Here's a thought I had, but probably will never get around to building.
Lots of people go to the expense and effort of building/buying radiators or using large tanks of water as the heatsink for their water-based CPU cooler systems.
Last year, I started measuring the temperature of the water in my toilet tank. After a flush, it drops to 5-6 degrees Celsius. Between flushes, it gradually reaches room temperature, of course, but this is still no worse than a radiator or bucket. In practice, however, it never actually gets above about 10C (while room temperature is about 20C).
In other words, it's a supply of cold water which you were going to simply flush away.
Place a small bucket inside the toilet tank. Put a submersible pump in there, run the water to the CPU coolers, bring the water back and drain it over the bucket in the tank.
Everytime you flush the 6 beers you went through while flaming me for my Linux isn't ready for the desktop article, you can rest assured that the water which cools your CPU is being replaced with fresh, cold water. No mold, no mildew.
The purpose of putting the pump in the bucket is so that there's always a supply of water for the pump, even during the flush. And the purpose of draining the return line over the bucket is so that if your toilet tank doesn't refill for some reason, you'll still keep your bucket full of water and buy some time for hardware monitors to shut the system down if it's getting too warm.
I don't know how hot the water in the toilet will get, but think about this:
The bucket full of water in the toilet tank is replaced during each flush but isn't actually available for a flush. You'll save water.
You'll be removing the CPU-heated water from the house and will therefore reduce the load on your air conditioning system.
You get to piss on the scourge of the overclocker, that excess CPU heat.
Warming liquids enhances their ability to dissolve things, including...dark matter. You might have to clean the toilet less often.
Of course, the only thing I'd worry about is the quality of the submersible pump. After all, if water leaked into the pump, then the water in the toilet could come into contact with one side of the AC line... the other side of which is grounded to your fusebox. If you happened to touch another grounded object while urinating (concrete floor, sink faucet, etc), then enough current could find that your stream of urine and urethral tissues are a more attractive ground path than the plastic sewer pipe. I think I'd invest in an isolation transformer (search ebay) to reduce the risk of highly...unpleasant... damage.
I think if one were pumping water through tubes soldered to the heatsinks of their power supply, the risks would be compounded, conceivably by a failure on the primary side of the power supply: I think I'd make a point of running the computer on an isolation transformer as well.
* One dragster's 500-inch Hemi makes more horsepower then the first 8
rows at Daytona
"500 inch" is a common expression meaning 500 CUBIC inches. Which is approximately equivalent to a metric car having an 8.3L engine.
Note that in the 1970s, Cadillac put out street vehicles (mostly Eldorados) with a 500 cubic inch (8.3L) V8 engine.
Granted, they were substantially milder than a Top-Fuel class derivative of the Chrysler Hemi, but it certainly should make any imbecile who thinks that Honda invented performance cars think twice before entering into a stoplight confrontation with a real car.
You *don't* need LEGISLATION to fix this problem (isn't that what technology is for?).
Especially since the legislation will do nothing.
Here goes: So (a) if a law like the one I propose is passed on a national level, and (b) it does not substantially reduce the level of spam, then (c) I will resign my job.
The problem is it's being addressed on a national level. That won't stop the African scam artists "whose money is tied up" - hopefully their oppressors will beat them in the face with a rusty camshaft - or the Chinese wishes of good fortune and prosperity that I was continually getting from some shitty company selling latex products until I finally decided to blackhole China from my mailserver.
This might keep the Florida 21-year-old unwed mother of 6 children from spamming me from her dial-up ISP of the week. But the funny thing about national laws is that they don't apply outside the nation...
If you're going to go to all that trouble, you may as well wire your water cooler into the supply line of the toilet: the tank fill pipe draws from your water reservoir, which draws from your water supply. Add a cutoff valve in the event that your water is cut off and you're done.
Are you suggesting that I pressurize the whole system to the water supply? That sounds dangerous to precious silicon. If you mean after the float valve, it's the same thing - inherent resistance in the line will result in a pressure gradient, and will probably put more pressure on the system than a fountain pump. Low pressure, high volume is the key to avoiding leaks while maintaining sufficient flow to transfer heat.
Of course this all smacks of the sort of thing a teenager would do to his honda - expensive, failure prone, and mostly useless.
Actually, if your computers are located near a bathroom, I see it as far less impractical than having a bucket or a fan-cooled automotive heater core parked adjacent. This lowers initial and operating cost, reduces noise of fans, and provides a greater temperature gradient to cool the processor(s). After all, subterranian water should be cooler than room temperature, and most people get their water from underground pipes.
I agree about what teenagers do to their Hondas. But the striking difference is that this is clearly not meant to impress bubble-headed boy-band crazy 16-year-old girls, nor does it say, "Yo, homeboy, I like to pretend that I've got money even though I'm driving a four cylinder economy shitbox that I think is a racecar".
This says, "Hi! I keep a case of plutonium under my bed so that I can fuel up my modded DeLorean."
Now, if you could run the hot water through a hollow toilet seat for those cold winter mornings, then you will have something useful going.
Oooh! Thank you!...Now how am I gonna cut a water jacket into a plastic toilet seat... Maybe a stainless steel industrial toilet seat with stainless tubing welded to the underside? Hmmm...
Here's a thought I had, but probably will never get around to building.
Lots of people go to the expense and effort of building/buying radiators or using large tanks of water as the heatsink for their water-based CPU cooler systems.
Last year, I started measuring the temperature of the water in my toilet tank. After a flush, it drops to 5-6 degrees Celsius. Between flushes, it gradually reaches room temperature, of course, but this is still no worse than a radiator or bucket. In practice, however, it never actually gets above about 10C (while room temperature is about 20C).
In other words, it's a supply of cold water which you were going to simply flush away.
Place a small bucket inside the toilet tank. Put a submersible pump in there, run the water to the CPU coolers, bring the water back and drain it over the bucket in the tank.
Everytime you flush the 6 beers you went through while flaming me for my Linux isn't ready for the desktop article, you can rest assured that the water which cools your CPU is being replaced with fresh, cold water. No mold, no mildew.
The purpose of putting the pump in the bucket is so that there's always a supply of water for the pump, even during the flush. And the purpose of draining the return line over the bucket is so that if your toilet tank doesn't refill for some reason, you'll still keep your bucket full of water and buy some time for hardware monitors to shut the system down if it's getting too warm.
I don't know how hot the water in the toilet will get, but think about this:
The bucket full of water in the toilet tank is replaced during each flush but isn't actually available for a flush. You'll save water.
You'll be removing the CPU-heated water from the house and will therefore reduce the load on your air conditioning system.
You get to piss on the scourge of the overclocker, that excess CPU heat.
Warming liquids enhances their ability to dissolve things, including...dark matter. You might have to clean the toilet less often.
Of course, the only thing I'd worry about is the quality of the submersible pump. After all, if water leaked into the pump, then the water in the toilet could come into contact with one side of the AC line... the other side of which is grounded to your fusebox. If you happened to touch another grounded object while urinating (concrete floor, sink faucet, etc), then enough current could find that your stream of urine and urethral tissues are a more attractive ground path than the plastic sewer pipe. I think I'd invest in an isolation transformer (search ebay) to reduce the risk of highly...unpleasant... damage.
The cabinet is in excellent condition, even the PET label just under the monitor. Haven't fired it up to see if it works, but there's an aftermarket accelerator/RAM expansion board resting on top of the motherboard right now - it looks complete but the expansion board is just *resting* on the motherboard, like someone tried to fix or upgrade it once. I have a suspicion that the machine is fine but the attempt was along the lines of "what do you mean I can't put a PCI video card into that?". FOB Ottawa, Canada.
Schematics would be cool so that I can sell it (or give it away if there are no good offers) as a working unit.
My own collection of old TI-99/4A, Amiga 1000, Vectrex and Coleco Telstar Alpha machines already occupies quite enough room, thank you very much. And I must confess that I haven't fired up even one of my prized TI-99/4A machines in over a decade.
Powerplants always have pretty interesting tours. Try and visit a couple of different types.. coal, nuclear, hydro, etc.
Research reactors are good, too. Atomic Energy of Canada Limited (AECL) used to offer tours of three nuclear reactors just 3 hours up the road from Ottawa, in Chalk River.
You were actually allowed to stand on the NRX nuclear reactor there, while it was running, and look down into the calandria to see the pretty blue glow. It was nice and warm up there, with a thrumming under your feet from all the pumps and support equipment running.
Of course, all tourists were outfitted with dosimeters and screened thoroughly before and after.
I worked there as a co-op student one summer, it was great fun. The NRX was my favorite, but I think it's decommissioned now.
That is a pretty broad statement to make. I played Wolf 3D when I was 9, 10 years old, Doom after that, Quake after that, so on. I knew full well it was a game. It was fun dammit! What makes you presume you know the maturity level of any given kid?
Exactly. As a kid, I played Popeye on my TI-99/4A all the time. The Old Hag threw bottles at me, and I learned how to punch them away. And look at how I turned out. I'm posting an unsolicited comment to a geek website at quarter to four in the morning while smoking a cigarette and waiting to take my next Prozac. Overall, I'm better adjusted than most people I know.
