The form-factor is pretty good, but the power consumption and feature set aren't quite there: This thing is going to draw at least 20W (that's a bit more than the C3 CPU requires alone) and it's lacking any sort of battery charging/monitoring circuitry (I'd complain about the lack of LCD interface, but you can get the board with an LVDS port, which should be close enough). On top of that, it doesn't have any serial ports and only has USB host ports (so you can't connect it to your desktop, for file syncronization or such, without using ethernet). Also, if they were supposed to have done away with 'legacy' ports, what's that PS/2 keyboard/mouse connector doing there?
For a moment there I was afraid that this would kill my own pet project, but after seeing what the power consumption is like, I am not worried.
First, you need to figure out how to interface to the print head. If you can't find documentation from print head manufacturers, you will need to break out the oscilloscope and poke at your cheap-printer-of-choice until you can get satisfactory control.
Next, you will need to rig a larger ink reservior for the print head. This should be pretty easy if you have any mechanical aptitude and access to a machine shop. You also need to secure a source of compatible ink.
I'm tempted to say you should skip the print head altogether and use a simple spray nozzle (disassemble a cheap airbrush with a fine nozzle and actuate the needle with a solenoid), but this would mean that you would be working at very low resolution (maybe 10 or 20 dpi). You will still need to maintain a reasonably constant distance from the media, but you tolerances will be less demanding and you will have a wider choice of inks.
Finally, you need a three axis robot to position the head over the media. You can sense the distance from the media optically (a couple photodiodes, some lenses and a diode laser should do the trick). Getting exact positioning of the head in the X-Y direction is a bit harder (good mechanical positioning over 2 meters will be expensive) but you can probably use an ultrasonic system and get reasonable results (there used to be digitizers for CAD use that had an ultrasonic clicker in the pen/puck and a pair of microphones on perpendicular edges of the tablet. Either the delay from click to reception, or the amplitude at the mics, was used to determine the pen/puck position). Just put a ultrasonic clicker on the print head, and click a couple of times on each X-Y positioning operation to fix the head location.
Assemble the above parts, along with some custom hardware and software for interface and control, and you should be good to go.
Oh, only a few thousand, but we don't know the real number because if the government had to reveal who it has secretly arrested THEN THE TERRORISTS HAVE WON!
Check out my journal for some discussion of exactly how to do such a thing. The first bit of advice is that you should consider a microcontroller as the heart of the system. There are lots of good candidates out there: PPC, ARM, and MIPS devices are common. You might be able to find some x86 based devices as well.
Many of the current MCUs are ball grid array (BGA) devices, which make them pretty hard to work with if you're not a professional, but a few can be had in PLCC or QFP packages, which means you can get an adapter board or socket.
You can also buy preassembled demo/development boards (this is the route I'm taking) and wire-up anything the board doesn't include by hand. Most of the MCUs on the market will have 32-bit memory busses (though they may not support more that 25 or 26 address lines), so you can attach just about anything you want to them.
This is exactly the kind of thing that Steve Ciarcia (of Ciarcia's Circuit Cellar) used to do: building personal computers from microcontrollers. Most of his designs used Z80 based devices, which was fine back in the mid- to late-eighties. Now, however, you can do a fair bit better.
As for speed, I don't know exactly what you're looking for, but the ARM devices can be had in speeds from 50MHz to 400MHz, and the same is true of the PPC and MIPS devices. That may not seem like much, compared to a 2GHz Pentium, but it's really quite nice.
Some good resources: Digi-Key is a reasonable source for all sorts of parts, Atmel makes some nice MCUs, programmable logic, and Flash RAM, Cirrus Logic makes some ARM MCUs and networking chips (amoung other things), Sharp, Samsung, Motorola, and AMD all make nice MCUs, Cogent Computers builds some nice development boards, and EarthLCD has good prices on LCDs and has an ARM based board in the works.
I am in the process of doing something very much like building a custom laptop, but I'm afraid I don't have very good news: it will be expensive, time consuming, and you will need to make many sacrifices.
If you check out my journal you will get a detailed description of what I've been doing and why. In a nutshell: I'm trying to build a handheld computer that delivers adequate performance with maximum flexibility and battery life. The target design will have a small (7-8 inch) VGA screen, a moderately slow processor (100-200 MHz ARM), moderately sized hard disk (10-20 GB, 2" IDE), a moderate amount of RAM (32-256 MB), and run off of conventional rechargable batteries (8 AA NiMH cells) for at least 8-hours on a single charge.
Just to build the prototype will take several months and cost at least $1000 (probably several times that, when all is said and done). If I go to production, I can probably get the cost down to $300 per unit (not counting NRE costs), but the selling price will still be up in the $500 range.
