My wife and I have just discovered the joys of digital audio on my/her itouch/iphone. It has not taken us very long to fill a 250 GB drive. People who keep movies must have a real storage problem.
The industry moved from 5" media to 3.5" media, and now 2" hard disks are coming in for more than just laptops. This has mostly come about for the sake of faster seeks, yes?
Given the disparate needs, perhaps the ordinary user needs different kinds of storage. A desktop that is running streaming video doesn't seek much. Do you care if the heads have a 20 ms average access time?
So you guys who have more knowledge about hardware than I do:
How cost effective would it be to create a 12 platter 5" form factor drive that spun at 3600 rpm, but had the same areal bit density as current 2" drives?
***
Part 2. At one point I remember 'tiered' storage. Back when disks were very expensive compared to tape. You had a set of spinning disks, and a silo full of tapes. When a file hadn't been accessed for N days, it was migrated to tape. When it was called for again, you had a long pause while the silo found the tape and loaded the file.
Are we going to have a window where this makes good sense, substituting SSD for disk, and HD for tape?
As a geezer currently in the job hunt I can identify with several comments.
I use interviews to learn about my possible employer.
* If the job doesn't sound interesting I won't bother. * To be interesting, it has to have elements I've not done before. * Is this a 'tie' shop? Excuse me, I think I left the roast in the oven.
If I were on the hiring side of the desk, I might ask, "What was the most recent language you learned, and how did you go about learning it?" Substitute any concept you wish for language.
Scenarios are good. I like the one response, "And then what..."
What do you know about Foo?, where foo is some aspect about the business the company is in. I interviewed with the Canadian Forest Service as a "Carbon Cycling Technician" I got the job because I had some experience programming, could obviously learn on my own, and also had spent the week before the interview reading up on the carbon cycle in forests. The interview asked me all sorts of questions about indicator species, eutrophication of lakes, reaction rates as a function of particle size. I was also asked if I had any experience snowshoeing, what my tolerance for mosquitoes, and was I afraid of heights. (Turns out that part of the job was taking samples at various times of year and climbing 40 meter towers calibrating CO2 sensors.)
It really helps if the programmers know enough about the business to make intelligent decisions.
My brother started out majoring in Math. Found that it was too abstract, and after a semester changed his major to Physics. Physics wasn't 'real' enough, so he switched to pre-med. Pre-med was full of very intense people who couldn't stand to get less than 95%. He switched to biology. One summer he worked in a lab culturing bacteria, so he took his masters in microbiology, and his doctoratal thesis was on some aspect of yeast.
I started in physics and wildlife management as dual majors and was able to keep it up for two years before schedule conflicts made it impossible. I finished with 1 course shy of a bio minor and 1 course shy of a math minor.
Unless she has both a huge amount of drive and there is a big pile of money around, don't send her to an ivy league school. There's lots of good schools that are less intense that have good departments in a broad spectrum of subjects. Encourage her to take a bunch of survey courses her first year. As my brother demonstrated, you can change your major 3 times and not lose any time.
I've been a sysadmin in 4 positions over 20 years. If done well, a sysadmin has time to read slashdot, and even the strawbale construction mailing list.
In brief a sysadmin's job is to make it possible for other people to do their job vis-a-vis computers.
There are several ways to look at the value. Some have already come up:
* Comparison to other forms of preventive maintenance. * Comparison to housecleaning functions, like janitor. (This may not be the comparison you want to raise. May go badly at your next salary review.) * Comparison to doing the job without computers. * Comparison to people doing it themselves.
The latter IMHO is the best of the bunch. * If you can fix a clerk's problem in 10 minutes that would take him all day, then in some sense your time is worth one clerk's weekly salary per hour. * If you aren't there then you have the issue of the president's secretary has Office 2007, while the VP's secretary has Office XP, and the Mailroom clerk has Office 95 running on a virus infected windows ME box. Nobody can send a file to anybody with any reliability of it being read.
For public values of the comparison, look up Gartner reports. At one point I vaguely recall they said that the cost of having a computer on a person's desk was about 5K per year over and above the cost of the computer itself. The figure was more or less independent of whether the company had an IT department or not. Overall, the money spent on IT was balanced by the increased production of the rest of the people. Now while this may have been true overall, I suspect that it had a large variance and weak correlation.
One of the ways I coped with this was to keep a vi window open on my machine, and I documented my day at a level of detail that showed every time I had to shift modes, answering the phone, fixing hot spots, writing scripts, analyizing log files, trying to prevent problems. After a month they would tell me to stop emailing them 500 lines of detail per day. I would say, "thanks. Now I can get back to work. Writing that email costs me about 1/3 of my day.
You really are being asked to put 10 lbs of shi -- potatoes into a 5 lb bag.
I'm a teacher who ended up being the school's IT geek, and I am daunted by the task you face. At my school the vast majority of the kids are clueless about comptuters. They can use a web browser, and can put words in a row in MS Word.
There are a few that have played with script hacks. And a very small number that actually know how to program.
I see IT education as having the following goals:
0. Teach touch typing. If you have to think about what your fingers are doing, you lose track of everything else.
1. Teach uses for the computer: * Internet: Searching strategies. * Internet: Safe use. * Text processing and layout. (use styles in word, darn it!) * Calculation, statistics, graphing and modeling in excel. * How to back up/move your data. Maintaining multiple versions.
These skills should be taught in conjunction with the kids other classes. E.g. Tie spreedsheet classes into their science or math classes. Tie text processing into their language arts classs.
2. Use the computers to teach troubleshooting/scientific method. Computers are a good way to teach reasoning.
3. Finding new tools. Present kids with a problem -- E.g. a photo with part of it in color, part in black and white, or a data set that needs perspective 3d to visualize. Let them search, download, install, and try out various things. (Lots of merit in each kid having his own VM for this.)
4. Show them logo. This is one of the best ways to get kids hooked on programming, as the feedback is so immediate, and it has elements to engage both left-brain and right-brain.
That's one course.
The second course is one in actual programming. Once the kids had mastered something a bit harder than "Hello WorldF" I would split the class in half, and give the two halves different easy problems. Say, the classic 4 banger calculator problem, for one, and a paragraph reformater for the other. Throughout this talk about the importance of code documentation. Now we add a feature to each problem, but the paragraph reformatter guys have to add a feature to someone's 4 banger, and vice versa. So each has to add a feature to someone else's code. Kids are marked not only on how well they did the original assignment, but how well the next person in line did with their code. Pairings are changed with each cycle.
For languages, I don't care, but given the sloppy nature that most kids today bring to academics, I think there is merit in a 'discipline and bondage' language that requires declarations and is strongly typed.
Some form of bottom up approach will work better than top down design will work better with kids, and overall will be easier to teach. You want a methodology that allows lots of immediate results, both so the kids get rewards for doing things right, and so that you can monitor their progress.
This second course is not one that I would make compulsory, or if it is, I'd want some provision for allowing kids flexibility as to which year they took it. I found teaching math that many young brains aren't ready for abstract thinking until they are 13-14 years old.
I would be interested in hearing what you decide. sgbotsford at gmail.com
Rsync can be versioned just fine. You tell it to rename the old version of foo.jpg to foo.jpg.back. After the backup runs, another script runs on the server that pulls the mtime of foo.jpg.back and uses that to rename it to foo.timestamp.jpg.
I implemented this in my last job where home directories were served by samba. The clients were winsooze. I had a large list of directories from the users' profile directory that were not backed up, because they changed too often, and few users would be aware of their existence. (browser caches, doc temp files...)
Anyway the backup folder for a user would typically be 2-3 times the size of their primary directory over the course of a year.
This size was maintained by another script that pruned the backups. The weekly prune removed all but the last version of the previous week. The monthly prune removed all but the last version of the previous month. The quarterly prune removed all but the last version of the previous quarter. So at any given time, you usually had dailies for this week and last week, weeklies for the current and previous month. Monthlies for the current and previous quarter.
Implementation: Assuming you have shell access on a unix server somewhere that you have space, then you need to establish a communication channel between the client and the server.
You need to put ssh and rsync on the client, and either run rsyncd as a service on the client, and pull from the server, or some form of cron on the client and push from the client.
If you have samba running on the server, you have a browse mechanism for users to restore files. Store the server's location as one of the user's "Network Places"
Of course all of this is run either through some form of VPN or each channel runs through ssh.
Caveats: As with many backup systems, this doesn't handle changes in the file system in a fully robust manner. There will be errors if the user moves or renames a file between the script starting and the file being backed up. Since this system keeps multiple versions, and since backups are handled at the file level, the occasional single bad file or missed directory is not the end of the world -- it just means that a less recent version is used.
It also doesn't handle huge files such as large MS Access file or Photoshop files with complete grace due to their size. However someone who uses these programs really needs an in house backup system.
In use the most common request from my users was for a file from 2-3 days ago. Usually what had happened was that they logged out without saving, and large files would not be saved completely. Most of the time this happened with powerpoint files.
Wait a minute. Something is off kilter here. First approximation the density of a gas is proportional to it's molecular weight. Methane = CH4 = 5. Water = H20=18 Air = O2+N2 ~~ 30. I'm not sure what the saturation point of water vapor in methane is, but even at 100% water vapour the density is only just over half that of air. Pure methane has a density of 1/6 that of air.
The density of air is around 1.2 Kg/cubic meter. So each cubic meter of the methane is going to have between 0.5 and 1.0 kg of lift.
Hot air saturates with water vapour at around 40 g/cubic meter. As a warm wet air mass rises, it cools, and that cloud is not very successful at keeping the all the water in tiny droplets. We call this process "rain"
We are asked to believe that methane can keep 10 to 20 times as much water as droplets in a fine enough dispersion without raining out for long enough to cover significant land masses?
What is the mechanism for creating these droplets? In a cloud the vapour condenses to form droplets. I can see no plausible mechanism for cold methane (Rising methane plumes would carry water from the ocean floor -- 4 degrees C roughly -- to evaporate at least it's own mass in water while rising through the water column.
Alternately what is the mechanism that produces sub fog sized water droplets is such profusion?