Hold on just a sec... Fscking cat won't stop meowing to be let out... SHIT! No more empty beer bottles! [digging around, lobs an empty rum bottle, makes a mental note to wear shoes when going to the bathroom after bedtime]
Also, we are talking EM radiation here, not nuclear. There is a difference.
Yeah, it's unclear what the original poster thinks that he's going to be getting "nuked" by.
Actually, several kinds of radiation leave a monitor, though it's only the last two that tend to cause worry:
Light - this is, after all, why you bought the monitor?
Heat.
Sound - ultrasonic and sonic, ranging from the ultrasonic whine of the power supply's inverter and the horizontal windings of the deflection yoke and the flyback transformer, to the rasp of the vertical deflection yoke windings, to the click of various relays as the sync rate changes.
Electromagnetic - primarily from the horizontal and vertical deflection circuits, these behave basically as VLF radio transmitters. Their actual effects on the human body are unknown, but the fact that the word "radiation" is used to describe them is enough to get every do-gooder high-school-educated hausfrau or trade union member up in arms. When a monitor says "Low Radiation", this is what is meant, but it's not what I worry about.
X-rays - X-rays are really a kind of electromagnetic radiation in theory, but in practice, they're quite different. They're ionizing, meaning that they cause a charge in something when they strike it. In a CRT, a very high voltage power supply is used to accelerate beam of electrons to an appreciable percentage of the speed of light, and then they're slammed into the phosphor coating on the inside of the screen. If you compare this with the theory of operation of an X-ray machine, you might be amused to note the similarities. The vacuum inside the CRT is only part of the reason why the glass is so thick... there have been stringent X-ray standards since the early 1960s, when most color TV sets like the RCA CT-100 could cloud photographic film. Of course, a failure in the high voltage regulator circuit of a modern monitor or TV could cause it to spoil all your vacation photos.
X-rays are ionizing radiation and are known to be carcinogenic and mutagenic. They're essentially man-made gamma rays. Geiger counters do tend to count them, though it depends on their energy.
The idea that Linux is harder to install than Windows has reached the status of an urban legend (or Microsoft FUD) -- this is 2003, not 1997.
I agree in almost every way with what you've said.
ISA hardware support is lacking, but fortunately that problem becomes less crucial with each day.
Some hardware remains poorly supported, especially video cards and winmodems. They're reality, and we have to keep up the pressure on the manufacturers to release information on programming device drivers for their stuff.
Software installation remains an issue. Debian and Red Hat deserve praise for these, but they're far less foolproof than the installers shipped with most Windows software. When was the last time that a Windows installer whined at you that you lacked a certain DLL? For the most part, they're simply installed. Maybe Linux applications need to be able to (?download and?) install their own libraries, independently of the operating system, and without user intervention. We have to find a way around the dependency hell which you encounter so frequently. And no, Joe Sixpack is not likely to compile his own stuff.
But the Windows comfort is more along the lines of familiarity with the locations of things, with how everything works, pitfalls to avoid, and familiarity with applications. Users will (and do!) crave that when moved to the Brave New World of Linux. We can address some of it, but much of it is inherent to changing operating systems. We have to be conscious of the fact that it's a force beconing many convertees back to Windows.
I'm actually an advocate of linux on the desktop (yes I am) and it seems those points you mentionned don't make much sense, here's why.
I'm an advocate of Linux on the desktop, too. I can't wait to see it. But from the perspective of an *average user*, I'm still convinced that it isn't ready. We *need* the average user to feel *more* comfortable working at a Linux desktop than a Windows machine, especially since he's gonna have to deal with a Windows world trying to suck him back into its comfortable embrace... at least until we've finished the takeover of the desktop.
- Linux GUIs are faster and faster at each version. Gnome2 for example was totally re-coded with performance in mind and behaves much better now, KDE 3.1 (still a release candidate but still) on this box is working SO much faster than XP did on the SAME box!
This is good. I cannot corroborate it using a pre-compiled distro. Why am I using a pre-compiled distro? Because that's what Joe Sixpack is gonna be using. More optimization is needed, and more carefully made binaries are required from the major distros; especially Red Hat in the current #1 off-the-shelf position.
I think part of the problem is that we need developers to try actually using the pre-compiled binaries of their works which end up being shipped with the Red Hats and the Mandrakes of the world.
- Since I've been running linux on my desktop, I have not yet had one problem reading any PPT, DOC, etc... documents... not once... sorry. And I get a lot of ppt and doc files sent to me daily
Most of them have formatting problems, cannot handle inline images (properly or at all). Table support from Word 2000 is lacking. I know this is a serious pain in the ass to reverse engineer, but it merely frustrates end-users who are already gonna be pissed off about having to learn something new when their company moves to Linux.
- I have had problems with some applications, contacted the mailing list, and the solution was sent to me a few minutes later... no RTFM.
You're not Joe Sixpack. "How come it says I cannot save my file in/bin? Huh? I didn't log in as root, whoever that is." Screams of RTFM or "Get a life" would abound on mailing lists or IRC, whereas a 1-900-DRONE would calmly answer, explain, and the user would be supported. Sure, the example I cited is an operating system issue instead of an application issue, but it's a problem every bit as simple, stupid and pervasive.
- I use Evolution for my email/calendar/tasklist/contact management stuff, it has everything I could ever use and more... I have used kmail in the past, I've never had any real problem with it.
KMail is great, the only programming complaint I've had with it is that it silently dies if it runs out of disk space. But the spellchecker is right out of 1995. We have to match feature-for-feature to be adopted. You're not going to sell Linux/KDE (or Linux/Gnome or OpenBSD/AfterStep or whatever) by screaming from the hilltops, "ALL THE FEATURES OF WINDOWS 3.1!" in a Windows XP world.
Evolution was too slow to be usable on my PIII-500. That's insane. It's just an e-mail client, not a genome sequencer, for Gawd's sake!
- Recent linux distributions based on more recent and less backward-compatible glibc usually have some kind of package management system that will not only save you from searching on freshmeat, but also install directly the application for you.
emerge gnucash
apt-get install gnucash
synaptic->gnucash
and so on... You have now installed the latest version of an excellent financial software, which, may I add, will read files from other windows software like Quickbook or Quicken without a glitch
It's a good start, yes.
But the biggest problem is that if a feature which an Excel user would take for granted is lacking, it's a negative perception. Most users will already resist the change to something new and "strange".
We've grown up with the idea of piping the output from one program to another; it's the Unix way. But it's *not* acceptable on a desktop system. You don't do your spreadsheet in OpenOffice Calc, then save it in some format that Gnumeric handles so that you can use the point-and-click data analysis tools, then open int up in OpenOffice again. If you're paying a secretary $20/hr to do this, it doesn't take more that a few months to make back what you would have spent to install Windows on the machine.
- I use daily applications for all my needs, none of them are poorly written at all. licq is stable as a rock, xmms plays music just perfectly, evolution still handles my emails (without a virus or worm or anything like that infesting my computer), mozilla works like a charm and KDE 3.1 is just a dream. Although all those applications work in a much superior fashion than equivalent applications on windows, they ARE skinnable indeed:)
I don't know how well Evolution handles e-mail. My main machine is over the hill, but easily captures video from my TV card in real-time. I find it hard to believe that responding to e-mail in Evolution should require such a fast computer as to be unusable on a machine which will capture NTSC video at 29.97FPS with 16 bit stereo sound with 0 dropped frames... (unless I open Evolution while I'm capturing video).
Mozilla is great. It's fast, attractive, and it works well. The only problems I have with it are fault tolerance (delete your JRE without telling Mozilla, then try to use a website infected with applets; it crashes with no warning), lack of ability to send a mailto: link to anything other than Mozilla's mail client, and the inability to tailor the browser string to be whatever I want without recompiling (at least one website I *have* to use will ban you if your browser doesn't say "MSIE" in its string).
- Companies such as the Kompany, RedHat, Suse, etc... actually DO have some marketing people that make your desktop look just like you want it to look like as a user and to behave.
My desktop right now looks simply amazing, yet is really fast and everything is at hand. My girlfriend uses it every time she comes, all my friends really love the way it's set up and even my mom used it and didn't have a problem doing everything she needed to do.
For sure. This is a good step. But part of the problem is with the overall look of it. Red Hat 7.3, for example, with probably the biggest marketing department in the Linux world, comes with a highly saturated eye-straining blue background.
Contrast this to the relatively neutral backgrounds of Windows and Mac environments, and it looks more like we're trying to sell a product than design something useful out of the box.
Even XP's default meadow is less eye-straining.
If some Joe Sixpacks can't figure out how to move the Windows taskbar to someplace they like better, do you really think they'll change the backgrounds and skins to something less displeasing? The desktop's defaults must be *neutral*, *inoffensive* and *non-eyestrain-inducing* out of the box with *every* distribution.
- and for the support thing, companies like Suse, RedHat, Mandrake, etc... DO offer commercial (cheap) support for pretty much all the applications shipped with their distributions, in fact, and I speak from experience, these companies go way beyond that by helping out users with applications not "officially" supported, and also collect bug-reports and offer patches to the original developer of the software to fix the problem for them (http://www.redhat.com/bugzilla) for example.
Who do you call when you need support with OpenOffice or xine? I haven't tried either; I've got the luxury of being able to pursue the source code.