The only way I am able to do any of this is that I am abandoning any kind of PC compatability: This device will never be able to run Windows or play flashy games (though it will run a regular, non-embedded, version of Linux). I've had to give up on all kinds of features that seemed like they should be simple: I don't have a clamshell case because the engineering is too complex for me to do myself and I don't have a keybaord because I couldn't find a source for laptop-style keyboards. I'm still holding out for polymer-LiION batteries, but I haven't found anywhere that builds them on contract, yet.
This is not the same sort of task as building a desktop computer from white-box parts (which is only slightly harder than plugging together lego blocks). This is much closer to actual engineering: I'll be constructing some of my own breadboards to connect the SBC to the LCD and the IDE hard disk, as well as custom hardware to charge and monitor the batteries.
First, learning this stuff is going to be hard, because the subject matter is hard. Don't let this disuade you, however, from the task: it is worth every bit of effort you put into it.
Second, some books:
go to the local community college or state university and invest in some college level text books. These books will be better written than highschool texts and will be more concerned with teaching than with covering some sanctioned set of facts and figures. Whatever you local college/university has is probably fine. Always buy used textbooks: they're cheaper.
consider getting an encyclopaedia: anything that isn't covered by one of your textbooks will be covered adequately, by the encyclopaedia, to answer your or your children's immediate questions.
If all this is a bit too pricey, you can substitute the local public library for almost all of this (I've never looked in the public library for college textbooks, but everything else should be there)
Third, don't skimp on the methematics! Mathematics, especially calculus, underlies all of modern science. You can't really understand most of the science without understanding calculus, and if you understand calculus, much of the science will simply fall into place.
Finally, don't be too proud or stubborn to actually go back to school. You can enroll in a night course at the local community college for less than the cost of a bare-bones PC-clone. A little actual instruction goes a lot farther than a whole lot of unaided reading. You might also have some fun. (it's amazing how much fun learning is when you're not worried about getting the piece-of-paper)
If you can learn the basics of physics and math, you will be able to cope with (if not master) just about anything. Anything you weren't actually taught, you will be able to get a reasonable grasp of after a few days (sometimes only a few hours) of reading and thinking. If something takes longer than that, you have hit upon a trully difficult subject and may need to look for a course to take.
I noticed that too (but you beat me to the post). I think that there is another explanation, however:
The Onyx4 either currently is, or will soon be, based on the Itanium rather than the MIPS. HP did something simlar with their recent platforms (shipped with PA-RISC but were plug-compatible with Itanium).
The marketing-speak "Industry Leading Processors" is awfully suspicious. The sad part is, SGI doesn't have any good options:
They already discredited the MIPS, so they can't admit to using that.
They can't brag about the Itanium, since it's not doing all that stellarly well (not, at least, as well as it was hyped to do).
They can't transition to x86, since they already tried that once and it was a disaster.
They can't transition to some other platform, since they haven't got any residual credability with which to fund such a move (anyone still using SGIs would rather jump ship entirely).
SGI has tried just about every dumb trick in the book (most pioneered by DEC) to find some way to move from thier ever shrinking niche (data visualization and computer animation) to something broader and more profitable. At each step along the way they have annoyed and alienated their loyal customers.
Even a lifespan of 3-5 years doesn't rise to the level of 'investment', at least not in my book. The residual value in a piece of computing equipment after 3 years is very close to zero (consider the combination of resale value, replacement cost, and costs of maintenance over the previous three years of use: Once you've spent as much in administrator time as it would take to buy a new widget, the widget's life is over). Desks, buildings and vehicles don't depreciate as rapidly as most IT hardware (I'll grant that high-ticket items -- large scale servers, storage farms, etc. -- tend to keep a fair amount of value over time), and are properly thought of as investments. Most IT hardware, however, has a shorter lifespan than a stapler or tape dispenser! Computers are, for the most part, consumables.
As for laying the blame for this attitude at Microsoft's feet: I don't think that's entirely fair. Microsoft certainly benefits from frequent replacement of IT equipment, but there are plenty of other benficiaries who have contributed: Intel is a obvious culprit, since they only stay in business if they can convince people to replace existing systems with brand-spanking-new kit. Another culprit, however, is not nearly so malign: Moore's Law has reduced the cost of all computing hardware to the point that many systems are cheaper to replace than to repair or, even, maintain.
Second: The author is just plain looney. IT is not an investment any more than sticky notes are, and for precisely the same reason: Most of your IT hardware will need replacement in very short order (only a few years).
If any part of the IT department is an investment it is the expertise of the people you employ to administer your IT, but either you (as a company) already know this (by treating all your employees as investments) or you don't (which, it appears, is the far more common attitude). Even the employess are not a very good investment, since there is damn little you can do to make sure they don't leave (you can't own your employees, at least not in most places) and they require a huge amount of maintenance (salaries, benefits, training, etc.).