Things the help design in my not very humble opinion.
1. If you are duplicating a commercial project then duplicate the interface unless you have good reason not to.
Since much of usability is 'what you're used to' then this makes it easier to get people to switch.
2. GUI and command line interface.
Another poster pointed out that autocad has GUI and commandline interfaces.
On NextStep you could usually do things either through a GUI or through a command line. The two weren't always the same -- e.g. it may take 3-4 commands to duplicate what was done in a wizard. The documentation for the commands was sparse. Much of this is true today for Mac OS X, but the command line documentation is even harder to find.
Although confusing as hell for other reasons, the configuration package for IBM AIX had a neat feature: As you selected options in the configuration panel, it built the command line to do the same thing in a 1 line window near the top of the panel. For people who were trying to become wizards, this was a useful way to find out what command did what.
3. Consistency, consistency, consistency.
One of the nice things about macs is that once you start to get to know your way around, much of what you learned for one program translates to a similar function in another. Those things that most programs do, (Open a file, print, preferences) are usually in the same place every time. The layout deeper into the program is more variable.
For those of you who worship the command line: Look at the huge variation even in the submission of options to the command line. E.g compare that of find, dd, rsync, and ls. This got me into trouble on my first sysadmin job. cp -i is interactive copy. rm -i is interactive remove. di -i must be interactive disk initialization. Nope. di was disk information. -i meant initialize. Fortunately it was a client machine, and had no user files on it. (I won't swear the program was di -- this was 20+ years ago on NextStep 2.0)
Yesterday we had to help a friend set up his email on his new used Mac. Much confusion ensued before we discovered that the email account setup interface for Leopard and for Tiger are quite different. There may be a good reason for this change.
One graphic design program I tried had numerical entries for some tools, drag and drop for some tools, and both for some tools. But in some cases with the both, it would show you the current numerical selection, but you couldn't change it, even though the widget appearance was no different from one that you could.
The school where I have may day job uses a school information management system called renweb. It does some cool stuff, and collects a lot of information under one roof, with modules for grades, assignments, lesson plans. It's very modular -- there are some 40 different modules in it.
But:
You can't have more than one module open at a time. Which means if you need information from another module, you are stuck.
You can't minimize the application when you have a module open.
If you click the exit button, in some modules you are prompted, "discard changes", some save automatically, some discard silently.
The exit button can be anywhere on the panel, and not always next to the save button.
A while ago my personal web page for my tree farm. (My other job) had dual navigation bars, a horizontal one under the logo, and a vertical one on the left margin. I suckered my inlaws into test driving the web site, and watched. In particular I watched the mouse. It kept wandering between the two nav bars.
I tore into the code, and consolidated into a single nav bar.
4. Code last.
One of my CS profs insisted, "Write the user manual first." I think that helps clarify a lot of how the internal coding happens. Open source needs fast UI proto typing tools. If you want to get non-coding designers in on the fun, then we nee
Our school had fiber brought in as part of the Alberta Supernet project. They had to lay 30 km of fiber at a cost of several hundred thousand dollars. That was laying it next to nice roads where the workmen had a 40 minute commute home at night.
Mines are difficult to move. The technology of cheaply transporting multiple cubic kilometers of granite crust is not well developed. (Cheap compared to a fiber link)
With a name like 'pitchblende" I suspect that his company is one of several that is mining uranium in northern Saskatchewan. If so this is some of the most unfriendly terrain for laying cable. Over most of this country there really is no 'dirt' It's granite, or it's bog. Some places it's just glacial till. Bad gravel roads cost a million dollars a mile. When the Gulf mine went in on Wollaston Lake, the initial road was so bad that the trucks carried 4-6 spare tires, and after each trip replace the entire set of tires.
Some of these places only have winter roads: Wait until the lakes freeze solid enough to support a D9 or D12 cat, run the cat over the lakes, push the trees aside. Then, depending on the quality of the road, bring equipment in with normal semis (fully chained wheels) or huge sleds towed by those same cats. Typically a winter road requires a few weeks of preparation each year, and is usable for 6-8 weeks. All the heavy equipment, fuel, non-perishible supplies come in during this time.
If the mine site has more than a few dozen people at it, they will probably carve a gravel air strip near it. Otherwise they use the lake and float/ski planes. If the latter, then for a few weeks during freezeup and breakup you are either stuck there, or you have arrangements with a helicopter charter company.
From the mining districts in northern Saskatchewan it's 200-300 kilometers to LaRonge, a town of about 5000 people. I don't think LaRonge has fiber, but it may have microwave relay to Prince Albert.
If Pitchblend has at least a winter road into his area, it might be possible to put up a chain of wireless relays. This would require a tower periodically. Given that the country is fairly flat, and the trees pretty scrubby (30 -40 feet tops) he might be able to get away with a tower every 30-40 miles. If a winter road maintenance would be a bitch. Except in winter, access would be by helicopter or canoe. Nothing quite like working in the northern muskeg during horsefly or blackfly season, I tell you.
I've got satellite internet. It's erratic. At non-peak times it's pretty good. At peak times it is still better than dialup. There is a lot of latency, and a huge amount of jitter. (I've had ping times as long as 5000 ms. 800 is more typical)
That said, Pitchblend may want to consider satphones. They are pricey, with plans typically running a few hundred dollars a month and time around $2/minute. But it may make a good backup alternative when VoIP goes pear shaped. The advantage is they use different satellites, ones closer to earth. The disadvantage is that depending on how far north he is he may not have continuous coverage. A given satellite is visible for only about 7 minutes, and the polar orbits I don't think are full.
Not sure how 'pre teen' your kids are. I caught the sci-fi bug in 4th grade. I'm pretty sure I had read most of the classics by the time I was in eighth grade -- when I turned 13.
Much depends on the maturity and determiniation of your kids. Kids are changing incredibly fast -- stuff that would be suitable at age 13 is too heavy at 8 or 9.
An earlier reader is correct -- things that the kids aren't ready for will generally be ignored. If there are too many bits that are hard they will drop the book. Things they are almost ready for will spark questions. Be ready for the questions, and you will have a priceless opportunity to shape your kid's values.
Don't be afraid to give them adult books. (Adult = normal book vocabulary/sentence structure, not sexual.) If their school reading scores say they are reading at a grade 8 or 9 level, let them go!
I read SF almost to the exclulsion of everyone else, and it has shaped my attitudes in the following way: * I'm constantly coming up with different ways of doing things. So far out of the box, that most of the time I can't find the box. * I generally believe in solutions. Every problem has a solution, if we are clever enough, and early enough to find it. (At work at present, the school is faced with closing. Most of the staff are resigned, and looking for other work, or carrying out the tasks of wrapping up the closure. I'm still bubbling with ideas.) * Very upbeat, and take the long view. Climate warming? What the global committee calls a catastrophe -- loss of the ice, 20 meter sea level rises -- is but an inconvenience. Turning the planet into Venus is a catastrophe.
For early exposure to the less kid like books for these authors, look at collections of their short stories.
So, what do I consider a classic:
* All of Heinlein's juveniles, and for almost teens, all of his middle period -- up to Moon is a Harsh Mistress. The stuff after that is a bit disjointed
* Asimov. As a kid I didn't find the Foundation books dry at all.
* Arthur C. Clarke -- Tales of the Whitehart, Fall of Moon Dust, Earthlight
* Alan Norse (Nourse?)
* Back issues of Analog. (One librarian had the foresight to have them bound -- 6 issues per volume. Extra polish on her halo.
* Madelaine L'Engle
* Lloyd Alexander
* Alan Garner
* Jules Verne -- He was very prolific. Lot of titles that are hard to find in English.
* H. Rider Haggard -- Not strictly Sci-fi
* Robert Louis Stevensen.
* Edgar Allen Poe. (Yes they are dark. Kids love gore.)
* Ursula LeGuin (Wizard of Earthsea, Tombs of Atuan)
* Lois McMaster Bujold
* Anne McCaffrey
* Clifford Simak
* Poul Anderson
* Fritz Leiber
* Hal Clement (Especially if they have an interest in science in general.)
In a word, start early, start often. Don't restrict it to the classroom.
I work with youth (I'm the IT department for a small school). One of the problems we have is that the kids we're getting aren't firmly connected with reality. My working hypothesis is that too many hours of TV and video games shows that you can make it all better with a restart. They don't believe that 'Actions have consequences' They horse around, and break a window. It doesn't come out of their pocket, but rather their parents. They get busted for dope, and are affronted that we expel them. They get caught bullying, and the other person's pain isn't real to them.
I use the 'actions have consequences' as a lietmotif in my conversations with them -- And I enforce it in my operation of the lab. I have scripts that can enable/disable accounts/squid access by cron. Last week the lab was a mess. I put up a sign -- no internet access until lab is spotless. 4 boys went for mops, brooms, and picked up the lab, and emptied the trash. I'm working on a script that during free time will reboot the entire lab every 15 minutes, putting a message on the screen. "I want the missing mouse back" or "Put the keys on the keyboard at station 10 back in their correct places."
In terms of teaching skepticism, I tell stories. All of them start plausibly, and get more and more outlandish:
One time with the western sky dark on a canoe trip, a kid asked if it was smoke or storm. (We'd been dodging forest fires for two weeks.) I replied 'neither. Two days ago a volcano in the Aleutian Islands off Alaska blew up and put tens of cubic kilometers of dust into the stratusphere.'
'How long with it last?'
'Oh about two years. The government is already sent notices nationalizing all the grain in the elevators, and is preparing to ask all the southern canadian farmers to replant with rye, as a possible crop, but right now they are predicting a total crop failure for most of Canada.'
'What will we eat!'
'Good question. How many cans of stew will it take to hold your family for two years?'
This goes on for about 15 minutes, with occasional reprises for late arrivals. During the first 10 mintues no one asked how I knew this, being in the middle of the bush. When I claimed to have a small shortwave radio in my camera box and to listen to BBC world report, they accepted that uncritically, although no one of them had ever heard of shortwave radio. No one asked to see/hear the radio.