I do know that at one of my former employers - a huge defense contractor staffed by engineers and computer scientists - we spent a lot of our IT budget on calls to Microsoft looking for support on how to create PowerPoint slides with embedded video and other dead-easy things like that.
Sucky as that may be, it's reality for lots of organizations. We have to address that.
- Whoever wrote that has NO idea of how much a business license for Microsoft Windows costs... it's not even close to $200. Tell this person to add many zeros to that number.
Sorry. $299, according to the Microsoft website, for Windows XP Professional, in single units, as a standalone operating system instead of an upgrade.
It remains that the purchase price is a very, very small part of the total cost of ownership.
I think linux is still very young on the desktop OS market but it's doing a great job and I'm very impressed by how fast it's moving forward...
Linux has made amazing strides since its inception a mere 10 years ago. It's already a secure and stable server operating system, with mature tools for sysadmins.
But it's still at workstation space. We can take heart; it's more usable on the desktop than a $30,000 Sun workstation, but it's still not ready to supplant Windows yet.
The biggest obstacles are not the Linux kernel, or even Linux itself, of course. The obstacles are a fast, feature-filled and stable desktop metaphore (be it KDE or Gnome or whatever) with good *USER* applications readily available. (Don't even bother sending me flames telling me that vi is the greatest word processor ever made because Joe Sixpack isn't gonna even gonna figure out how to bring up the help screen.)
KDE, Gnome, Evolution, OpenOffice, etc... all these software are working on a new development version right now that's purely amazing... I can't wait to see what it will be like by the end of the year 2003!
I can't wait to see what it's like 20 years from now.
I've been waiting 15 years to see the end of Windows.
I'm not a huge advocate of Linux on the desktop (yet)
I want to be, but I can't (yet). [grin]
Here's the problem:
To put Linux on the desktop, we're asking them to give up the comfort, familiarity and applications of Windows. For what benefits?
A user interface which is slow, designed by computer geeks for what *we* like, rather than designed by marketing departments for what *the public* likes, and usually ships, by default, with color schemes which are somehow even more garish and offensive than Windows XP.
Inconsistent support. If Joe Sixpack were to look for support on a Linux program, usually there's no 1-900 number. If he were to dig up the mailing list info and send in a question, how long would it be before someone says "RTFM!"? What's he gonna do when TFM is half-written or poorly translated from some strange Tibetan dialect?
Poor applications. Quoting an e-mail I received: "But a lot of it - and mainly the GUI stuff - is still lagging behind, being a slower and buggier version of a half-decent program on Windows. And priorities are wonderful - when we build a GUI application, the most important thing is that it's skinnable. Bugs? Features? Competition? Who cares?! It's skinnable!"
The same writer continues... "And for the biggest question: Mr. Rupert wants a financial software for Linux (his son installed it for him). So he calls his son over to install a simple financial software - just something which can calculate his loan repayments. His son opens google (or freshmeat), and finds 31 financial programs. Each has a different set of features, of course. He downloads and compiles each of them (ah, yes, the rpm was compiled using an ancient glibc version, and no, Mr. Rupert doesn't know what glibc is). The only two candidates which could actually be compiled (and didn't require libobscure.so.2) and actually have this option in their ugly programmer-designed-GUI menus die as soon as you choose the option. That's right - the operating system is stable as a rock, but the programs die immediately. What's Mr. Rupert going to use? hmm.... Maybe a respectable program from a respectable company (on Windows, of course).
But wait! John Rupert (the little 15 year old) can program - he's got some C tutorials, and he's written a few small programs. Why can't he write the program for his father? And the 32nd version is on its way."
Good stability and core networking and filesystems. (Joe Sixpack really seems to care about this, after all, he's still running Windows 98 with FAT32. But he's happy, 'cause it's 98SE.)
Free to download, cheap to buy. Ahh, but if you're in business, you're paying people to use computers. You're paying people to surf the 'Net and try to figure out why OpenOffice Calc won't do the polynomial regression that Excel 95 and up will do in two mouseclicks. You're paying people to punch Ignore/Ignore/Ignore as KMail chews through an e-mail with the names of people it doesn't recognize, rather than quietly underlining them so that you may passively ignore them. You're paying people to wait 1/2 hour as KDE parses a directory full of JPG images of the latest marketing brochures. Suddenly, the $200 or whatever Microsoft is currently charging for Windows is pretty unimportant.
An ordeal every time someone sends you a Microsoft Office file. These are basically standard in the business world, and while you expect this to be a problem with an alternative desktop, it's incredible how pervasive the damned things are. Are you gonna tell a potential employer to re-send his offer of employment in HTML because you can't read a Word file properly? Wouldn't it be even worse if you were a large company dealing with clients who sent you stuff in XLS, PPT, DOC?
Sorry. But this is simultaneously an amazing project and an act of butchery.
A few words of advice for anyone trying this themselves: use electronics solder- it doesn't stick to breadboard. I later got curious and tested regular solder on an old NIC and it stuck everywhere. Watch what you buy.
"Regular solder"? Is that plumber's solder, like you'd use to sweat two pipes together?
Solder includes chemicals (flux) which help to clean the pieces of metal which are being attached. Electronic solder is either rosin core or organic core; they're fairly gentle. Acid core is used by plumbers and the post-soldering remains of the flux attack electronic components over the years.
Based on some of the pictures, I believe that the soldering to the PC board was done with an overly-large soldering iron. A fine grounded-tip 15-25W soldering iron is *essential*. My favorites are the tiny little Ungar irons from just before Weller bought them out.
BTW, it *is* possible to solder to aluminum, but it's very different. For one thing, aluminum coats itself almost instantly with a very fine layer of very hard aluminum oxide. Solder will not stick to this layer. The other problem is that aluminum conducts heat away from the attempted soldered connection. A large-wattage iron with a sharpened tip will do the trick. Apply a puddle of solder and, scrape the aluminum under the puddle with the tip of the hot iron. A bond will form, and the strength will be mostly dependent on how well you scraped the aluminum under the puddle.
Gaffer tape has a tendency to dry out, shrink and peel off over time. This limits the longevity of this particular modification. Of course, you probably don't want to cut up the typewriter since you chose it because you like it, and they don't make 'em anymore, so I applaud the reversibility.
Rather than attempting to make a distinct "switch" for each button, why not simply have a scheme where each keypress will ground a wire to the crossbar? Of course, that won't work with the keyboard's matrix arrangement, but that can be easily solved in a minute.
PC board looks like a good way to insulate parts of your switch assemblies, since it's cheap, readily-available, easy to work with, and you can use it to make narrow switches.
Onto each one of the levers (which are steel) you could solder a small piece of printed circuit board with a custom pattern. (You can buy printed circuit board etching supplies at Radio Shack.) One part of this tiny board's pattern is used to solder it to the lever. Then a gap, with no copper on the board. The opposite end of this pattern is a place to which you can solder a flexible wire and a small fold of some springy nickel-plated steel.
The tiny board then rides with the lever and the new springy-steel contact then connects the wire to the crossbar.
As for connecting the keyboard to the matrix, my first thought is to use optocouplers. Optocouplers are merely an LED and a photocell built into the same case. They're meant to isolate different parts of electronic circuits.
Solder a piece of flexible wire to the crossbar, and connect that directly to the keyboard's ground on the PS/2 or DIN connector. (You can get the pinout from the Internet.) Take the +5V lead from the keyboard and put about a 500 ohm resistor in series with it, then carry that across to the anodes of the LEDs in the optocouplers. Connect each cathode to each wire coming off the lever boards. Now, when you hit a key, the LED in the corresponding optocoupler should be lit.
The photocell in an optocoupler is actually a kind of transistor, which is essentially an electronic switch. Connect each optocoupler to its corresponding pair of contacts on the keyboard's matrix. Make sure you get the polarity right, a quick check of the keyboard with a voltmeter should do it.
Choose an optocoupler with a good transfer characteristic; probably Darlington-outputs. LEDs need current limiting, and that's what the 500 ohm resistor is for. Now, there's ONE resistor being used to power all those LEDs in the optocoupler, and it limits the single or total LED current to 10mA. This is done because motherboards often have fuses to keep keyboard current below about 50mA. Even if you hold down all the keys on the keyboard, you will not draw more than 10mA. Blowing the keyboard fuse on your motherboard would suck.
Now, when the Smithsonian comes calling and wants your typewriter because FDR used to keep it in the trunk of the Presidential Limo? Desolder the wire from the crossbar, desolder all the little pieces of printed circuit board from the levers, and the typewriter is undamaged.
Yeah Hard drive noise is bad but it's nothing like it used to be. I remember some of the older drives I had that made so much noise it scared away my cat.
Still, there's nothing like the seismic rumble of one of the good old full-height 5.25" 20 megabyte Seagates starting up. And the squeak-squeak sounds of band-stepper actuators.
Ahhh... remember the good old days where you always let the drive warm up for 20 minutes before you saved anything, because the old actuators didn't account for expansion and contraction of the platters and arm?
I must resurrect one of those just for the fun of being able to stick, like, 4 MP3s on a drive which draws 12V @ 6A at startup.
Of course I used to buy crappy hardware so the drives were bad to begin with.