Everything else about IT is obviosly a simple cost (possibly amortized over a short period of time): All the hardware will be obsolete within only a few years. All the data you collect has an almost equally short shelf life (how long are individual customer demographics good for? how about marketing materials? maybe financial data has a reasonable lifespan, but it's not much more than 5 or 7 years) and have a continuing cost of maintenance in order to keep them up-to-date.
The longest lived part of your IT infrastructure is probably the network cables, which can last 10 years or so. None of these items have the lifespan of a building, however, which is the canonical example of an investment item.
This is by no means the culmination of "Apple Technology."... Good marketing, good engineering, overall a good job but nothing that stunning that originated in Steve's back yard.
In fact, even if none of the electronics were designed by Apple, the most important part "originated in Steve's back yard:" What other company has the guts and determination to mass-produce non-x86 (and, for that matter, non-Windows) hardware targeted at the consumer desktop? Apple is the only company left standing from the great microcomputer ecology of the seventies and eighties, and the only one that was willing to stay the course, no matter how difficult.
So, sure, Apple farms out huge parts of their engineering effort, but the only way this sets them apart from any other consumer manufacturer, is that they do more custom engineering than any of their competition! (Does anyone really think that Dell or Gateway are doing anything more than throwing together a bunch of white-box parts and sticking them in a fancy box?)
I'll admit that calling this the culminatio of Apple's technology may be gilting the rose, but it is certainly the culmination of Apple's vision and dedication. Without Apple this sort of technology would never have been brought to a desktop near you, or, at best, it would still be years in the future.
I don't get it. Why is the transportation of hydrogen any different from the above examples? Or is the writer just making careless statements?
Because hydrogen molecules are really small, much smaller than any of the other gasses you mentioned (methane, butane, propane and acetylene are all hydrocarbons, which means they are fairly large: at leat one carbon and four hydrogens. Even the noble gasses are pretty big, with the exception of helium) and have a tendancy to leak through solid metal pipes! See this site for some details.
That said, I can think of a few ways to build storage containers and pipelines that would ensure that leaked hydrogen re-entered the atmosphere as water rather than pure hydrogen. An obvious solution is to surrond the storage container or pipeline in a bed of powdered catalyst and pump oxygen through the bed. Any hydrogen that escapes the container or pipe will combine with oxygen in the catalytic bed, producing water.
This wouldn't solve the leakage problem (you would still lose a fair percentage of your product) but it would easily take care of environmental effects: the qantity of waste hydrogen (emerging as water vapor) would never be more than a small fraction of the total water vapor released into the atmosphere by natural action (plant respiration, evaporation, volcanism, etc.)
I can't help it: every time I read the headline I see "(Still Life) in (the Apple II Community)", as in "Still Life in Sepia", rather than "([There is] Still) (Life in) (the Apple II Community)". I guess it's just my misspent youth in the visual arts.
On Topic, when I was still wasting my youth as a painter, I spent about 6 months writing a simple drawing package for the Apple IIe, complete with a simple vector font library so that I could place text at any size or orientation into my images. This was around 1985-86 and I was using an Apple Lisa 2, at home, to prototype my paintings before committing to canvas (much more expensive than fan-fold paper) but my high-school only had Apple IIes.
I'd already been programming for a number of years, but this was the first time I actually tried to write something big and of my own devising. I stumbled upon the concept of top-down design and step-wise refinement almost immediately, which could probably be taken as an indication that Art was not my real calling.
The IIe was a nice machine, but was awfully slow, even in '85. I remember that, even with a fairly well optimized circle drawing algorithm (a lookup table for sines and cosines and reflection along two axes), you actually had to wait for the figure to draw on screen.
While I don't think that VoD is going to take off, much less erradicate
DVDs (or any other long-term storage du jour), I don't think that it's
reasonable to expect that new and original movies, by any delivery
medium, will be essentially free (I'm calling anything under $5.00
"free"). If you will bear with me, this is my back-of-the-envelope
anaylsis:
One movie probably requires the creative skills of several hundred people
(writers, editors, costume designers, set designers, make-up artists,
carpenters, tailors/seamstresses, camera operators, sound technicians,
lighting technicians, special-effects technicians, and human resource
administrators, not to mention the actors and directors) for, at least,
the better part of a year or two. The cost of equipping and employing
those people for that time is likely to be in the range of $100,000,000
(yes, I know that indie-folk can turn out a compelling and original
story in under a year for less than $100,000, but that kind of production
only goes so far. Every once in a while I'd like to watch a real Hollywood
blockbuster with aliens blowing up the Whitehouse and explosions chasing
people down hallways instead of Clerks or
Smoke). That money has to be
recouped relatively quickly in order to ensure that the studio can go
on to the next project without laying off it's staff (all the techs and
admins, not the actors and directors).