After I broke out with a grin, the kids were embarassed at being sucked in. That's when I gave my speach about critical thinking, and said it was a game: As a teacher it was not my job to give them facts. It was my job to teach them to think. Every 'factoid' (Something that looks like a fact but is unverified) has to be examined on the basis of: * It's source. Scientific American in general is a better source than National Enquirer. * The size of it's claims. Reasonable claims are more credible. Unreasonable claims need better evidence. * The process surrounding the collection of this information. Anecdotes are less credible than measurements. * The plausibility of this story in connection to everything esle you know. Does it fit?
Then I point out: * I have a reputation for story telling, both as a raconteur (never let the facts get in the way of a good story) and as someone who uses stories to teach lessons. * This was making extroidinary claims. None of them were old enough to experience a volcanic eruption that had local effects. (I remember the dark skies of Mt. St. Helens) * The process was implausible. All of them had seen my camera box, and its contents. None of them had seen anything but camera stuff. * Volcanoes are unusual events, with a total disconnect from everything they know. They should have been asking questions like mad: ** How often does this happen? ** What will happen in the U.S? ** Will Europe be affected? ** What kind of plans does the g
Maybe they can make the file system layout easier.
on
OS X Snow Leopard Details
·
· Score: 2, Interesting
I'm revealing my age here, when I say that my first unix based OS was NeXTstep.
One of the things I liked about Next was that an application kept to itself: Wherever you installed it, everything was in a directory "ApplicationName.app" This made uninstalling easy. It also meant that installing an application on a network file system made it available to all NextStations on the local network. (In some cases a 'dwrite global applicationname value' was needed for licensing for individual machines.)
Apple has not insisted on this. While many applications will work this way, now files are also stuffed into various Library directories. Uninstalling applications manually is no longer trivial.
Furthermore, some applications insist on writing to their own program directory.
I wish that apple and other OS's would implement a new security model regarding file spaces.
1. There are three file spaces: OS, Application, and user. Each can be divided.
2. The OS space consists of the distro along with applications from the distro vendor. For Windows the OS would include WordPad, but not Office (sold separately) For Mac it would include Mail, but not Aperture. For linux it would include/bin and/sbin. X and it's support files could go either way.
2a. The OS space has at least the following three subsections:/var contains files that change on a frequent basis./something contains files that change on an infrequent basis and depending on security setup could require special privilege or mode to modify./everythingelse which in normal operation is read only. -- Not even root can modify/evertythingelse/bin/ps without jumping through hoops.
3. User space. By default user space has a directory for each user, with access restricted to and controlled by that user. This is pretty much the way things are now.
3a. User space/group space. Methods for collaborating and sharing documents.
4. Application space. app space is done on 1 top level directory per vendor. Acrobat reader goes in/Apps/Adobe/AcrobatReader. Photoshop can go in/Apps/Adobe/Photoshop or/Apps/Adobe/CS3/Photoshop depending on the whims of Adobe.
The key here is that the adobe installer does not have write privileges outside of the/Apps/Adobe tree. Just as user smith can't write to user jone's files, nor should Adobe be able to write to microsoft files. This implies that some program equivalent to Next's 'buildservices' needs to periodically run to pick out what programs provided services for other programs.
5. In a general setup, no user should be able to execute a file in a directory they have write access to. Some mechanism for installations, and for developers needs to be made, but as a general rule this would go a long way to intercept malware. For users (as opposed to developers) having executable code in their directories is not a benefit.
Everyone is harping on the evils of SUV's but the original article is about aspects of racing car design that can make for better (less polluting, more fuel efficient) cars.
My ideas:
1. Require all cars to have an 'instantaneous' fuel economy guage. One that tells you how many mpg you are getting right now. If well designed, it takes into account your change in speed.
2. Increase the CAFE limits. Use taxation to enforce them. E.g. for every mpg your fleet average is above the limit, your corporate income tax goes up 2%. For every mpg your fleet average is below the limits, your income tax rate drops.
3. All vehicles come with a trailer hitch. People are encouraged to use a trailer when they need extra hauling space. It's almost always more efficient.
4. Part of driver's ed is dealing with a trailer.
5. New road construction builds for higher pressure tires. (Higher pressure = smaller footprint = more load per inch tire width = higher road wear.) Higher pressure = better gas mileage. There is a reason that big rigs run at 70 psi.
6. Modify the vehicle registration and insurance laws so that people aren't penalized for owning special purpose vehicles. (About 3 times per year I wish I had an old 3 ton farm grain truck. I can buy them for $1-2K. But it costs me $600 to insure it. And $100 to keep plates on it. Not worth it.)
***
Construction details: While race cars are one source of inspiration, much can also be learned from small aircraft. A Cessna 150 gets something like 13 mpg -- but it's moving at 125 mph, and has these big things sticking out in the breeze. I can't believe that a non-ducted air screw with half the path blocked by the cabin can be an efficient way to turn energy into momentum.
The three largest contributors to fuel economy:
1. GVW. More mass = more energy to change velocity. More friction.
2. Cross section. Above residential speeds, air resistance is the largest factor slowing you down. How 'bout a line of cars with the passenger compartment build like the cabin of a Cessna 180? Two narrow seats wide. Wheels outside the main cabin line for stability. Or even like a piper cub: 1 seat wide. Sure you couldn't use these for everything, but look how many miles are driven by single occupant cars.
3. Tire pressure. Look at the difference between the rolling resistance of a highway racing bike and a commuter tired mountain bike at half the air pressure. Same thing works with cars. At one point I had a beater that was going to need new tires before winter anyway. So with each fill up, I put two more psi in the tires. Fuel economy went up by 2% for each psi added. Of course the middle of the tread wore out pretty fast.
Sure right now tire pressure is part of the suspension system. Some serious engineering required. And if you make the tires too skinny, there are problems with stopping. (Higher friction per square inch means more heating, = tire melting = lower deceleration) On the other hand a skinnier tire has less problem in snow and is harder to hydroplane one wet roads.
Ideally it also has a viewfinder, so can be used without nearly as much battery drain. (You also get sharper pix when you brace the camera against your face.
Finally: Get it 6 months ahead of time, and practice, practice, practice. The last thing you need is to find that you have 2000 murky blurry photos because you didn't understand how to use the camera.
I teach photography at my school. One of my students consistently has photos that are 3 stops underexposed, and a blur that looks like tehre is a permanent film of vasaline on his lens. I point out to him that perhaps he should read the manual. (It's actually a class on how to follow directions.)
You're walking down the street, wearing a bunch of reprogramable rfids. The lamppost pings you and the people around you. Your receiver picks up responses at random. Reprograms your chips to imitate those around you. For the distance to the next lamp post, you become a mishmash of the people around you at the last post. Next post it happens again.
Of course now you can be tracked by the duplicated signals. You are a disturbance in the order.
Or you have a normal identity, and a false one. Get aboard a crowded bus. As you step off the bus, you switch to your false set of rfids.
Or you seek out other people who are wearing by chance reprogramable rfids. And you duplicate your ID onto them, so the RFID tracking system has 30 copies of you running around, and none of them are you. Do this a lot when you have nothing to hide, so that you are tagged as a shit disturber, but harmless in the system.
When Adobe allowed FrameMaker to escape there were lots of library problems. I managed to get it to run under Redhat 6 and 7, and Fedora Core 3. It wouldn't run under 4. Seg fault. It runs under 6.
One of the 'virtues' of Microsloth is that degree of backward compatibility that seems to work a lot of the time.
One of the virtues of unix systems generally is that libraries can be in use by multiple programs at once with only one copy loaded.
If every application has to include a full set of libraries, then you to the winsnooze model.
How does a commercial vendor work it so that their application can run on many distributions of linux?
My ideas: 1. The application starts from a script. The script uses a library search path that checks the application library directory first.
2. The installer checks the libaries it knows that it needs. Perhaps even a 'hello world' program that uses every library call that the main program uses to see if the results are sane.
3. For those libraries it needs but aren't in the main systems libraries, it installs known working copies in the application library folder.
4. As an alternative the vendor provides an open source 'shim' library for all of it's application's calls. Because this part is open source, the translation and recompilation can be left in the hands of the groups that want it, with the vendor responding to demand by providing shims for the more common distributions.
You go to radio shack and buy a metal foil wallet. It has an outside pocket of nylon for rfid cards you don't care about. Or it has a pocket on the inside so you just open the wallet and wave it at the reader.
Or if you are really ticked, you make a transmitter that listens for RFID triggering transmissions, and responds with a blast that knocks the RFID response detector off line. (wear your metal underwear.)
If you used multiple antennas you could in principle get a bearing on the source, and useing phased array transmission blast just the source. How practical would it be to build this into a brief case? Put it in the back of a pickup with a fiberglass canopy. Knock out every RFID detector along your route. (Ok, ok, So you need a welding generator as a power supply. So shoot me.)
As the entire IT department for my small school, I exclusively use OS OS's for my servers: Three freebsd boxes for file server, data server, backup (at the other end of the building) and an OpenBSD box for my firewall.
Periodically I look at using freebsd or linux on the desktop, but end up turning it down, as a huge increase in complexity.
We have two applications at present that require winsnooze or mac to run. One is "The Learning Equation" which is a grades 7 through 12 set of math programs. The other is "Accelerated Reader" which basically acts as a record keeper/testing progress program for running a recreational reading program.
Accelerated Reader REQUIRES adobe acrobat reader for the reports. It doesn't require a pdf reader, it requires AAR. Further not just any version of AAR. Has to be either 5 or 7. And it does this by checking registry entries. Yes I've written to the company. It also requires flash. AR works. We have kids who never read a book before they came here now reading books for pleasure.
TLE may run under wine. Haven't tried it. Somehow I don't think that wine running on a 64 MByte 500 MHz PC will have reasonable performance. Correct me if I'm wrong.
TLE only sort of works. Most of our teachers find that it has a few strong points for illustrating concepts that are hard to do on a whiteboard, but it is ineffective at presenting new material, giving practice in using it, or assessing mastery of it.