There was a company called Kalok which was producing mega-cheap hard disk drives in about 1995, before they got bought out by an even more fledgling JTS. They had a 100 megabyte 3.5" hard disk drive which was selling for about $75-$100 less than a comparable name-brand drive. 'Course, there was a catch... in fact, two of them.
The Kalok had a band-stepper actuator - which is impressive because everything else from 40 megabytes and up seemed to have a voice-coil actuator. Needless to say, installing Windows 95 on a Kalok hard drive was a bad idea, since the system wrote to boot up logs and the like during startup - before the hard drive had a chance to warm up.
The other catch reads like a bad joke: The hard drives were made in India.
You are assuming a non-catastrophic failure. What if you are doing 80mph on the freeway and your front tire (which has been running on low pressure for 2 months) blows out.
Right there, you are doing 2 things which directly contribute to the accident, speeding and low tire pressure.
Posted speed limits on some Michigan freeways were 75MPH, last time I was there. 80MPH is within the +/-10% accuracy that is mostly assumed of automobile speedometers, so it's unlikely that you'd be pulled over for doing 80 in a 75MPH zone.
*HOWEVER*, only 20 years ago, every freeway gas station had a garage, well-stocked with tires.
Tires have made tremendous leaps in their technology in the past 20 years. In 1982, blow-outs were common, and in 1972, even more so.
Back then, any driver knew how to handle the car when a tire blew out, and would manage to get it off to the side of the road without killing anyone else.
What we're saying is that a mere 20 years after these events were last commonplace, the motoring public's incompetence is such that it has now turned what used to be a nuisance failure into a deadly failure.
Look at the Ford Explorers on Firestone ATX P235-75/R15 tires. There's obviously a problem with that combination of tires and suspension design. But the failure is a *tread separation*. The tire merely loses the outer inch or so of rubber, this is a far less catastrophic failure than the blow-outs of bias-ply tires on the cars of the 1970s and before. And yet people apparently have no idea how to control their vehicles when this happens.
It's pathetic.
If you cannot control your vehicle through a mere tire blowout (let alone something as trivial as a tread separation), you shouldn't be allowed on the road.
Should you be held accountable for that? What if someone dies in the accident? Does the family of the victim deserve to know that you killed their loved one? Do the cops deserve to have access to information that could prove you commited a crime which caused an accident?
Being unable to control a vehicle through a tire blowout is negligence. There's no excuse for not being able to handle the car. No part of any vehicle is fail-safe - especially not tires - and you have to be competent at coping with at least small failures like tires.
People need to take their cars more seriously. They need to learn how to drive, not by dad, but by people who are professional drivers. They need to know that the oil, water, wiper fluid, tire pressure, tread, and a myrid of other things are important.
I agree. There should also be a mechanical proficiency requirement to getting a driver's license. People need to know how wheel bearings work so that they know when a grinding sound is a potentially deadly failure. They need to know how brakes work so they can appreciate the dangers of the clear oily patch on the driveway. They need to know how their steering and suspension systems work so that they don't fuck around when the steering wheel play is getting excessive because a ball joint is getting loose enough to separate.
My animated paperclip went on a bender and refuses to speak to me.
The last time I heard mine was a dying scream as I mounted my FAT32 partition, navigated to it, and typed the magic letters:
# rm -rf *
It was high, blood-curdling, but strangely satisfying. Like the sound of the welds in a Honda's body popping as the car crusher takes it down to 3 apples tall, then the wet thunk of a cast-aluminum engine block cracking like a flowerpot in a vise.
Mercifully, when I had to install Excel on Wine because OpenOffice doesn't do something as fscking simple as a polynomial regression, the damned paperclip didn't work.
Of course Microsoft is their own worst enemy. Who else would allow IIS or Outlook - a security hole which masquerades as an e-mail client - to be some of their flagship products?
The security holes are even more annoying than the damned animated paperclip.
Hunt the Wumpus for the TI-99/4 and TI-99/4A
on
Top Ten Shameful Games
·
· Score: 5, Funny
But why all the "worst games" lists?
The usual hype. But I was surprised by the lack of one hugely important game:
The TI-99/4A (and its rare older brother, the TI-99/4) had a 16 bit TMS9900 processor chip (in 1979 and 1981, boys and girls!), a kick-butt video chip (the TMS9918) which had 32 sprites and a video overlay feature. But Texas Instruments, a company which is/was making more chips than Frito-Lay, hobbled the machine by using the video chip's RAM as the console's main memory, bottlenecking the expanded memory down to 8 bits, and creating the single slowest BASIC interpreter ever designed by having it interpreted TWICE (from BASIC to GPL - "Graphics Programming Language" - then to machine language).
Research indicates that during the first year of life an uncircumcised male infant has at most about a 1 in 100 chance of developing a UTI, while a circumcised male has about a 1 in 1000 chance.
Studies conclude that the risk of an uncircumcised man developing penile cancer is more than three-fold that of a circumcised man.
Some research suggests that circumcised men may be at a reduced risk for developing syphilis and HIV infections. However, the AAP policy states that behavioral factors continue to be far more important in determining a person's risk of contracting sexually transmitted diseases than circumcision status. (I have never once in my life had sex without a condom. However, if we're talking about HIV, I'll keep every advantage in my corner that I can, thank you very much.)
Considerable new evidence shows that newborns circumcised without analgesia experience pain and stress measured by changes in heart rate, blood pressure, oxygen saturation and cortisol levels. Other studies suggest that the circumcision experience may cause infants to respond more strongly to pain of future immunization than those who are uncircumcised. (Note the use of the word "infants" in the phrase "infants to respond more strongly". This indicates that the adverse effect is limited to infancy, and does not imply future harm. Otherwise, by extension, almost all North American men would shrink from a paper cut for fear of reliving their circumcisions.)
Quoting from one of your previous posts:
Now I can build a virtual 3D woman on my linux system. THANK YOU. I have had sex only once in my life, and that was when I was in mid 20's (No Joke). I LOVE technology!
[sigh] Yet another NOHARMM or NOCIRC wacko. Let me guess: You're fat, bald, lazy and ugly. Your life sucks, but you're absolutely convinced that you'd be as dashing, financially and sexually successful as James Bond, if only you'd been allowed to keep your foreskin.
Even today, I believe that circumcisions are performed without anaesthesia. I remember seeing a news story a few years ago about a circumcision conference. The doctors there had previously thought that babies could not feel pain (?!) but they were beginning to reconsider this claim..
Generally, it's a topical anaesthetic used now. (topical: applied to the skin as a lotion, rather than injected)
And while it seems inhumane, consider:
Despite lots of scientific research by the AAP and a few other groups, there's been no evidence whatsoever that the pain is processed by a baby the same way we do, nor is there any evidence that it causes any harm.
I've yet to meet anyone *credible* who had it done at birth and remembers it.
General anaesthetics are extremely dangerous and expensive to administer on babies.
Circumcision reduces HIV infection rates, absolutely obliterates penile cancer, and has no scientifically provable effect on enjoyment of sex or comfort in daily living. These benefits are scientifically provable, whereas the allegations that it is traumatic are not. If you reject that, then you reject the entire scientific basis for medicine.
The Jews have been doing it without anaesthetic for over 5,750 years. They tend to be extremely family-oriented, hard-working, successful and are still quite capable of having a good time. So if that's the result of a little infant trauma, sign me up.
My own circumcision was only moderately more unpleasant than having a dentist do a filling. I dreaded my circumcision more than I dreaded having my wisdom teeth out - in reality, the wisdom teeth were far more unpleasant, even though I got Percocet.
If my foreskin were to grow back tomorrow, I'd be on the telephone to the Mohel immediately. I would do it over again in a heartbeat. Knowing what I know now, if I could go back in time and have it done at birth, I would.
Ignoring the rest of your comment because it looks like it could easily become a circular argument, let me just stick with your single most telling comment:
Hmmm, I just loaded up my directory of MP3s and Oggs. I used the xmms "+DIR" button to add the entire directory of 2808 songs. It took 62 seconds. Over NFS on 100mbit ethernet. From a 200MHz pentium with PIO mode 4 disks. Moreover it took less than a second to sort the entire list by filename. And I haven't been listening to my entire music collection this week (Christmas music)- so except for 25-some songs, none of that was in the buffer cache of *either* machine.
Okay. What you're suggesting is that I use another application (xmms) as my file browser, because the file browser built into the leading desktop metaphor is so krufty. You're advocating a work-around which bypasses one of the most fundamental features of KDE, and you apparently don't see a problem with that.
Even so, the speed is too slow. In my case, it does take me over 2 minutes to use the +DIR button, and that's using a PIII-500 with 512 megs of RAM and the MP3 collection being on a separate drive from/. Both drives (/ and/mnt/mp3) are running at UDMA/66 with ext2fs. There's no NFS involved, the MP3 collection is local to that machine.
One way or another, I'm not sure how your workaround would upset my expectations of being able to choose a song and play it immediately when I feel like it.
You see, from my experimentation with your workaround here, it seems that I would have to open xmms, choose +DIR, then wait 2 minutes before I can scroll through and choose the song I wish to listen to.
The alternative is, of course, to save the list as a playlist, but opening large playlists takes as much time as using +DIR. And the list gets stale; music which I've just downloaded isn't available in my playlists until I've updated them - which takes 2 minutes.
