrologers make the prediction, THEN ask them for their birthday.)
(Alternate: The astrologer interviews the person, but certain questions aren't allowed -- e.g. what is your birthdate -- were you the oldest, youngest or in the middle of your grade regarding relative age? When the astrologer is fairly certain, they record the sign, and move on to the next interview.)
The manager's request is unreasonable.
Ask him to demonstrate that it distracts.
If he is uncooperative then everyone calls in sick for a week.
Alternate strategies:
Grab a conference room large enough to seat all the programmers and have a head hunter from come in to talk about the best way to transition. Make sure your boss knows about the purpose of the 'development meeting' after the meeting has started.
When I worked at YottaYotta one of the developers set up a spare box as music server. Bosses didn't care. They were mostly willing to do anything to keep the developers happy.
While I was there my supervisor said, "XXX (the company CEO) thinks we should all be working on the same computers. What would the developers do if we said they couldn't run linux desktops, but used winsooze instead?
I told him that it was easy:
* Half of them would quit.
* The company would have to pay several thousand dollars per seat to license development software equivalent to what they were using under linux for no fee.
* Development would essentially come to a halt for weeks to months while people learned new tools
Fortunately with this, my supervisor accepted the advice, passed it up the chain, and we heard no more about this.
I'm the opposite. I can write and code with instrumental music on, with familiar songs on. I can't read, write or code with spoken words. It's as if I only have a single channel word processor in my head, and the ears have a higher priority interrupt.
This can be perceived as a good thing.
If I'm struggling with a hunk of opaque perl, the last thing I want is someone talking to me. If I break state at this time, it can take 5-20 minutes to get back to work.
At my last computer job I had a sign on my door: "If this door is closed there better be smoke coming out of the server room. Do NOT interrupt. As the entire IT department, I was paid to be bothered, so I didn't close the door very often.
But what you don't get is a mature forest. Plantations will have much lower diversity.
Even in our limited northern forests in Canada compare:
* Plantation poplar will often be a single clone through the entire plantation. Natural poplar clones are typically only a few hundred feet across.
* Best results in plantation poplar are to keep bare earth underneath. Serious erosion issues. Second best is a shallow, low water usage grass such as sheep's fescue. A natural poplar grove here has an understory with wild rose, dogwood, hazelnut, chokecherry, lungwort, violets, bedstraw, thistles, nettles, bunchberry, false soloman's seal, wild clematis, harebell, dandelion...
* Plantations are done at uniform density. Wild poplar groves have edges and holes in them. Even more diversity there.
* Plantations are even aged. Most natural groves will have a distribution of ages. (Not true for fire succession species such as jack pine, and lodgepole pine.)
* Poplar plantations here are typically cropped in 8-12 years. The entire grove is clearcut. Natural poplar groves have a life span of 50-120 years, with new trees starting at the edges, and spruce taking over in the interior.
This does not mean that plantations are evil. But they are not as diverse as natural forests. They aren't planned that way, and they don't have enough time.
You won't find as many kinds of birds, kinds of mammals, kinds of bugs.
Most low diversity ecologies are unstable: Populations fluctuate wildly. Even natural systems here have low enough diversity that some animals have wild swings in population. E.g. rabbits have a 7-9 year cycle. Predators that have strong preference for rabbit show a similar cycle. This year we have a huge spike in the meadow vole population locally. No idea why.
Overall I would rather see a mixed strategy of some uncut natural forest and intense plantation rather than cutting down all the forest on a longer cycle.
I suspect that using wood for paper is a passing fad. GM versions of crops like switch grass and wheat grass may have more potential for producing cellulose. Trees have way too much lignum in their make up.
So I downloaded the paper. It gives as the address:
patent_quality_comments@uspto.gov
So I fired up, sent them 6 ideas
A second or so later I get the dreaded:
Delivery to the following recipient failed permanently.
Sigh.
If any others care, this is what I suggested:
0. Establish a web site forum where this whole idea can be discussed.
1. Public Search for prior art.
When an application comes it, the claims and abstract are published and are publicly available. Further, there is an automated email system that users can set up with keyword searches using Google syntax to get weekly notifications in areas of interest.
Each patent is the start of a forum thread where people can point out prior art. An apprentice patent examiner acts as a moderator of the forum. The patent applicant also is a participant.
The moderator awards 'karma' points to participants based on the clarity and insight of their commentary. Writers with higher karma are given greater attention.
In some cases the questions brought up on the forum will cause the applicant to modify the language of his patent.
2. All patents must be backed up with working examples. It is no longer sufficient to patent an idea. However an application can start before the completion of the working example. Initially this is for a year, but perhaps should be extendible on payment of a reasonable fee.
Because of the public review process, however, a person who takes too long after making application may find that someone else beats them to it.