The problem is more fundamental. Educational software is expensive. I told one software company this, and came back with surprise: "It's not even $30 per seat!" I pointed out that their particular package could only be used for about 6 hours of instruction during a course. A textbook, on the other hand, was typically used for half an hour a day in class, and another half hour to hour for homework. The TCO of a textbook worked out to being under 25 cents per instructional hour. If we use the same 5 year schedule for software it works out to about a buck an hour, not counting time for maintenance. I told him that I want educational software that could compete with a textbook.
Worse than this, however, is that using ed software is much like bringing auxilary materials into the classroom. It makes a nice change, but you can't base a course on it. As far as I've been able to find, no one has a completely computerized math curriculum. That is, a system that would be suitable for home schooling, distance learning, or self taught student.
What OS education needs are good authoring systems for educational software -- A system that enables a good teacher to create a 'smart textbook'
I am partially deaf in one ear, I think due to sitting for 6 years next to the raid array for the department. Three of those years a disk had a squealing bearing. The company claimed that this wasn't sign that the disk would crash, so my boss wouldn't replace it.
Do not live in the server room.
In another job, the server room was kept at 55 degrees, with an air exchange every 60 seconds. We had hooks at the server room entry for people to keep a jacket or sweatshirt when they had to come in to work on the systems. We racks of dell 1U poweredge servers with their 4 tiny fans that sounded like jet turbines spinning up. I started wearing ear protectors in there, partly to protect my ears, partly so I could ignore people who tried to talk to me, mostly to keep my ears from getting cold.
Do not live in the server room.
In my present job, despite the air conditioning going all the time, and a fan pulling air out of the room through a hole in the wall, I've not been able to get the server room below 80F.
Do not live in the server room.
Racks: Building lifespans are typically 15-25 years before major renovations on the interior. Plan for 10 racks. Size the air conditioning for three times the initial power demand. (HVAC engineers chronically underestimate the amount of cooling AND you want enough capacity for expansion BUT 10 years from now, you'll probably be using less power per box.) You won't need them now, but you will later. If your bosses only have three-site (they lack foresite) then insist on an equipment storage area adjacent to the server room to expand into later.
Elevated floors. If you can possibly arrange it, ask for 16" elevations on the floors. During the construction, insist on the subfloor being polished smooth. One job required that I use the floor space as a crawlway to install cableing. I got really tired of rough concrete, dust, spiders, cockroaches, and bangomg my head on the cross pieces. It was only 12"
Network plumbing. Assume that the number of cables coming to the room with go up by at least a factor of 8 over the lifespan. The whole building needs to be done with mind to adaptable networking.
Run a dozen cables from the server room to your office. I don't know what you will use them for. But one job we ran 4 from the main wiring closet to the Sourcerer's Dungeon, and a year later we wanted more. (Let's see: One connected to the firewall which lived in the closet, 1 went to the cluster of managed switches, 1 was the 'copy port' on the switch cluster so we could track down problems, 1 was the interface on the firewall that copied all firewall traffic for the IDS. Oh yeah. Then 4th floor became part of our jurisdiction too, so we needed another cluster of switches...)
Service Bench. Either in the server room, or close to it you want a place where you can pull a computer, put it at a convenient height to work on. It's a real pain to work with bits of computer on the floor, or worse, having to stand on a step ladder. Having your office in the repair room isn't bad. People will see all the cruft and clutter, and assume you are busy as hell.
Wheeled racks: You can get some increase in density by having your racks on dollies, so they can live close together, but be moved around to work on. This is what you will be doing if you only plan for 4 racks, and three years from now have 6.
Your office:
Ask for one of the fancy 6 monitor surround displays. You will probably only get 4. Meanwhile, if you don't already, get at least two on your desk.
It's very likely that you will want one or more boxes for ancillary purposes. E.g. experimenting with new OS's and such.
Remember too, that all systems change. Six months from now comes an OS upgrade. You want enough spare machines to test this out before attacking the production servers. This may mean another rack.
Give the phone people their own space. A phone person is not your employee generally. You don't want them in the server room without a keeper. (Not if security is a concern, anyway).
... In the computer lab at the school I work at.
* I've got the license.
* It runs ok on a 700 MHz machine with 256M ram.
So far I've seen nothing of XP that makes me want to upgrade.
(I know, I should be running linux. But these computers have to support several ed packages, somehow running a an application under Wine under linux on machines that aready are low on memory just doesn't hack it.)
It's not like we're short of muskeg...
on
Google Goes Green
·
· Score: 1
Define destruction: This land is fire succession: In the normal cycle of events it burns to the ground every 30 to 50 years. Part of the cycle. Tarsand mining is more disruptive, because it messes up the drainage patterns some. But the companies have to put it back more or less as they found it, and they are limited in what chemicals they can leave behind.
If you have ever spent any time walking the ground in that part of Alberta it is tedius country. Once you are even a couple hundred meters away from the Athatbasca River it is so flat you can see only 50 meters or so. Much of it is either spagnum peat bog interspersed with black spruce or tamarack, or its sandy soil covered with caribou moss and lodgepole pine all in various stages of recovery from the most recent fire.
Given that Canada has tens of thousands of square miles of terrain that is indistinguishable from this, and another million or so that is very similar, sacrificing a few tens or even hundreds of square miles on a temporary basis is not an unreasonable price to pay.
For the companies do have to reclaim it. Is the reclamation perfect? No. It doesn't have the same degree of randomness as the orginal landscape. Overall it takes about 20 years.
I've also seen the waste heaps at Uranium City. El Dorado Nuclear shut down decades ago. There was a war. They needed uranium now. It was an era when we cared even less about polution. Beaverlodge lake had acid leaching into it for years from the refuse heaps of the concentrator plant. When I saw it the first time, twenty years after the plant closed, in 1975 the lake was sterile. No fish. No algae. Could see the bottom 30 feet down through clear blue water. I saw it again 20 years later. The lake now had trees growing on the margin. Grasses and reeds were colonizing some of the shallow bays.
Uranium City is on the Crackingstone Peninsula on the north shore of Lake Athabasca. It's Canadian Shield -- granite. The land has almost no carbonates in its soil so acidic waters take a very long time to neutralize. Yes despite all this, the land is recovering. Not fast, but steadily, without help.
The last mine at UC closed shortly after my first visit. Within weeks UC was a ghost town. Now about 50 people live in a town that once housed 5,000. Some of the houses were put on barges and taken to other settlements on Lake Athabasca. The rest? Some are burned -- victims of school kids pranks during the last months. Many have collapsed from the snow loads. Most of the windows are broken. A few have trees growing in the foundations. It was a sad experience as I walked through this place. I looked at the empty places, and wondered who had been there? What were their dreams? UC is now a graveyard of dreams.
The tar sands are on the edge of the shield. Mostly sandstone and glacial till. Fair amount of buffering compounds to keep the soil pH from getting too out of whack. This is land that left on it's own would heal. But the companies give it a shove along that direction.
Query: Was the land a wasteland all the way from the processing plant to the cutting edge where the equipment was? I bet not. Yet much of that land has been mined and reclaimed.
Is it zero impact? In the short scale no. In a thousand years you'll have a hard time telling it happened.
I spend several weeks each summer traveling the wild country on and near the Canadian Shield. I love that country. But I don't begrudge the islands of temporary ugliness that allow our society the raw materials that we need. Canada is not short of wilderness.
One of the things that I've come to appreciate from my trips is that with all of their foibles and faults, people are precious. After three weeks of seeing no one but other members of our group, no sign that anyone had ever been here before, no contrails across the sky of jets taking people to distant lands the discovery of a fifty year old tobacco tin where a game trail between to lakes meets the shore, is a source of wonder.
While conventional oil is nearing the peak, as is conventional natural gas, there is an awful lot of carbon available in tarsand, heavy oil, oil shale, coal bed methane, ocean bed ice methane clathrates. And we are no where near out of coal. Probably have enough fossil fuels to run the CO2 up to a percent of the air. (1% = 10000 ppm In the past two centuries I think the levels have risen from about 330 ppm to about 500 ppm.)
I left a sysadmin job at a university for a sysadmin job in private industry, for a 30% raise in pay. However the raise in pay was accompanied by: A 75 minute commute; Only 3 weeks holiday/year; no banking of overtime; life in a dilbert box, and three signatures required to buy a $100 ethernet card.
I left that position after a year and a half, taking a job at a local school as a sysadmin for a bit more than half the pay. Curiously, we found that the transition to half the money wasn't nearly as hard as we expected. We weren't putting as much into retirement plans, but between saving $6000 per year on the commute, and that lower salary is taxed at a lower rate, we have pretty much the same life style.
But I now had 12 weeks holiday per year, a 15 minute commute, 4 weeks per year where they paid me to go with the kids on canoe trips.
In my experience most jobs dealing with computers turn into infinite time sinks. You will be spending at least half your waking time at whichever of the two jobs you choose. Go for the one you will like.
Some years ago at one of the usenix conferences someone reported a similar project. However, instead of trying for chords and harmony, they chose natural sounds -- frogs croaking, crickets, bird chirps, water, wind. Their take: As former hunter gatherers we detect change in this environment more quickly, and few combinations are dissonant, further, the 'normal' environment swiftly becomes non-distracting.
So, for example, disk activity can be represented by the sound of bees. Incoming mail to the server by a chickadee chirp, successful network logons by a duck quack, non-succesful ones by a piliated woodpecker call, disk full warnings by various levels of thunder, depending on how full and how fast they were filling, network activity as wind in trees noise, Internet access as rushing water, cpu usage by frog choruses.
Implementation: Since this would require more than just vmstat as it's input, I would suggest a that it be implemented as a simple udp socket. Various programs then can run at various places on your network, and send packets to the socket, with an event type, and whatever addtional info you wish. Possibly the same mechanism that Big Brother uses for the data layer.
So...
Use existing BB modules to gather information.
Send to central receiver.
Receiver does two things:
A: Writes the BB webpage
B: sends an event to the sound generator.
Sound gnerator uses the event to set some parameter of the sound environment. This environmental sound can be a single short event, such as a bird chirp, or can set some level for reporting a continous event.