By contrast, I'll move over to my Windows 2000 machine (PII-233, 128 megs RAM, NE2000 network card). I hate Windows, and I resent using it, but it does still have a better UI than KDE or Gnome.
Okay. I'll start Samba up on my Linux machine with the MP3 collection. SMBD: [OK] NMBD: [OK]. Now, I'll navigate to my MP3 collection through the Samba share... there. Now let me click on the folder and open it.
One mississippi... two mississippi... three mississippi... four mississippi... five mississippi... six mississip
There. It's done. About 6 seconds, and all 2,751 MP3s currently in the folder are visible. I can click on and play anything from Arlo Guthrie to Warren Zevon.
The moral of the story? Windows 2000 on a slower machine and with an ancient network card on a 10Mbps network will still browse a directory of 2,751 MP3s faster than KDE 3.01 will browse the same locally-mounted directory of MP3s.
By an order of magnitude.
That's a usability problem which cannot be ignored.
A person who drives a train is called a train driver. They are not an engineer unless they are a member of a chartered institution (unlikely unless they drive trains for fun). Equally the guy who fixes your car is a mechanic, not an engineer.
I do have to point out that, much to my dismay, in Canada there are instituted policies of political correctness.
The largest city in Canada erected a "Holiday Tree" in the town square. Similarly, they employ "sanitation engineers" to load household waste into the back of trucks. Apparently, they have to have nicer titles than "garbageman", when the unions have them paid $40 an hour for work far less challenging than a McDonalds job.
I resent the use of this terminology because it undermines the value of my iron ring, but the forces responsible are so many Bachelors of Arts that they refuse to learn something "technical" like the difference between someone who has been through four years of hell, and someone who hasn't.
Having said that, I do know a gas station engineer, and a Wal*Mart customer service engineer. One's BEng. Electrical, the other is a BEng. Mechanical. Both were fresh graduates in Ottawa when Nortel started its plunge toward oblivion.
Slashdot Mirror
Here's a thought I had, but probably will never get around to building.
Lots of people go to the expense and effort of building/buying radiators or using large tanks of water as the heatsink for their water-based CPU cooler systems.
Last year, I started measuring the temperature of the water in my toilet tank. After a flush, it drops to 5-6 degrees Celsius. Between flushes, it gradually reaches room temperature, of course, but this is still no worse than a radiator or bucket. In practice, however, it never actually gets above about 10C (while room temperature is about 20C).
In other words, it's a supply of cold water which you were going to simply flush away.
Place a small bucket inside the toilet tank. Put a submersible pump in there, run the water to the CPU coolers, bring the water back and drain it over the bucket in the tank.
Everytime you flush the 6 beers you went through while flaming me for my Linux isn't ready for the desktop article, you can rest assured that the water which cools your CPU is being replaced with fresh, cold water. No mold, no mildew.
The purpose of putting the pump in the bucket is so that there's always a supply of water for the pump, even during the flush. And the purpose of draining the return line over the bucket is so that if your toilet tank doesn't refill for some reason, you'll still keep your bucket full of water and buy some time for hardware monitors to shut the system down if it's getting too warm.
I don't know how hot the water in the toilet will get, but think about this:
Of course, the only thing I'd worry about is the quality of the submersible pump. After all, if water leaked into the pump, then the water in the toilet could come into contact with one side of the AC line... the other side of which is grounded to your fusebox. If you happened to touch another grounded object while urinating (concrete floor, sink faucet, etc), then enough current could find that your stream of urine and urethral tissues are a more attractive ground path than the plastic sewer pipe. I think I'd invest in an isolation transformer (search ebay) to reduce the risk of highly ...unpleasant... damage.
I think if one were pumping water through tubes soldered to the heatsinks of their power supply, the risks would be compounded, conceivably by a failure on the primary side of the power supply: I think I'd make a point of running the computer on an isolation transformer as well.
Ahh... the joys of being an eccentric genius.
"500 inch" is a common expression meaning 500 CUBIC inches. Which is approximately equivalent to a metric car having an 8.3L engine.
Note that in the 1970s, Cadillac put out street vehicles (mostly Eldorados) with a 500 cubic inch (8.3L) V8 engine.
Granted, they were substantially milder than a Top-Fuel class derivative of the Chrysler Hemi, but it certainly should make any imbecile who thinks that Honda invented performance cars think twice before entering into a stoplight confrontation with a real car.
You *don't* need LEGISLATION to fix this problem (isn't that what technology is for?).
Especially since the legislation will do nothing.
Here goes: So (a) if a law like the one I propose is passed on a national level, and (b) it does not substantially reduce the level of spam, then (c) I will resign my job.The problem is it's being addressed on a national level. That won't stop the African scam artists "whose money is tied up" - hopefully their oppressors will beat them in the face with a rusty camshaft - or the Chinese wishes of good fortune and prosperity that I was continually getting from some shitty company selling latex products until I finally decided to blackhole China from my mailserver.
This might keep the Florida 21-year-old unwed mother of 6 children from spamming me from her dial-up ISP of the week. But the funny thing about national laws is that they don't apply outside the nation...
If you're going to go to all that trouble, you may as well wire your water cooler into the supply line of the toilet: the tank fill pipe draws from your water reservoir, which draws from your water supply. Add a cutoff valve in the event that your water is cut off and you're done.
Are you suggesting that I pressurize the whole system to the water supply? That sounds dangerous to precious silicon. If you mean after the float valve, it's the same thing - inherent resistance in the line will result in a pressure gradient, and will probably put more pressure on the system than a fountain pump. Low pressure, high volume is the key to avoiding leaks while maintaining sufficient flow to transfer heat.
Of course this all smacks of the sort of thing a teenager would do to his honda - expensive, failure prone, and mostly useless.Actually, if your computers are located near a bathroom, I see it as far less impractical than having a bucket or a fan-cooled automotive heater core parked adjacent. This lowers initial and operating cost, reduces noise of fans, and provides a greater temperature gradient to cool the processor(s). After all, subterranian water should be cooler than room temperature, and most people get their water from underground pipes.
I agree about what teenagers do to their Hondas. But the striking difference is that this is clearly not meant to impress bubble-headed boy-band crazy 16-year-old girls, nor does it say, "Yo, homeboy, I like to pretend that I've got money even though I'm driving a four cylinder economy shitbox that I think is a racecar".
This says, "Hi! I keep a case of plutonium under my bed so that I can fuel up my modded DeLorean."
Now, if you could run the hot water through a hollow toilet seat for those cold winter mornings, then you will have something useful going.
Oooh! Thank you! ...Now how am I gonna cut a water jacket into a plastic toilet seat... Maybe a stainless steel industrial toilet seat with stainless tubing welded to the underside? Hmmm...
Here's a thought I had, but probably will never get around to building.
Lots of people go to the expense and effort of building/buying radiators or using large tanks of water as the heatsink for their water-based CPU cooler systems.
Last year, I started measuring the temperature of the water in my toilet tank. After a flush, it drops to 5-6 degrees Celsius. Between flushes, it gradually reaches room temperature, of course, but this is still no worse than a radiator or bucket. In practice, however, it never actually gets above about 10C (while room temperature is about 20C).
In other words, it's a supply of cold water which you were going to simply flush away.
Place a small bucket inside the toilet tank. Put a submersible pump in there, run the water to the CPU coolers, bring the water back and drain it over the bucket in the tank.
Everytime you flush the 6 beers you went through while flaming me for my Linux isn't ready for the desktop article, you can rest assured that the water which cools your CPU is being replaced with fresh, cold water. No mold, no mildew.
The purpose of putting the pump in the bucket is so that there's always a supply of water for the pump, even during the flush. And the purpose of draining the return line over the bucket is so that if your toilet tank doesn't refill for some reason, you'll still keep your bucket full of water and buy some time for hardware monitors to shut the system down if it's getting too warm.
I don't know how hot the water in the toilet will get, but think about this:
Of course, the only thing I'd worry about is the quality of the submersible pump. After all, if water leaked into the pump, then the water in the toilet could come into contact with one side of the AC line... the other side of which is grounded to your fusebox. If you happened to touch another grounded object while urinating (concrete floor, sink faucet, etc), then enough current could find that your stream of urine and urethral tissues are a more attractive ground path than the plastic sewer pipe. I think I'd invest in an isolation transformer (search ebay) to reduce the risk of highly ...unpleasant... damage.
Ahh... the joys of being an eccentric genius.
Hey, anyone want a Commodore PET 8096?
The cabinet is in excellent condition, even the PET label just under the monitor. Haven't fired it up to see if it works, but there's an aftermarket accelerator/RAM expansion board resting on top of the motherboard right now - it looks complete but the expansion board is just *resting* on the motherboard, like someone tried to fix or upgrade it once. I have a suspicion that the machine is fine but the attempt was along the lines of "what do you mean I can't put a PCI video card into that?". FOB Ottawa, Canada.
Schematics would be cool so that I can sell it (or give it away if there are no good offers) as a working unit.
My own collection of old TI-99/4A, Amiga 1000, Vectrex and Coleco Telstar Alpha machines already occupies quite enough room, thank you very much. And I must confess that I haven't fired up even one of my prized TI-99/4A machines in over a decade.