3. There are occasions when there is an idea whose time has come. There may be multiple submissions for what amounts to the same thing. In this case the patent office has the choice of:
* Granting the patent to the best implementation of the idea.
* Granting the patent to the first completed application.
* Granting a joint patent in multiple names.
* Grant a patent to one individual, encumbered with a fee sharing agreement to the other.
This may result in situations where A proposes a patent. During the review period B sees it, and writes an application for the same idea, but demonstrates a better application of that idea. B gets the patent.
4. Elimination of overly broad claims.
If a patent claims results "using a catalyst from the Palladium group of elements" he has to demonstrate that several different elements from that group are effective as catalysts. It's not sufficient just to show that Palladium works. If somone clones a sheep, they can't claim that the method works for all mammals. Once they have demonstrated that the method works for several different mammals, they can broaden their claims.
5. Clarity of application.
The application should be clear enough that a third party equipped only with capability in the prior art can duplicate the results.
6. Pre-licensing
Someone who wishes to license a prospective patent may enter into an agreement with the applicant. License fees are held in excrow until the application is accepted. In the event that an application is turned down, license fees revert. This needs to be carefully thought out to deal with multiple overlapping applications.
Our schools don't really teach science. They give lip service to the 'scientific method' and it's application in experimental sciences. (It's really hard to do experiments in geology, stellar evolution, cosmology, meteorology.)
Most of the science is presented as tablets presented from on high. (This is how you do stoichiometry problems...This is Charles law, Boyles law...)
We need to teach people:
1. Science is a process. Not a result.
2. If we teach science well, we teach people to doubt, to question, to be critical in their evaluation of facts, and authorities. Coupled with the ability to think numerically -- to estimate problems with 10-20% errors, this is a valuable life skill.
3. Science is a process done by human beings. Scientists have their foibles, biases, prejudices just like the rest of us.
My fantasy science class is to walk in on the first day, and talk about "Earth, air, fire, and water" and teach it as fact.
Then the next day, come in and contradict myself.
***
Show a picture of the night sky. Show a sequence. Point out the wanderers. Get the kids to figure out how to keep track. Get them to figure out how to keep track without using paper. Show the weird motions. Teach about the celestial crystal sphere. That the heavens are perfect, and the circle is the only perfect shape, therefore, planets must move in circular motion. Epicycles, Eccentrics, the whole Ptolomeic structure of the universe.
Flash forward to the Kepler and Copernicus crowd.
Have detour to Galileo. Consternation from extra things -- moons of Jupiter. Venus with phases. Mountains on the moon. Sunspots.
***
Throughout the course start out with folk wisdom -- common sense, go through the Aristotealian view, then more modern views. In all cases present each layer as fact. Stress that 'current view is that..." when we get to modern times.
Talk about proof, and the nature of proof.
Talk about "What would you consider to be evidence" for odd ball things such as UFO's re-incarnation, Mu and Atlantis, astrology, sasquatch,
$8/hour is grunt work. Even telephone help desks start at $13 and there isn't that much difference between the U.S. and Canada.
I suspect you won't learn much. I would certainly go and look very carefully at both the working conditions and at the job requirements before starting in.
Intern positions here in technical fields get 2-3 times minimum wage. Engineering interns make enough during the semester to pay for their next two semesters education.
My advice: Keep looking.
Do you want to be a manager? Some people take to this, some don't. That you are asking the question here makes me think that you either don't want it, or aren't ready for it.
What are the duties? Do you want to do that?
One possible answer: "I don't feel ready for an administrative role at this time. I would like to remain in my present position for 2-3 more years."
I have seen people get sucked into a job they hate because of the pay, and then get trapped there because they think they are addicted tot he pay. I have cleverly negotiated a 66% pay cut over the last 4 years, but now I'm my own boss. I sleep to 7:30. We don't miss the money.
To me 5 day week and on call would be major losses. Be sure you get it nailed down what the compensation is for both being on call, and for responding to on call. If my wife and I are in town on an on-call day, we either have to travel in separate vehicles, or she has to be ready to either wait for me or come in and pick me up when the problem is fixed. It means I can't go hunting,hiking, canoeing -- anything that takes me more than a few minutes further from the city.
I currently live an hours drive from the city of Edmonton. If I were on call:
1. two hours wages at overtime rates for each on-call shift.
2. 1.5 hour response time.
3. Company picks up fee for cell phone.
4. Connectivity to solve IT problems remotely -- they pick up the charge.
5. If have to go in, they pay for my travel time + 25 cents per kilometer.
6. At least 50% of my weekends have no on-call shift.
7. As a bargaining chip: 1 more day paid holiday for each three on call days.
8. I can specify 10 days per year that I cannot be on call: Wife's/kids birthday, aniversary.
9. Certain family time holidays the remuneration is at 3 times standard wage. (Christmas...)
10. If I am in charge of uptime, I get a yearly bonus based on how few times someone had to come in to fix something. And I get a reasonable equipment budget to inplement that failsafe system.