Sound generator has a library of sounds, either some form of sound clips, or parameters to hand to external programs.
Slashdot Mirror
My wife and I have just discovered the joys of digital audio on my/her itouch/iphone. It has not taken us very long to fill a 250 GB drive. People who keep movies must have a real storage problem.
The industry moved from 5" media to 3.5" media, and now 2" hard disks are coming in for more than just laptops. This has mostly come about for the sake of faster seeks, yes?
Given the disparate needs, perhaps the ordinary user needs different kinds of storage. A desktop that is running streaming video doesn't seek much. Do you care if the heads have a 20 ms average access time?
So you guys who have more knowledge about hardware than I do:
How cost effective would it be to create a 12 platter 5" form factor drive that spun at 3600 rpm, but had the same areal bit density as current 2" drives?
***
Part 2. At one point I remember 'tiered' storage. Back when disks were very expensive compared to tape. You had a set of spinning disks, and a silo full of tapes. When a file hadn't been accessed for N days, it was migrated to tape. When it was called for again, you had a long pause while the silo found the tape and loaded the file.
Are we going to have a window where this makes good sense, substituting SSD for disk, and HD for tape?
Lots of good stuff posted already.
As a geezer currently in the job hunt I can identify with several comments.
I use interviews to learn about my possible employer.
* If the job doesn't sound interesting I won't bother.
* To be interesting, it has to have elements I've not done before.
* Is this a 'tie' shop? Excuse me, I think I left the roast in the oven.
If I were on the hiring side of the desk, I might ask, "What was the most recent language you learned, and how did you go about learning it?" Substitute any concept you wish for language.
Scenarios are good. I like the one response, "And then what..."
What do you know about Foo?, where foo is some aspect about the business the company is in. I interviewed with the Canadian Forest Service as a "Carbon Cycling Technician" I got the job because I had some experience programming, could obviously learn on my own, and also had spent the week before the interview reading up on the carbon cycle in forests. The interview asked me all sorts of questions about indicator species, eutrophication of lakes, reaction rates as a function of particle size. I was also asked if I had any experience snowshoeing, what my tolerance for mosquitoes, and was I afraid of heights. (Turns out that part of the job was taking samples at various times of year and climbing 40 meter towers calibrating CO2 sensors.)
It really helps if the programmers know enough about the business to make intelligent decisions.
My brother started out majoring in Math. Found that it was too abstract, and after a semester changed his major to Physics. Physics wasn't 'real' enough, so he switched to pre-med. Pre-med was full of very intense people who couldn't stand to get less than 95%. He switched to biology. One summer he worked in a lab culturing bacteria, so he took his masters in microbiology, and his doctoratal thesis was on some aspect of yeast.
I started in physics and wildlife management as dual majors and was able to keep it up for two years before schedule conflicts made it impossible. I finished with 1 course shy of a bio minor and 1 course shy of a math minor.
Unless she has both a huge amount of drive and there is a big pile of money around, don't send her to an ivy league school. There's lots of good schools that are less intense that have good departments in a broad spectrum of subjects. Encourage her to take a bunch of survey courses her first year. As my brother demonstrated, you can change your major 3 times and not lose any time.
I've been a sysadmin in 4 positions over 20 years. If done well, a sysadmin has time to read slashdot, and even the strawbale construction mailing list.
In brief a sysadmin's job is to make it possible for other people to do their job vis-a-vis computers.
There are several ways to look at the value. Some have already come up:
* Comparison to other forms of preventive maintenance.
* Comparison to housecleaning functions, like janitor. (This may not be the comparison you want to raise. May go badly at your next salary review.)
* Comparison to doing the job without computers.
* Comparison to people doing it themselves.
The latter IMHO is the best of the bunch.
* If you can fix a clerk's problem in 10 minutes that would take him all day, then in some sense your time is worth one clerk's weekly salary per hour.
* If you aren't there then you have the issue of the president's secretary has Office 2007, while the VP's secretary has Office XP, and the Mailroom clerk has Office 95 running on a virus infected windows ME box. Nobody can send a file to anybody with any reliability of it being read.
For public values of the comparison, look up Gartner reports. At one point I vaguely recall they said that the cost of having a computer on a person's desk was about 5K per year over and above the cost of the computer itself. The figure was more or less independent of whether the company had an IT department or not. Overall, the money spent on IT was balanced by the increased production of the rest of the people. Now while this may have been true overall, I suspect that it had a large variance and weak correlation.
One of the ways I coped with this was to keep a vi window open on my machine, and I documented my day at a level of detail that showed every time I had to shift modes, answering the phone, fixing hot spots, writing scripts, analyizing log files, trying to prevent problems. After a month they would tell me to stop emailing them 500 lines of detail per day. I would say, "thanks. Now I can get back to work. Writing that email costs me about 1/3 of my day.
You really are being asked to put 10 lbs of shi -- potatoes into a 5 lb bag.
I'm a teacher who ended up being the school's IT geek, and I am daunted by the task you face. At my school the vast majority of the kids are clueless about comptuters. They can use a web browser, and can put words in a row in MS Word.
There are a few that have played with script hacks. And a very small number that actually know how to program.
I see IT education as having the following goals:
0. Teach touch typing. If you have to think about what your fingers are doing, you lose track of everything else.
1. Teach uses for the computer:
* Internet: Searching strategies.
* Internet: Safe use.
* Text processing and layout. (use styles in word, darn it!)
* Calculation, statistics, graphing and modeling in excel.
* How to back up/move your data. Maintaining multiple versions.
These skills should be taught in conjunction with the kids other classes. E.g. Tie spreedsheet classes into their science or math classes. Tie text processing into their language arts classs.
2. Use the computers to teach troubleshooting/scientific method. Computers are a good way to teach reasoning.
3. Finding new tools. Present kids with a problem -- E.g. a photo with part of it in color, part in black and white, or a data set that needs perspective 3d to visualize. Let them search, download, install, and try out various things. (Lots of merit in each kid having his own VM for this.)
4. Show them logo. This is one of the best ways to get kids hooked on programming, as the feedback is so immediate, and it has elements to engage both left-brain and right-brain.
That's one course.
The second course is one in actual programming. Once the kids had mastered something a bit harder than "Hello WorldF" I would split the class in half, and give the two halves different easy problems. Say, the classic 4 banger calculator problem, for one, and a paragraph reformater for the other. Throughout this talk about the importance of code documentation. Now we add a feature to each problem, but the paragraph reformatter guys have to add a feature to someone's 4 banger, and vice versa. So each has to add a feature to someone else's code. Kids are marked not only on how well they did the original assignment, but how well the next person in line did with their code. Pairings are changed with each cycle.
For languages, I don't care, but given the sloppy nature that most kids today bring to academics, I think there is merit in a 'discipline and bondage' language that requires declarations and is strongly typed.
Some form of bottom up approach will work better than top down design will work better with kids, and overall will be easier to teach. You want a methodology that allows lots of immediate results, both so the kids get rewards for doing things right, and so that you can monitor their progress.
This second course is not one that I would make compulsory, or if it is, I'd want some provision for allowing kids flexibility as to which year they took it. I found teaching math that many young brains aren't ready for abstract thinking until they are 13-14 years old.
I would be interested in hearing what you decide.
sgbotsford at gmail.com
Rsync can be versioned just fine. You tell it to rename the old version of foo.jpg to foo.jpg.back. After the backup runs, another script runs on the server that pulls the mtime of foo.jpg.back and uses that to rename it to foo.timestamp.jpg.
I implemented this in my last job where home directories were served by samba. The clients were winsooze. I had a large list of directories from the users' profile directory that were not backed up, because they changed too often, and few users would be aware of their existence. (browser caches, doc temp files...)
Anyway the backup folder for a user would typically be 2-3 times the size of their primary directory over the course of a year.
This size was maintained by another script that pruned the backups. The weekly prune removed all but the last version of the previous week. The monthly prune removed all but the last version of the previous month. The quarterly prune removed all but the last version of the previous quarter.
So at any given time, you usually had dailies for this week and last week, weeklies for the current and previous month. Monthlies for the current and previous quarter.
Implementation:
Assuming you have shell access on a unix server somewhere that you have space, then you need to establish a communication channel between the client and the server.
You need to put ssh and rsync on the client, and either run rsyncd as a service on the client, and pull from the server, or some form of cron on the client and push from the client.
If you have samba running on the server, you have a browse mechanism for users to restore files. Store the server's location as one of the user's "Network Places"
Of course all of this is run either through some form of VPN or each channel runs through ssh.
Caveats: As with many backup systems, this doesn't handle changes in the file system in a fully robust manner. There will be errors if the user moves or renames a file between the script starting and the file being backed up. Since this system keeps multiple versions, and since backups are handled at the file level, the occasional single bad file or missed directory is not the end of the world -- it just means that a less recent version is used.
It also doesn't handle huge files such as large MS Access file or Photoshop files with complete grace due to their size. However someone who uses these programs really needs an in house backup system.
In use the most common request from my users was for a file from 2-3 days ago. Usually what had happened was that they logged out without saving, and large files would not be saved completely. Most of the time this happened with powerpoint files.
Wait a minute. Something is off kilter here. First approximation the density of a gas is proportional to it's molecular weight. Methane = CH4 = 5. Water = H20=18
Air = O2+N2 ~~ 30. I'm not sure what the saturation point of water vapor in methane is, but even at 100% water vapour the density is only just over half that of air. Pure methane has a density of 1/6 that of air.
The density of air is around 1.2 Kg/cubic meter. So each cubic meter of the methane is going to have between 0.5 and 1.0 kg of lift.
Hot air saturates with water vapour at around 40 g/cubic meter. As a warm wet air mass rises, it cools, and that cloud is not very successful at keeping the all the water in tiny droplets. We call this process "rain"
We are asked to believe that methane can keep 10 to 20 times as much water as droplets in a fine enough dispersion without raining out for long enough to cover significant land masses?
What is the mechanism for creating these droplets? In a cloud the vapour condenses to form droplets. I can see no plausible mechanism for cold methane (Rising methane plumes would carry water from the ocean floor -- 4 degrees C roughly -- to evaporate at least it's own mass in water while rising through the water column.