Powerplants always have pretty interesting tours. Try and visit a couple of different types.. coal, nuclear, hydro, etc.
Research reactors are good, too. Atomic Energy of Canada Limited (AECL) used to offer tours of three nuclear reactors just 3 hours up the road from Ottawa, in Chalk River.
You were actually allowed to stand on the NRX nuclear reactor there, while it was running, and look down into the calandria to see the pretty blue glow. It was nice and warm up there, with a thrumming under your feet from all the pumps and support equipment running.
Of course, all tourists were outfitted with dosimeters and screened thoroughly before and after.
I worked there as a co-op student one summer, it was great fun. The NRX was my favorite, but I think it's decommissioned now.
That is a pretty broad statement to make. I played Wolf 3D when I was 9, 10 years old, Doom after that, Quake after that, so on. I knew full well it was a game. It was fun dammit! What makes you presume you know the maturity level of any given kid?
Exactly. As a kid, I played Popeye on my TI-99/4A all the time. The Old Hag threw bottles at me, and I learned how to punch them away. And look at how I turned out. I'm posting an unsolicited comment to a geek website at quarter to four in the morning while smoking a cigarette and waiting to take my next Prozac. Overall, I'm better adjusted than most people I know.
Hold on just a sec... Fscking cat won't stop meowing to be let out... SHIT! No more empty beer bottles! [digging around, lobs an empty rum bottle, makes a mental note to wear shoes when going to the bathroom after bedtime]
Also, we are talking EM radiation here, not nuclear. There is a difference.
Yeah, it's unclear what the original poster thinks that he's going to be getting "nuked" by.
Actually, several kinds of radiation leave a monitor, though it's only the last two that tend to cause worry:
X-rays are ionizing radiation and are known to be carcinogenic and mutagenic. They're essentially man-made gamma rays. Geiger counters do tend to count them, though it depends on their energy.
The idea that Linux is harder to install than Windows has reached the status of an urban legend (or Microsoft FUD) -- this is 2003, not 1997.
I agree in almost every way with what you've said.
ISA hardware support is lacking, but fortunately that problem becomes less crucial with each day.
Some hardware remains poorly supported, especially video cards and winmodems. They're reality, and we have to keep up the pressure on the manufacturers to release information on programming device drivers for their stuff.
Software installation remains an issue. Debian and Red Hat deserve praise for these, but they're far less foolproof than the installers shipped with most Windows software. When was the last time that a Windows installer whined at you that you lacked a certain DLL? For the most part, they're simply installed. Maybe Linux applications need to be able to (?download and?) install their own libraries, independently of the operating system, and without user intervention. We have to find a way around the dependency hell which you encounter so frequently. And no, Joe Sixpack is not likely to compile his own stuff.
But the Windows comfort is more along the lines of familiarity with the locations of things, with how everything works, pitfalls to avoid, and familiarity with applications. Users will (and do!) crave that when moved to the Brave New World of Linux. We can address some of it, but much of it is inherent to changing operating systems. We have to be conscious of the fact that it's a force beconing many convertees back to Windows.
I'm actually an advocate of linux on the desktop (yes I am) and it seems those points you mentionned don't make much sense, here's why.
I'm an advocate of Linux on the desktop, too. I can't wait to see it. But from the perspective of an *average user*, I'm still convinced that it isn't ready. We *need* the average user to feel *more* comfortable working at a Linux desktop than a Windows machine, especially since he's gonna have to deal with a Windows world trying to suck him back into its comfortable embrace... at least until we've finished the takeover of the desktop.
- Linux GUIs are faster and faster at each version. Gnome2 for example was totally re-coded with performance in mind and behaves much better now, KDE 3.1 (still a release candidate but still) on this box is working SO much faster than XP did on the SAME box!This is good. I cannot corroborate it using a pre-compiled distro. Why am I using a pre-compiled distro? Because that's what Joe Sixpack is gonna be using. More optimization is needed, and more carefully made binaries are required from the major distros; especially Red Hat in the current #1 off-the-shelf position.
I think part of the problem is that we need developers to try actually using the pre-compiled binaries of their works which end up being shipped with the Red Hats and the Mandrakes of the world.
- Since I've been running linux on my desktop, I have not yet had one problem reading any PPT, DOC, etc... documents... not once... sorry. And I get a lot of ppt and doc files sent to me dailyMost of them have formatting problems, cannot handle inline images (properly or at all). Table support from Word 2000 is lacking. I know this is a serious pain in the ass to reverse engineer, but it merely frustrates end-users who are already gonna be pissed off about having to learn something new when their company moves to Linux.
- I have had problems with some applications, contacted the mailing list, and the solution was sent to me a few minutes later... no RTFM.You're not Joe Sixpack. "How come it says I cannot save my file in /bin? Huh? I didn't log in as root, whoever that is." Screams of RTFM or "Get a life" would abound on mailing lists or IRC, whereas a 1-900-DRONE would calmly answer, explain, and the user would be supported. Sure, the example I cited is an operating system issue instead of an application issue, but it's a problem every bit as simple, stupid and pervasive.
- I use Evolution for my email/calendar/tasklist/contact management stuff, it has everything I could ever use and more... I have used kmail in the past, I've never had any real problem with it.KMail is great, the only programming complaint I've had with it is that it silently dies if it runs out of disk space. But the spellchecker is right out of 1995. We have to match feature-for-feature to be adopted. You're not going to sell Linux/KDE (or Linux/Gnome or OpenBSD/AfterStep or whatever) by screaming from the hilltops, "ALL THE FEATURES OF WINDOWS 3.1!" in a Windows XP world.
Evolution was too slow to be usable on my PIII-500. That's insane. It's just an e-mail client, not a genome sequencer, for Gawd's sake!
- Recent linux distributions based on more recent and less backward-compatible glibc usually have some kind of package management system that will not only save you from searching on freshmeat, but also install directly the application for you. emerge gnucash apt-get install gnucash synaptic->gnucash and so on... You have now installed the latest version of an excellent financial software, which, may I add, will read files from other windows software like Quickbook or Quicken without a glitchIt's a good start, yes.
But the biggest problem is that if a feature which an Excel user would take for granted is lacking, it's a negative perception. Most users will already resist the change to something new and "strange".
We've grown up with the idea of piping the output from one program to another; it's the Unix way. But it's *not* acceptable on a desktop system. You don't do your spreadsheet in OpenOffice Calc, then save it in some format that Gnumeric handles so that you can use the point-and-click data analysis tools, then open int up in OpenOffice again. If you're paying a secretary $20/hr to do this, it doesn't take more that a few months to make back what you would have spent to install Windows on the machine.
- I use daily applications for all my needs, none of them are poorly written at all. licq is stable as a rock, xmms plays music just perfectly, evolution still handles my emails (without a virus or worm or anything like that infesting my computer), mozilla works like a charm and KDE 3.1 is just a dream. Although all those applications work in a much superior fashion than equivalent applications on windows, they ARE skinnable indeedI don't know how well Evolution handles e-mail. My main machine is over the hill, but easily captures video from my TV card in real-time. I find it hard to believe that responding to e-mail in Evolution should require such a fast computer as to be unusable on a machine which will capture NTSC video at 29.97FPS with 16 bit stereo sound with 0 dropped frames... (unless I open Evolution while I'm capturing video).
Mozilla is great. It's fast, attractive, and it works well. The only problems I have with it are fault tolerance (delete your JRE without telling Mozilla, then try to use a website infected with applets; it crashes with no warning), lack of ability to send a mailto: link to anything other than Mozilla's mail client, and the inability to tailor the browser string to be whatever I want without recompiling (at least one website I *have* to use will ban you if your browser doesn't say "MSIE" in its string).
- Companies such as the Kompany, RedHat, Suse, etc... actually DO have some marketing people that make your desktop look just like you want it to look like as a user and to behave. My desktop right now looks simply amazing, yet is really fast and everything is at hand. My girlfriend uses it every time she comes, all my friends really love the way it's set up and even my mom used it and didn't have a problem doing everything she needed to do.For sure. This is a good step. But part of the problem is with the overall look of it. Red Hat 7.3, for example, with probably the biggest marketing department in the Linux world, comes with a highly saturated eye-straining blue background.
Contrast this to the relatively neutral backgrounds of Windows and Mac environments, and it looks more like we're trying to sell a product than design something useful out of the box.
Even XP's default meadow is less eye-straining.
If some Joe Sixpacks can't figure out how to move the Windows taskbar to someplace they like better, do you really think they'll change the backgrounds and skins to something less displeasing? The desktop's defaults must be *neutral*, *inoffensive* and *non-eyestrain-inducing* out of the box with *every* distribution.
- and for the support thing, companies like Suse, RedHat, Mandrake, etc... DO offer commercial (cheap) support for pretty much all the applications shipped with their distributions, in fact, and I speak from experience, these companies go way beyond that by helping out users with applications not "officially" supported, and also collect bug-reports and offer patches to the original developer of the software to fix the problem for them (http://www.redhat.com/bugzilla) for example.Who do you call when you need support with OpenOffice or xine? I haven't tried either; I've got the luxury of being able to pursue the source code.