One my peaves is that I cannot use the keyboard with the phone in my ear. I would like the device to come with a separate ear piece/mike, or alternately, a handset about the size of a soup spoon. Either would have a single button for pickup, hangup. As long as I'm in bluetooth range of the netbook, I can use my phone.
I also want voice recognition.
Netbook -- Grocery list -- onions, oranges, crackers. End.
Netbook -- phone Laura's cell. Retry every 10 minutes until answer. Don't leave message End.
Netbook -- Put short order on Apple. Sell 2000 shares at 185 End.
... at the airport when you go through Homeland Security's gates with a bag of shit?
I think I'll get it at Home depot instead. $5.49 is worth it to not carry it so far.
I've attended only one conference, and left two days early.
I've looked at attending other conferences but came up against the following stumbling blocks:
1. Expense: A 5 day conference will typically cost me or my boss $400-500 per day to attend. Add another $100-200/day for hotel and meals, and $500-1000 to fly there and back. As someone who had a $8000 budget for servers, network, internet connectivity and repairs I could not justify this to my boss.
2. Subject matter: There needs to be enough material that isn't just a rehash of an O'Reilly book that interests me. Going to a conference and finding only one session per day that has any interest for me is a loss.
3. Time: Most conferences are non-local. A three day conference takes a full week with transport.
The one advantage of these conferences is they give me leverage: I put together a proposal to my boss one year to go to SANS. He had no budget for it, but at the end of the talk I asked for an open ticket for any computer books I wanted. He ended up agreeing to up to $1000/year in books for my 'professional development'.
The conference mechanism right now has a single tier: SANS is a one shot event in a really big city with several thousand attendees. And for the top couple thousand people in the trade, this is a good thing. However there is a market for a lower level tier of conferences.
In this tier a better model would be to cooperate with local learning institutions to provide venues and provide 'micro schools' These would be local: You drive there in the morning, and go home at night. These would be small: Typical attendance would be 300-600 people. They would be cheap: Attendance at a session would be at a price comparable in per hour costs to going to a movie or university extension course costs.
At such a venue some of the costs would be defrayed by local reps of major companies putting on infomercial sessions on their product. May or may not have an associated trade show.
So, for example, it may make sense after SANS to have SANS road trip. Or take vids of all the sessions, and for a much cheaper fee, you can subscribe to the vids. Or get a printed or PDF copy of all the papers presented at the conference.
We have an annual green conference here in Edmonton every fall. It's primarily for horticulture, turf, arborists and the associated trades. They have a public trade show. They also have 3 streams of sessions over the 2.5 day course. The price for attending the show is $350 per person. This year there was one session that would have held interest for me. I didn't attend. $160/day would have been worth it had there been 3-4 sessions per day I wanted to go to. If I could have bought a ticket to the one session I wanted to go to for $20 I probably would have gone.
So I played with the dictionary. Not bad. I like the multiple definitions, and possible links to chase down.
But what I *really* want is a 'distinctive thesaurus' -- a dictionary that distinguishes between synonyms so that you can get closer to the perfect word.
As an example,
Consider the differences between
Irony
Sarcasm
Sardony (Ok sardonic)
Facetiousness
All of them involve some degree of humour by stating things as they aren't.
If I look up sarcasm on thesaurus.com I get a longer list, yielding words that range from near to distant in their connotations.
acrimony, aspersion, banter, bitterness, burlesque, causticness, censure, comeback, contempt, corrosiveness, criticism, cut*, cynicism, derision, dig*, disparagement, flouting, invective, irony, lampooning, mockery, mordancy, put-down, raillery, rancor, ridicule, satire, scoffing, scorn, sharpness, sneering, superciliousness, wisecrack.
Yes creating my own distinctions is possible. So is writing my own definition possible. But trying to define a word from my own experience with a word is hard, and frought with potential pitfalls where my mental model of the word world is defective, so even harder is it to define the differences between closely allied words.
Anybody know of an online thesaurus that distinguishes between synonyms?
My own crack at the above four.
Irony applies to both statements and description. In events has a perverseness to it, poetic justice. In statements it has has less connotation of derision and mocking.
Sardony has a bitter, derisive quality to it. The object of sardony is most often the speaker, less often the world generally. Self-deprecating on steroids.
Sarcasm is a contrary statement intended to hurt someone else, to express contempt.
Facetiousness is similar to sarcasm, but humour is it's main goal. There is no intent to hurt.