Alternately what is the mechanism that produces sub fog sized water droplets is such profusion?
I need more convincing.
Things the help design in my not very humble opinion.
1. If you are duplicating a commercial project then duplicate the interface unless you have good reason not to.
Since much of usability is 'what you're used to' then this makes it easier to get people to switch.
2. GUI and command line interface.
Another poster pointed out that autocad has GUI and commandline interfaces.
On NextStep you could usually do things either through a GUI or through a command line. The two weren't always the same -- e.g. it may take 3-4 commands to duplicate what was done in a wizard. The documentation for the commands was sparse. Much of this is true today for Mac OS X, but the command line documentation is even harder to find.
Although confusing as hell for other reasons, the configuration package for IBM AIX had a neat feature: As you selected options in the configuration panel, it built the command line to do the same thing in a 1 line window near the top of the panel. For people who were trying to become wizards, this was a useful way to find out what command did what.
3. Consistency, consistency, consistency.
One of the nice things about macs is that once you start to get to know your way around, much of what you learned for one program translates to a similar function in another. Those things that most programs do, (Open a file, print, preferences) are usually in the same place every time. The layout deeper into the program is more variable.
For those of you who worship the command line: Look at the huge variation even in the submission of options to the command line. E.g compare that of find, dd, rsync, and ls. This got me into trouble on my first sysadmin job. cp -i is interactive copy. rm -i is interactive remove. di -i must be interactive disk initialization.
Nope. di was disk information. -i meant initialize. Fortunately it was a client machine, and had no user files on it. (I won't swear the program was di -- this was 20+ years ago on NextStep 2.0)
Yesterday we had to help a friend set up his email on his new used Mac. Much confusion ensued before we discovered that the email account setup interface for Leopard and for Tiger are quite different. There may be a good reason for this change.
One graphic design program I tried had numerical entries for some tools, drag and drop for some tools, and both for some tools. But in some cases with the both, it would show you the current numerical selection, but you couldn't change it, even though the widget appearance was no different from one that you could.
The school where I have may day job uses a school information management system called renweb. It does some cool stuff, and collects a lot of information under one roof, with modules for grades, assignments, lesson plans. It's very modular -- there are some 40 different modules in it.
But:
You can't have more than one module open at a time. Which means if you need information from another module, you are stuck.
You can't minimize the application when you have a module open.
If you click the exit button, in some modules you are prompted, "discard changes", some save automatically, some discard silently.
The exit button can be anywhere on the panel, and not always next to the save button.
A while ago my personal web page for my tree farm. (My other job) had dual navigation bars, a horizontal one under the logo, and a vertical one on the left margin. I suckered my inlaws into test driving the web site, and watched. In particular I watched the mouse. It kept wandering between the two nav bars.
I tore into the code, and consolidated into a single nav bar.
4. Code last.
One of my CS profs insisted, "Write the user manual first." I think that helps clarify a lot of how the internal coding happens. Open source needs fast UI proto typing tools. If you want to get non-coding designers in on the fun, then we nee
I don't think fiber will cost under 30K.
Our school had fiber brought in as part of the Alberta Supernet project. They had to lay 30 km of fiber at a cost of several hundred thousand dollars. That was laying it next to nice roads where the workmen had a 40 minute commute home at night.
Mines are difficult to move. The technology of cheaply transporting multiple cubic kilometers of granite crust is not well developed. (Cheap compared to a fiber link)
With a name like 'pitchblende" I suspect that his company is one of several that is mining uranium in northern Saskatchewan. If so this is some of the most unfriendly terrain for laying cable. Over most of this country there really is no 'dirt' It's granite, or it's bog. Some places it's just glacial till. Bad gravel roads cost a million dollars a mile. When the Gulf mine went in on Wollaston Lake, the initial road was so bad that the trucks carried 4-6 spare tires, and after each trip replace the entire set of tires.
Some of these places only have winter roads: Wait until the lakes freeze solid enough to support a D9 or D12 cat, run the cat over the lakes, push the trees aside. Then, depending on the quality of the road, bring equipment in with normal semis (fully chained wheels) or huge sleds towed by those same cats. Typically a winter road requires a few weeks of preparation each year, and is usable for 6-8 weeks. All the heavy equipment, fuel, non-perishible supplies come in during this time.
If the mine site has more than a few dozen people at it, they will probably carve a gravel air strip near it. Otherwise they use the lake and float/ski planes. If the latter, then for a few weeks during freezeup and breakup you are either stuck there, or you have arrangements with a helicopter charter company.
From the mining districts in northern Saskatchewan it's 200-300 kilometers to LaRonge, a town of about 5000 people. I don't think LaRonge has fiber, but it may have microwave relay to Prince Albert.
If Pitchblend has at least a winter road into his area, it might be possible to put up a chain of wireless relays. This would require a tower periodically. Given that the country is fairly flat, and the trees pretty scrubby (30 -40 feet tops) he might be able to get away with a tower every 30-40 miles. If a winter road maintenance would be a bitch. Except in winter, access would be by helicopter or canoe. Nothing quite like working in the northern muskeg during horsefly or blackfly season, I tell you.
I've got satellite internet. It's erratic. At non-peak times it's pretty good. At peak times it is still better than dialup. There is a lot of latency, and a huge amount of jitter. (I've had ping times as long as 5000 ms. 800 is more typical)
That said, Pitchblend may want to consider satphones. They are pricey, with plans typically running a few hundred dollars a month and time around $2/minute. But it may make a good backup alternative when VoIP goes pear shaped. The advantage is they use different satellites, ones closer to earth. The disadvantage is that depending on how far north he is he may not have continuous coverage. A given satellite is visible for only about 7 minutes, and the polar orbits I don't think are full.
Welcome to the true north
Not sure how 'pre teen' your kids are. I caught the sci-fi bug in 4th grade. I'm pretty sure I had read most of the classics by the time I was in eighth grade -- when I turned 13.
Much depends on the maturity and determiniation of your kids. Kids are changing incredibly fast -- stuff that would be suitable at age 13 is too heavy at 8 or 9.
An earlier reader is correct -- things that the kids aren't ready for will generally be ignored. If there are too many bits that are hard they will drop the book. Things they are almost ready for will spark questions. Be ready for the questions, and you will have a priceless opportunity to shape your kid's values.
Don't be afraid to give them adult books. (Adult = normal book vocabulary/sentence structure, not sexual.) If their school reading scores say they are reading at a grade 8 or 9 level, let them go!
I read SF almost to the exclulsion of everyone else, and it has shaped my attitudes in the following way:
* I'm constantly coming up with different ways of doing things. So far out of the box, that most of the time I can't find the box.
* I generally believe in solutions. Every problem has a solution, if we are clever enough, and early enough to find it. (At work at present, the school is faced with closing. Most of the staff are resigned, and looking for other work, or carrying out the tasks of wrapping up the closure. I'm still bubbling with ideas.)
* Very upbeat, and take the long view. Climate warming? What the global committee calls a catastrophe -- loss of the ice, 20 meter sea level rises -- is but an inconvenience. Turning the planet into Venus is a catastrophe.
For early exposure to the less kid like books for these authors, look at collections of their short stories.
So, what do I consider a classic:
* All of Heinlein's juveniles, and for almost teens, all of his middle period -- up to Moon is a Harsh Mistress. The stuff after that is a bit disjointed
* Asimov. As a kid I didn't find the Foundation books dry at all.
* Arthur C. Clarke -- Tales of the Whitehart, Fall of Moon Dust, Earthlight
* Alan Norse (Nourse?)
* Back issues of Analog. (One librarian had the foresight to have them bound -- 6 issues per volume. Extra polish on her halo.
* Madelaine L'Engle
* Lloyd Alexander
* Alan Garner
* Jules Verne -- He was very prolific. Lot of titles that are hard to find in English.
* H. Rider Haggard -- Not strictly Sci-fi
* Robert Louis Stevensen.
* Edgar Allen Poe. (Yes they are dark. Kids love gore.)
* Ursula LeGuin (Wizard of Earthsea, Tombs of Atuan)
* Lois McMaster Bujold
* Anne McCaffrey
* Clifford Simak
* Poul Anderson
* Fritz Leiber
* Hal Clement (Especially if they have an interest in science in general.)
* E. E. "Doc" Smith
* George O. Smith
* Ben Bova
* Jerry Pournelle
* Larry Niven (Short stories esp.)
* Lloyd Biggle Jr.
In a word, start early, start often. Don't restrict it to the classroom.
I work with youth (I'm the IT department for a small school). One of the problems we have is that the kids we're getting aren't firmly connected with reality. My working hypothesis is that too many hours of TV and video games shows that you can make it all better with a restart. They don't believe that 'Actions have consequences' They horse around, and break a window. It doesn't come out of their pocket, but rather their parents. They get busted for dope, and are affronted that we expel them. They get caught bullying, and the other person's pain isn't real to them.
I use the 'actions have consequences' as a lietmotif in my conversations with them -- And I enforce it in my operation of the lab. I have scripts that can enable/disable accounts/squid access by cron. Last week the lab was a mess. I put up a sign -- no internet access until lab is spotless. 4 boys went for mops, brooms, and picked up the lab, and emptied the trash. I'm working on a script that during free time will reboot the entire lab every 15 minutes, putting a message on the screen. "I want the missing mouse back" or "Put the keys on the keyboard at station 10 back in their correct places."
In terms of teaching skepticism, I tell stories. All of them start plausibly, and get more and more outlandish:
One time with the western sky dark on a canoe trip, a kid asked if it was smoke or storm. (We'd been dodging forest fires for two weeks.) I replied 'neither. Two days ago a volcano in the Aleutian Islands off Alaska blew up and put tens of cubic kilometers of dust into the stratusphere.'
'How long with it last?'
'Oh about two years. The government is already sent notices nationalizing all the grain in the elevators, and is preparing to ask all the southern canadian farmers to replant with rye, as a possible crop, but right now they are predicting a total crop failure for most of Canada.'
'What will we eat!'