I do know that at one of my former employers - a huge defense contractor staffed by engineers and computer scientists - we spent a lot of our IT budget on calls to Microsoft looking for support on how to create PowerPoint slides with embedded video and other dead-easy things like that.
Sucky as that may be, it's reality for lots of organizations. We have to address that.
- Whoever wrote that has NO idea of how much a business license for Microsoft Windows costs... it's not even close to $200. Tell this person to add many zeros to that number.Sorry. $299, according to the Microsoft website, for Windows XP Professional, in single units, as a standalone operating system instead of an upgrade.
It remains that the purchase price is a very, very small part of the total cost of ownership.
I think linux is still very young on the desktop OS market but it's doing a great job and I'm very impressed by how fast it's moving forward...This is true, but let's stop kidding ourselves about it being ready. It's not ready for the desktop yet.
Linux has made amazing strides since its inception a mere 10 years ago. It's already a secure and stable server operating system, with mature tools for sysadmins.
But it's still at workstation space. We can take heart; it's more usable on the desktop than a $30,000 Sun workstation, but it's still not ready to supplant Windows yet.
The biggest obstacles are not the Linux kernel, or even Linux itself, of course. The obstacles are a fast, feature-filled and stable desktop metaphore (be it KDE or Gnome or whatever) with good *USER* applications readily available. (Don't even bother sending me flames telling me that vi is the greatest word processor ever made because Joe Sixpack isn't gonna even gonna figure out how to bring up the help screen.)
KDE, Gnome, Evolution, OpenOffice, etc... all these software are working on a new development version right now that's purely amazing... I can't wait to see what it will be like by the end of the year 2003!I can't wait to see what it's like 20 years from now.
I've been waiting 15 years to see the end of Windows.
I'm not a huge advocate of Linux on the desktop (yet)
I want to be, but I can't (yet). [grin]
Here's the problem:
To put Linux on the desktop, we're asking them to give up the comfort, familiarity and applications of Windows. For what benefits?
Poor applications. Quoting an e-mail I received: "But a lot of it - and mainly the GUI stuff - is still lagging behind, being a slower and buggier version of a half-decent program on Windows. And priorities are wonderful - when we build a GUI application, the most important thing is that it's skinnable. Bugs? Features? Competition? Who cares?! It's skinnable!"
The same writer continues... "And for the biggest question: Mr. Rupert wants a financial software for Linux (his son installed it for him). So he calls his son over to install a simple financial software - just something which can calculate his loan repayments. His son opens google (or freshmeat), and finds 31 financial programs. Each has a different set of features, of course. He downloads and compiles each of them (ah, yes, the rpm was compiled using an ancient glibc version, and no, Mr. Rupert doesn't know what glibc is). The only two candidates which could actually be compiled (and didn't require libobscure.so.2) and actually have this option in their ugly programmer-designed-GUI menus die as soon as you choose the option. That's right - the operating system is stable as a rock, but the programs die immediately. What's Mr. Rupert going to use? hmm.... Maybe a respectable program from a respectable company (on Windows, of course).
But wait! John Rupert (the little 15 year old) can program - he's got some C tutorials, and he's written a few small programs. Why can't he write the program for his father? And the 32nd version is on its way."
We need to work on this stuff. Linux still isn't ready for the desktop.
...that this guy should learn how to solder.
Sorry. But this is simultaneously an amazing project and an act of butchery.
A few words of advice for anyone trying this themselves: use electronics solder- it doesn't stick to breadboard. I later got curious and tested regular solder on an old NIC and it stuck everywhere. Watch what you buy."Regular solder"? Is that plumber's solder, like you'd use to sweat two pipes together?
Solder includes chemicals (flux) which help to clean the pieces of metal which are being attached. Electronic solder is either rosin core or organic core; they're fairly gentle. Acid core is used by plumbers and the post-soldering remains of the flux attack electronic components over the years.
Based on some of the pictures, I believe that the soldering to the PC board was done with an overly-large soldering iron. A fine grounded-tip 15-25W soldering iron is *essential*. My favorites are the tiny little Ungar irons from just before Weller bought them out.
BTW, it *is* possible to solder to aluminum, but it's very different. For one thing, aluminum coats itself almost instantly with a very fine layer of very hard aluminum oxide. Solder will not stick to this layer. The other problem is that aluminum conducts heat away from the attempted soldered connection. A large-wattage iron with a sharpened tip will do the trick. Apply a puddle of solder and, scrape the aluminum under the puddle with the tip of the hot iron. A bond will form, and the strength will be mostly dependent on how well you scraped the aluminum under the puddle.
Gaffer tape has a tendency to dry out, shrink and peel off over time. This limits the longevity of this particular modification. Of course, you probably don't want to cut up the typewriter since you chose it because you like it, and they don't make 'em anymore, so I applaud the reversibility.
Rather than attempting to make a distinct "switch" for each button, why not simply have a scheme where each keypress will ground a wire to the crossbar? Of course, that won't work with the keyboard's matrix arrangement, but that can be easily solved in a minute.
PC board looks like a good way to insulate parts of your switch assemblies, since it's cheap, readily-available, easy to work with, and you can use it to make narrow switches.
Onto each one of the levers (which are steel) you could solder a small piece of printed circuit board with a custom pattern. (You can buy printed circuit board etching supplies at Radio Shack.) One part of this tiny board's pattern is used to solder it to the lever. Then a gap, with no copper on the board. The opposite end of this pattern is a place to which you can solder a flexible wire and a small fold of some springy nickel-plated steel.
The tiny board then rides with the lever and the new springy-steel contact then connects the wire to the crossbar.
As for connecting the keyboard to the matrix, my first thought is to use optocouplers. Optocouplers are merely an LED and a photocell built into the same case. They're meant to isolate different parts of electronic circuits.
Solder a piece of flexible wire to the crossbar, and connect that directly to the keyboard's ground on the PS/2 or DIN connector. (You can get the pinout from the Internet.) Take the +5V lead from the keyboard and put about a 500 ohm resistor in series with it, then carry that across to the anodes of the LEDs in the optocouplers. Connect each cathode to each wire coming off the lever boards. Now, when you hit a key, the LED in the corresponding optocoupler should be lit.
The photocell in an optocoupler is actually a kind of transistor, which is essentially an electronic switch. Connect each optocoupler to its corresponding pair of contacts on the keyboard's matrix. Make sure you get the polarity right, a quick check of the keyboard with a voltmeter should do it.
Choose an optocoupler with a good transfer characteristic; probably Darlington-outputs. LEDs need current limiting, and that's what the 500 ohm resistor is for. Now, there's ONE resistor being used to power all those LEDs in the optocoupler, and it limits the single or total LED current to 10mA. This is done because motherboards often have fuses to keep keyboard current below about 50mA. Even if you hold down all the keys on the keyboard, you will not draw more than 10mA. Blowing the keyboard fuse on your motherboard would suck.
Now, when the Smithsonian comes calling and wants your typewriter because FDR used to keep it in the trunk of the Presidential Limo? Desolder the wire from the crossbar, desolder all the little pieces of printed circuit board from the levers, and the typewriter is undamaged.
Yeah Hard drive noise is bad but it's nothing like it used to be. I remember some of the older drives I had that made so much noise it scared away my cat.
Still, there's nothing like the seismic rumble of one of the good old full-height 5.25" 20 megabyte Seagates starting up. And the squeak-squeak sounds of band-stepper actuators.
Ahhh... remember the good old days where you always let the drive warm up for 20 minutes before you saved anything, because the old actuators didn't account for expansion and contraction of the platters and arm?
I must resurrect one of those just for the fun of being able to stick, like, 4 MP3s on a drive which draws 12V @ 6A at startup.
Of course I used to buy crappy hardware so the drives were bad to begin with.There was a company called Kalok which was producing mega-cheap hard disk drives in about 1995, before they got bought out by an even more fledgling JTS. They had a 100 megabyte 3.5" hard disk drive which was selling for about $75-$100 less than a comparable name-brand drive. 'Course, there was a catch... in fact, two of them.
The Kalok had a band-stepper actuator - which is impressive because everything else from 40 megabytes and up seemed to have a voice-coil actuator. Needless to say, installing Windows 95 on a Kalok hard drive was a bad idea, since the system wrote to boot up logs and the like during startup - before the hard drive had a chance to warm up.
The other catch reads like a bad joke: The hard drives were made in India.
You are assuming a non-catastrophic failure. What if you are doing 80mph on the freeway and your front tire (which has been running on low pressure for 2 months) blows out.
Right there, you are doing 2 things which directly contribute to the accident, speeding and low tire pressure.
Posted speed limits on some Michigan freeways were 75MPH, last time I was there. 80MPH is within the +/-10% accuracy that is mostly assumed of automobile speedometers, so it's unlikely that you'd be pulled over for doing 80 in a 75MPH zone.
*HOWEVER*, only 20 years ago, every freeway gas station had a garage, well-stocked with tires.
Tires have made tremendous leaps in their technology in the past 20 years. In 1982, blow-outs were common, and in 1972, even more so.
Back then, any driver knew how to handle the car when a tire blew out, and would manage to get it off to the side of the road without killing anyone else.
What we're saying is that a mere 20 years after these events were last commonplace, the motoring public's incompetence is such that it has now turned what used to be a nuisance failure into a deadly failure.
Look at the Ford Explorers on Firestone ATX P235-75/R15 tires. There's obviously a problem with that combination of tires and suspension design. But the failure is a *tread separation*. The tire merely loses the outer inch or so of rubber, this is a far less catastrophic failure than the blow-outs of bias-ply tires on the cars of the 1970s and before. And yet people apparently have no idea how to control their vehicles when this happens.
It's pathetic.
If you cannot control your vehicle through a mere tire blowout (let alone something as trivial as a tread separation), you shouldn't be allowed on the road.
Should you be held accountable for that? What if someone dies in the accident? Does the family of the victim deserve to know that you killed their loved one? Do the cops deserve to have access to information that could prove you commited a crime which caused an accident?Being unable to control a vehicle through a tire blowout is negligence. There's no excuse for not being able to handle the car. No part of any vehicle is fail-safe - especially not tires - and you have to be competent at coping with at least small failures like tires.
People need to take their cars more seriously. They need to learn how to drive, not by dad, but by people who are professional drivers. They need to know that the oil, water, wiper fluid, tire pressure, tread, and a myrid of other things are important.I agree. There should also be a mechanical proficiency requirement to getting a driver's license. People need to know how wheel bearings work so that they know when a grinding sound is a potentially deadly failure. They need to know how brakes work so they can appreciate the dangers of the clear oily patch on the driveway. They need to know how their steering and suspension systems work so that they don't fuck around when the steering wheel play is getting excessive because a ball joint is getting loose enough to separate.
My animated paperclip went on a bender and refuses to speak to me.
The last time I heard mine was a dying scream as I mounted my FAT32 partition, navigated to it, and typed the magic letters:
# rm -rf *
It was high, blood-curdling, but strangely satisfying. Like the sound of the welds in a Honda's body popping as the car crusher takes it down to 3 apples tall, then the wet thunk of a cast-aluminum engine block cracking like a flowerpot in a vise.
Mercifully, when I had to install Excel on Wine because OpenOffice doesn't do something as fscking simple as a polynomial regression, the damned paperclip didn't work.
Of course Microsoft is their own worst enemy. Who else would allow IIS or Outlook - a security hole which masquerades as an e-mail client - to be some of their flagship products?
The security holes are even more annoying than the damned animated paperclip.
But why all the "worst games" lists?
The usual hype. But I was surprised by the lack of one hugely important game:
Hunt the Wumpus.
Hunt the Wumpus was apparently an old Unix text-based RPG, which Texas Instruments brought to life on their under-rated but massively overbuilt TI-99/4A home computer in 1980 or so.
The TI-99/4A (and its rare older brother, the TI-99/4) had a 16 bit TMS9900 processor chip (in 1979 and 1981, boys and girls!), a kick-butt video chip (the TMS9918) which had 32 sprites and a video overlay feature. But Texas Instruments, a company which is/was making more chips than Frito-Lay, hobbled the machine by using the video chip's RAM as the console's main memory, bottlenecking the expanded memory down to 8 bits, and creating the single slowest BASIC interpreter ever designed by having it interpreted TWICE (from BASIC to GPL - "Graphics Programming Language" - then to machine language).
With this nasty kludge, they released a graphical version of Hunt The Wumpus. Horrible sound effects, and game play which made you feel like you were drunk and on LSD. Oh, and attempting to add graphics to an old text-only game is doomed to fail, don't even attempt it.
[sigh] Quoting from aap.org:
Research indicates that during the first year of life an uncircumcised male infant has at most about a 1 in 100 chance of developing a UTI, while a circumcised male has about a 1 in 1000 chance.
Studies conclude that the risk of an uncircumcised man developing penile cancer is more than three-fold that of a circumcised man.
Some research suggests that circumcised men may be at a reduced risk for developing syphilis and HIV infections. However, the AAP policy states that behavioral factors continue to be far more important in determining a person's risk of contracting sexually transmitted diseases than circumcision status. (I have never once in my life had sex without a condom. However, if we're talking about HIV, I'll keep every advantage in my corner that I can, thank you very much.)
Considerable new evidence shows that newborns circumcised without analgesia experience pain and stress measured by changes in heart rate, blood pressure, oxygen saturation and cortisol levels. Other studies suggest that the circumcision experience may cause infants to respond more strongly to pain of future immunization than those who are uncircumcised. (Note the use of the word "infants" in the phrase "infants to respond more strongly". This indicates that the adverse effect is limited to infancy, and does not imply future harm. Otherwise, by extension, almost all North American men would shrink from a paper cut for fear of reliving their circumcisions.)
Quoting from one of your previous posts:
Now I can build a virtual 3D woman on my linux system. THANK YOU. I have had sex only once in my life, and that was when I was in mid 20's (No Joke). I LOVE technology!
[sigh] Yet another NOHARMM or NOCIRC wacko. Let me guess: You're fat, bald, lazy and ugly. Your life sucks, but you're absolutely convinced that you'd be as dashing, financially and sexually successful as James Bond, if only you'd been allowed to keep your foreskin.
Oh, poor, poor you.
Even today, I believe that circumcisions are performed without anaesthesia. I remember seeing a news story a few years ago about a circumcision conference. The doctors there had previously thought that babies could not feel pain (?!) but they were beginning to reconsider this claim..
Generally, it's a topical anaesthetic used now. (topical: applied to the skin as a lotion, rather than injected)
And while it seems inhumane, consider:
...but I was 22 at the time! Ha!
Oh, the requirements of dating an Orthodox Jew. The relationship didn't work out, but I have no regrets. I feel so free and unencumbered, it's great.
Ignoring the rest of your comment because it looks like it could easily become a circular argument, let me just stick with your single most telling comment:
Hmmm, I just loaded up my directory of MP3s and Oggs. I used the xmms "+DIR" button to add the entire directory of 2808 songs. It took 62 seconds. Over NFS on 100mbit ethernet. From a 200MHz pentium with PIO mode 4 disks. Moreover it took less than a second to sort the entire list by filename. And I haven't been listening to my entire music collection this week (Christmas music)- so except for 25-some songs, none of that was in the buffer cache of *either* machine.Okay. What you're suggesting is that I use another application (xmms) as my file browser, because the file browser built into the leading desktop metaphor is so krufty. You're advocating a work-around which bypasses one of the most fundamental features of KDE, and you apparently don't see a problem with that.
Even so, the speed is too slow. In my case, it does take me over 2 minutes to use the +DIR button, and that's using a PIII-500 with 512 megs of RAM and the MP3 collection being on a separate drive from /. Both drives (/ and /mnt/mp3) are running at UDMA/66 with ext2fs. There's no NFS involved, the MP3 collection is local to that machine.
One way or another, I'm not sure how your workaround would upset my expectations of being able to choose a song and play it immediately when I feel like it.
You see, from my experimentation with your workaround here, it seems that I would have to open xmms, choose +DIR, then wait 2 minutes before I can scroll through and choose the song I wish to listen to.
The alternative is, of course, to save the list as a playlist, but opening large playlists takes as much time as using +DIR. And the list gets stale; music which I've just downloaded isn't available in my playlists until I've updated them - which takes 2 minutes.
By contrast, I'll move over to my Windows 2000 machine (PII-233, 128 megs RAM, NE2000 network card). I hate Windows, and I resent using it, but it does still have a better UI than KDE or Gnome.
Okay. I'll start Samba up on my Linux machine with the MP3 collection. SMBD: [OK] NMBD: [OK]. Now, I'll navigate to my MP3 collection through the Samba share... there. Now let me click on the folder and open it.
One mississippi... two mississippi... three mississippi... four mississippi... five mississippi... six mississip
There. It's done. About 6 seconds, and all 2,751 MP3s currently in the folder are visible. I can click on and play anything from Arlo Guthrie to Warren Zevon.
The moral of the story? Windows 2000 on a slower machine and with an ancient network card on a 10Mbps network will still browse a directory of 2,751 MP3s faster than KDE 3.01 will browse the same locally-mounted directory of MP3s.
By an order of magnitude.
That's a usability problem which cannot be ignored.
A person who drives a train is called a train driver. They are not an engineer unless they are a member of a chartered institution (unlikely unless they drive trains for fun). Equally the guy who fixes your car is a mechanic, not an engineer.
I do have to point out that, much to my dismay, in Canada there are instituted policies of political correctness.
The largest city in Canada erected a "Holiday Tree" in the town square. Similarly, they employ "sanitation engineers" to load household waste into the back of trucks. Apparently, they have to have nicer titles than "garbageman", when the unions have them paid $40 an hour for work far less challenging than a McDonalds job.
I resent the use of this terminology because it undermines the value of my iron ring, but the forces responsible are so many Bachelors of Arts that they refuse to learn something "technical" like the difference between someone who has been through four years of hell, and someone who hasn't.
Having said that, I do know a gas station engineer, and a Wal*Mart customer service engineer. One's BEng. Electrical, the other is a BEng. Mechanical. Both were fresh graduates in Ottawa when Nortel started its plunge toward oblivion.
Right, what KDE needs is a Blue Screen of Death. Or maybe teal.
[pondering...] Nothing beats the Amiga Guru Meditation Error!