'Good question. How many cans of stew will it take to hold your family for two years?'
This goes on for about 15 minutes, with occasional reprises for late arrivals. During the first 10 mintues no one asked how I knew this, being in the middle of the bush. When I claimed to have a small shortwave radio in my camera box and to listen to BBC world report, they accepted that uncritically, although no one of them had ever heard of shortwave radio. No one asked to see/hear the radio.
After I broke out with a grin, the kids were embarassed at being sucked in. That's when I gave my speach about critical thinking, and said it was a game: As a teacher it was not my job to give them facts. It was my job to teach them to think. Every 'factoid' (Something that looks like a fact but is unverified) has to be examined on the basis of:
* It's source. Scientific American in general is a better source than National Enquirer.
* The size of it's claims. Reasonable claims are more credible. Unreasonable claims need better evidence.
* The process surrounding the collection of this information. Anecdotes are less credible than measurements.
* The plausibility of this story in connection to everything esle you know. Does it fit?
Then I point out:
* I have a reputation for story telling, both as a raconteur (never let the facts get in the way of a good story) and as someone who uses stories to teach lessons.
* This was making extroidinary claims. None of them were old enough to experience a volcanic eruption that had local effects. (I remember the dark skies of Mt. St. Helens)
* The process was implausible. All of them had seen my camera box, and its contents. None of them had seen anything but camera stuff.
* Volcanoes are unusual events, with a total disconnect from everything they know. They should have been asking questions like mad:
** How often does this happen?
** What will happen in the U.S?
** Will Europe be affected?
** What kind of plans does the g
I'm revealing my age here, when I say that my first unix based OS was NeXTstep.
One of the things I liked about Next was that an application kept to itself: Wherever you installed it, everything was in a directory "ApplicationName.app" This made uninstalling easy. It also meant that installing an application on a network file system made it available to all NextStations on the local network. (In some cases a 'dwrite global applicationname value' was needed for licensing for individual machines.)
Apple has not insisted on this. While many applications will work this way, now files are also stuffed into various Library directories. Uninstalling applications manually is no longer trivial.
Furthermore, some applications insist on writing to their own program directory.
I wish that apple and other OS's would implement a new security model regarding file spaces.
1. There are three file spaces: OS, Application, and user. Each can be divided.
2. The OS space consists of the distro along with applications from the distro vendor. For Windows the OS would include WordPad, but not Office (sold separately) For Mac it would include Mail, but not Aperture. For linux it would include
2a. The OS space has at least the following three subsections:
3. User space.
By default user space has a directory for each user, with access restricted to and controlled by that user. This is pretty much the way things are now.
3a. User space/group space. Methods for collaborating and sharing documents.
4. Application space.
app space is done on 1 top level directory per vendor. Acrobat reader goes in
The key here is that the adobe installer does not have write privileges outside of the
Just as user smith can't write to user jone's files, nor should Adobe be able to write to microsoft files.
This implies that some program equivalent to Next's 'buildservices' needs to periodically run to pick out what programs provided services for other programs.
5. In a general setup, no user should be able to execute a file in a directory they have write access to. Some mechanism for installations, and for developers needs to be made, but as a general rule this would go a long way to intercept malware. For users (as opposed to developers) having executable code in their directories is not a benefit.
Everyone is harping on the evils of SUV's but the original article is about aspects of racing car design that can make for better (less polluting, more fuel efficient) cars.
My ideas:
1. Require all cars to have an 'instantaneous' fuel economy guage. One that tells you how many mpg you are getting right now. If well designed, it takes into account your change in speed.
2. Increase the CAFE limits. Use taxation to enforce them. E.g. for every mpg your fleet average is above the limit, your corporate income tax goes up 2%. For every mpg your fleet average is below the limits, your income tax rate drops.
3. All vehicles come with a trailer hitch. People are encouraged to use a trailer when they need extra hauling space. It's almost always more efficient.
4. Part of driver's ed is dealing with a trailer.
5. New road construction builds for higher pressure tires. (Higher pressure = smaller footprint = more load per inch tire width = higher road wear.)
Higher pressure = better gas mileage. There is a reason that big rigs run at 70 psi.
6. Modify the vehicle registration and insurance laws so that people aren't penalized for owning special purpose vehicles. (About 3 times per year I wish I had an old 3 ton farm grain truck. I can buy them for $1-2K. But it costs me $600 to insure it. And $100 to keep plates on it. Not worth it.)
***
Construction details:
While race cars are one source of inspiration, much can also be learned from small aircraft. A Cessna 150 gets something like 13 mpg -- but it's moving at 125 mph, and has these big things sticking out in the breeze. I can't believe that a non-ducted air screw with half the path blocked by the cabin can be an efficient way to turn energy into momentum.
The three largest contributors to fuel economy:
1. GVW. More mass = more energy to change velocity. More friction.
2. Cross section. Above residential speeds, air resistance is the largest factor slowing you down. How 'bout a line of cars with the passenger compartment build like the cabin of a Cessna 180? Two narrow seats wide. Wheels outside the main cabin line for stability. Or even like a piper cub: 1 seat wide. Sure you couldn't use these for everything, but look how many miles are driven by single occupant cars.
3. Tire pressure. Look at the difference between the rolling resistance of a highway racing bike and a commuter tired mountain bike at half the air pressure.
Same thing works with cars. At one point I had a beater that was going to need new tires before winter anyway. So with each fill up, I put two more psi in the tires.
Fuel economy went up by 2% for each psi added. Of course the middle of the tread wore out pretty fast.
Sure right now tire pressure is part of the suspension system. Some serious engineering required. And if you make the tires too skinny, there are problems with stopping. (Higher friction per square inch means more heating, = tire melting = lower deceleration) On the other hand a skinnier tire has less problem in snow and is harder to hydroplane one wet roads.
I would add to this:
Get a camera that is at least weather resistant.
Ideally it also has a viewfinder, so can be used without nearly as much battery drain. (You also
get sharper pix when you brace the camera against your face.
Finally: Get it 6 months ahead of time, and practice, practice, practice. The last thing you
need is to find that you have 2000 murky blurry photos because you didn't understand how to use
the camera.
I teach photography at my school. One of my students consistently has photos that are 3 stops underexposed,
and a blur that looks like tehre is a permanent film of vasaline on his lens. I point out to him that perhaps
he should read the manual. (It's actually a class on how to follow directions.)
If you can make reprogramable rfids ...
You're walking down the street, wearing a bunch of reprogramable rfids. The lamppost pings you and the people around you. Your receiver picks up responses at random. Reprograms your chips to imitate those around you. For the distance to the next lamp post, you become a mishmash of the people around you at the last post. Next post it happens again.
Of course now you can be tracked by the duplicated signals. You are a disturbance in the order.
Or you have a normal identity, and a false one. Get aboard a crowded bus. As you step off the bus, you switch to your false set of rfids.
Or you seek out other people who are wearing by chance reprogramable rfids. And you duplicate your ID onto them, so the RFID tracking system has 30 copies of you running around, and none of them are you. Do this a lot when you have nothing to hide, so that you are tagged as a shit disturber, but harmless in the system.
When Adobe allowed FrameMaker to escape there were lots of library problems. I managed to get it to run under Redhat 6 and 7, and Fedora Core 3. It wouldn't run under 4. Seg fault. It runs under 6.
One of the 'virtues' of Microsloth is that degree of backward compatibility that seems to work a lot of the time.
One of the virtues of unix systems generally is that libraries can be in use by multiple programs at once with only one copy loaded.
If every application has to include a full set of libraries, then you to the winsnooze model.
How does a commercial vendor work it so that their application can run on many distributions of linux?
My ideas:
1. The application starts from a script. The script uses a library search path that checks the application library directory first.
2. The installer checks the libaries it knows that it needs. Perhaps even a 'hello world' program that uses every library call that the main program uses to see if the results are sane.
3. For those libraries it needs but aren't in the main systems libraries, it installs known working copies in the application library folder.
4. As an alternative the vendor provides an open source 'shim' library for all of it's application's calls. Because this part is open source, the translation and recompilation can be left in the hands of the groups that want it, with the vendor responding to demand by providing shims for the more common distributions.
You go to radio shack and buy a metal foil wallet. It has an outside pocket of nylon for rfid cards you don't care about. Or it has a pocket on the inside so you just open the wallet and wave it at the reader.
Or if you are really ticked, you make a transmitter that listens for RFID triggering transmissions, and responds with a blast that knocks the RFID response detector off line. (wear your metal underwear.)
If you used multiple antennas you could in principle get a bearing on the source, and useing phased array transmission blast just the source. How practical would it be to build this into a brief case?
Put it in the back of a pickup with a fiberglass canopy. Knock out every RFID detector along your route.
(Ok, ok, So you need a welding generator as a power supply. So shoot me.)
As the entire IT department for my small school, I exclusively use OS OS's for my servers: Three freebsd boxes for file server, data server, backup (at the other end of the building) and an OpenBSD box for my firewall.
Periodically I look at using freebsd or linux on the desktop, but end up turning it down, as a huge increase in complexity.
We have two applications at present that require winsnooze or mac to run. One is "The Learning Equation" which is a grades 7 through 12 set of math programs. The other is "Accelerated Reader" which basically acts as a record keeper/testing progress program for running a recreational reading program.
Accelerated Reader REQUIRES adobe acrobat reader for the reports. It doesn't require a pdf reader, it requires AAR. Further not just any version of AAR. Has to be either 5 or 7. And it does this by checking registry entries. Yes I've written to the company. It also requires flash. AR works. We have kids who never read a book before they came here now reading books for pleasure.
TLE may run under wine. Haven't tried it. Somehow I don't think that wine running on a 64 MByte 500 MHz PC will have reasonable performance. Correct me if I'm wrong.
TLE only sort of works. Most of our teachers find that it has a few strong points for illustrating concepts that are hard to do on a whiteboard, but it is ineffective at presenting new material, giving practice in using it, or assessing mastery of it.
The problem is more fundamental. Educational software is expensive. I told one software company this, and came back with surprise: "It's not even $30 per seat!"
I pointed out that their particular package could only be used for about 6 hours of instruction during a course. A textbook, on the other hand, was typically used for half an hour a day in class, and another half hour to hour for homework. The TCO of a textbook worked out to being under 25 cents per instructional hour. If we use the same 5 year schedule for software it works out to about a buck an hour, not counting time for maintenance. I told him that I want educational software that could compete with a textbook.
Worse than this, however, is that using ed software is much like bringing auxilary materials into the classroom. It makes a nice change, but you can't base a course on it. As far as I've been able to find, no one has a completely computerized math curriculum. That is, a system that would be suitable for home schooling, distance learning, or self taught student.
What OS education needs are good authoring systems for educational software -- A system that enables a good teacher to create a 'smart textbook'
Do not live in the server room.
I am partially deaf in one ear, I think due to sitting for 6 years next to the raid array for the department. Three of those years a disk had a squealing bearing. The company claimed that this wasn't sign that the disk would crash, so my boss wouldn't replace it.
Do not live in the server room.
In another job, the server room was kept at 55 degrees, with an air exchange every 60 seconds. We had hooks at the server room entry for people to keep a jacket or sweatshirt when they had to come in to work on the systems. We racks of dell 1U poweredge servers with their 4 tiny fans that sounded like jet turbines spinning up. I started wearing ear protectors in there, partly to protect my ears, partly so I could ignore people who tried to talk to me, mostly to keep my ears from getting cold.
Do not live in the server room.
In my present job, despite the air conditioning going all the time, and a fan pulling air out of the room through a hole in the wall, I've not been able to get the server room below 80F.
Do not live in the server room.
Racks: Building lifespans are typically 15-25 years before major renovations on the interior. Plan for 10 racks. Size the air conditioning for three times the initial power demand. (HVAC engineers chronically underestimate the amount of cooling AND you want enough capacity for expansion BUT 10 years from now, you'll probably be using less power per box.) You won't need them now, but you will later. If your bosses only have three-site (they lack foresite) then insist on an equipment storage area adjacent to the server room to expand into later.
Elevated floors. If you can possibly arrange it, ask for 16" elevations on the floors. During the construction, insist on the subfloor being polished smooth. One job required that I use the floor space as a crawlway to install cableing. I got really tired of rough concrete, dust, spiders, cockroaches, and bangomg my head on the cross pieces. It was only 12"
Network plumbing. Assume that the number of cables coming to the room with go up by at least a factor of 8 over the lifespan. The whole building needs to be done with mind to adaptable networking.
Run a dozen cables from the server room to your office. I don't know what you will use them for. But one job we ran 4 from the main wiring closet to the Sourcerer's Dungeon, and a year later we wanted more. (Let's see:
One connected to the firewall which lived in the closet, 1 went to the cluster of managed switches, 1 was the 'copy port' on the switch cluster so we could track down problems, 1 was the interface on the firewall that copied all firewall traffic for the IDS. Oh yeah. Then 4th floor became part of our jurisdiction too, so we needed another cluster of switches...)
Service Bench. Either in the server room, or close to it you want a place where you can pull a computer, put it at a convenient height to work on. It's a real pain to work with bits of computer on the floor, or worse, having to stand on a step ladder. Having your office in the repair room isn't bad. People will see all the cruft and clutter, and assume you are busy as hell.
Wheeled racks: You can get some increase in density by having your racks on dollies, so they can live close together, but be moved around to work on. This is what you will be doing if you only plan for 4 racks, and three years from now have 6.
Your office:
Ask for one of the fancy 6 monitor surround displays. You will probably only get 4. Meanwhile, if you don't already, get at least two on your desk.
It's very likely that you will want one or more boxes for ancillary purposes. E.g. experimenting with new OS's and such.
Remember too, that all systems change. Six months from now comes an OS upgrade. You want enough spare machines to test this out before attacking the production servers. This may mean another rack.
Give the phone people their own space. A phone person is not your employee generally. You don't want them in the server room without a keeper. (Not if security is a concern, anyway).
... In the computer lab at the school I work at. * I've got the license. * It runs ok on a 700 MHz machine with 256M ram. So far I've seen nothing of XP that makes me want to upgrade. (I know, I should be running linux. But these computers have to support several ed packages, somehow running a an application under Wine under linux on machines that aready are low on memory just doesn't hack it.)
Define destruction: This land is fire succession: In the normal cycle of events it burns to the ground every 30 to 50 years. Part of the cycle. Tarsand mining is more disruptive, because it messes up the drainage patterns some. But the companies have to put it back more or less as they found it, and they are limited in what chemicals they can leave behind.
If you have ever spent any time walking the ground in that part of Alberta it is tedius country. Once you are even a couple hundred meters away from the Athatbasca River it is so flat you can see only 50 meters or so. Much of it is either spagnum peat bog interspersed with black spruce or tamarack, or its sandy soil covered with caribou moss and lodgepole pine all in various stages of recovery from the most recent fire.
Given that Canada has tens of thousands of square miles of terrain that is indistinguishable from this, and another million or so that is very similar, sacrificing a few tens or even hundreds of square miles on a temporary basis is not an unreasonable price to pay.
For the companies do have to reclaim it. Is the reclamation perfect? No. It doesn't have the same degree of randomness as the orginal landscape. Overall it takes about 20 years.
I've also seen the waste heaps at Uranium City. El Dorado Nuclear shut down decades ago. There was a war. They needed uranium now. It was an era when we cared even less about polution. Beaverlodge lake had acid leaching into it for years from the refuse heaps of the concentrator plant. When I saw it the first time, twenty years after the plant closed, in 1975 the lake was sterile. No fish. No algae. Could see the bottom 30 feet down through clear blue water. I saw it again 20 years later. The lake now had trees growing on the margin. Grasses and reeds were colonizing some of the shallow bays.
Uranium City is on the Crackingstone Peninsula on the north shore of Lake Athabasca. It's Canadian Shield -- granite. The land has almost no carbonates in its soil so acidic waters take a very long time to neutralize. Yes despite all this, the land is recovering. Not fast, but steadily, without help.
The last mine at UC closed shortly after my first visit. Within weeks UC was a ghost town. Now about 50 people live in a town that once housed 5,000. Some of the houses were put on barges and taken to other settlements on Lake Athabasca. The rest? Some are burned -- victims of school kids pranks during the last months. Many have collapsed from the snow loads. Most of the windows are broken. A few have trees growing in the foundations. It was a sad experience as I walked through this place. I looked at the empty places, and wondered who had been there? What were their dreams? UC is now a graveyard of dreams.
The tar sands are on the edge of the shield. Mostly sandstone and glacial till. Fair amount of buffering compounds to keep the soil pH from getting too out of whack. This is land that left on it's own would heal. But the companies give it a shove along that direction.
Query: Was the land a wasteland all the way from the processing plant to the cutting edge where the equipment was? I bet not. Yet much of that land has been mined and reclaimed.
Is it zero impact? In the short scale no. In a thousand years you'll have a hard time telling it happened.
I spend several weeks each summer traveling the wild country on and near the Canadian Shield. I love that country. But I don't begrudge the islands of temporary ugliness that allow our society the raw materials that we need. Canada is not short of wilderness.
One of the things that I've come to appreciate from my trips is that with all of their foibles and faults, people are precious. After three weeks of seeing no one but other members of our group, no sign that anyone had ever been here before, no contrails across the sky of jets taking people to distant lands the discovery of a fifty year old tobacco tin where a game trail between to lakes meets the shore, is a source of wonder.
While conventional oil is nearing the peak, as is conventional natural gas, there is an awful lot of carbon available in tarsand, heavy oil, oil shale, coal bed methane, ocean bed ice methane clathrates. And we are no where near out of coal. Probably have enough fossil fuels to run the CO2 up to a percent of the air. (1% = 10000 ppm In the past two centuries I think the levels have risen from about 330 ppm to about 500 ppm.)
I left a sysadmin job at a university for a sysadmin job in private industry, for a 30% raise in pay.
However the raise in pay was accompanied by: A 75 minute commute; Only 3 weeks holiday/year; no banking of overtime;
life in a dilbert box, and three signatures required to buy a $100 ethernet card.
I left that position after a year and a half, taking a job at a local school as a sysadmin for a bit
more than half the pay. Curiously, we found that the transition to half the money wasn't nearly as hard as we expected. We weren't putting as much into retirement plans, but between saving $6000 per year on the commute, and that lower salary is taxed at a lower rate, we have pretty much the same life style.
But I now had 12 weeks holiday per year, a 15 minute commute, 4 weeks per year where they paid me to go with
the kids on canoe trips.
In my experience most jobs dealing with computers turn into infinite time sinks. You will be spending at least half your waking time at whichever of the two jobs you choose. Go for the one you will like.
Some years ago at one of the usenix conferences someone reported a similar project. However, instead of trying for chords and harmony, they chose natural sounds -- frogs croaking, crickets, bird chirps, water, wind. Their take: As former hunter gatherers we detect change in this environment more quickly, and few combinations are dissonant, further, the 'normal' environment swiftly becomes non-distracting.
So, for example, disk activity can be represented by the sound of bees. Incoming mail to the server by a chickadee chirp, successful network logons by a duck quack, non-succesful ones by a piliated woodpecker call, disk full warnings by various levels of thunder, depending on how full and how fast they were filling, network activity as wind in trees noise, Internet access as rushing water, cpu usage by frog choruses.
Implementation:
Since this would require more than just vmstat as it's input, I would suggest a that it be implemented as a simple udp socket. Various programs then can run at various places on your network, and send packets to the socket, with an event type, and whatever addtional info you wish. Possibly the same mechanism that Big Brother uses for the data layer.
So...
Use existing BB modules to gather information.
Send to central receiver.
Receiver does two things:
A: Writes the BB webpage
B: sends an event to the sound generator.
Sound gnerator uses the event to set some parameter of the sound environment. This environmental sound can
be a single short event, such as a bird chirp, or can set some level for reporting a continous event.
Sound generator has a library of sounds, either some form of sound clips, or parameters to hand to external programs.