Let's assume that a reasonable size domestic (U.S.) audience is on the
order of 25,000,000 people (about 1/8 of the U.S. population). These are
the people who want to see the movie and will bother to shell out for a
theater ticket. Let's also assume that the theatrical run will be about
2 months. At $8.00/ticket the movie will gross $200,000,000 over it's
run. Since some of that money goes back to the theater owner, and some
to other middle men (distributers, promoters, etc.) let's assume that we
only see half of the ticket-counter gross: $100,000,000. We just manage
to break-even.
Now we need to bank on long-term sales for home-viewing (mass-media
broadcast, rental and direct sale) to get any profit out of this turkey.
We can assume that there is another 25,000,000 people who wouldn't have
watched the movie in the theaters but will see it on TV, or rent/buy it
from the local video store. The TV audience is limited by the amount of
time we can hold the movie back from the rental/retail sales channel, so
we can only expect to get a few showings. I have no idea what the big
networks will pay for the rights to debut a movie on TV, but we can
assume that it's not all that much money, and that it's pretty much a
one-shot deal (After the movie has made it's TV debut, the amount that
the networks are willing to pay to show it again will plummet). We can
assume that a fair amount of the money we get from TV will be eaten up
in the effort needed to make the sale in the first place. Now we are left
with only the rental/retail market to make our profit.
As I said, our rental/retail market is probably about 25,000,000 people.
Some fraction of these folk will probably have already seen the movie
in either the theater or on TV, but let's just neglect them for the moment.
The rental/retail market has a much longer life-span that either the
theatrical release or the TV release (virtually infinite) but the rate
of return is correspondingly lower. Lets assume that we will move only
5,000,000 units per year. At the end of a 5 year period we will have
saturated the remaining market and our income will have dropped to a
trickle (as close to zero as makes no odds).
We need to put out some amount of effort to market the movie to the
rental/retail channel (there is limited shelf space in any bricks and
mortar store) so there is a non-zero cost to every movie that we sell.
We also need to allow the retail outlets to make some profit, so our
return is diminised even further. I will assume, purely for the sake
of argument, that reproduction, packag
The problem with a lot of systems these days is more than just the speakser and the amp. The A/D converter (when playing CDs) has quite a huge impact.
I own a 30K System (yes, that's a lot of money) and come to realize that most AD converters in consumer grade CD Players are just horrible.
...
right now you cannot get a digital out on the players (they are afraid someone is making high grade copies of the CDs) so you have to rely on the internal A/D convert to get the sound out.
Don't you mean D/A (also known as a DAC or digital to analog converter)? The data on the CD is already in digital (D) format and needs to be converted to an analog (A) signal for your speakers. I don't see any place for an analog to digital converter in a CD player.
most of the people I know do not want to feed the amp with an analog signal if possible.
This might just be my EE bias, but my understanding of the definition of the term amplifier is that it refers to an analog defice. Sending digital signals (two values signals, usually V+ and V- or V+ and 0) to an amplifier doesn't do anything, since the amplitude of the signal doesn't carry the information (the information is carried in the pattern of high and low signals: 0V,+5V,0V,+5V,+5V has the same meaning as +3V,+20V,+3V,+20V,+20V).
I suppose, if you wanted to run you digital signal over a very long cable, you might feed it through an amplifier to get sufficient current and voltage to get to the other end, but it won't change the amplitude of the output sound one whit (or Watt, for that matter). More likely, however, you would re-encode the digital data in a format that could detect and correct multi-bit errors (using a Hamming code, for instance, though I'd bet that data on CDs is already encoded in just such a way).
<sarcasm> <irony>Oh yes, import tarriffs have done a wonderfull job of keeping other sorts of jobs in the U.S., just ask all those busy garment workers in New York, or the textile or steel mill workers in the nerth east. They'll certainly agree that import tarriffs have made all the difference in their lives.</irony> </sarcasm>
The real problem is, of course, that the cost of living the U.S. is way too high. In other countries you can live comfortably on half or a third of what it costs to live in the U.S. (and if your willing to live in the developing world, you can get by on a tenth or less) Almost everything is cheaper: housing, food, clothes, the only major exception is gasoline.
The main reason that the U.S. cost of living is insane is that we have never recovered from the inflation of late-seventies/early-eighties: prices went up when comodities surged, but never went down again. Salaries have had to follow suit just to maintain parity.
And all the while there are talking heads claiming that we are the most productive and efficient work-force in the world. If that's so, why do I pay 10-times as much for a shirt here as I would in Viet Nam? It's the exact same shirt! (check the tags at your local department store, at least half will say Made in Vietname or China) Someone is making out like a bandit on this stuff, and I don't think it's the seamstress' and tailors in Hoi An! (They're not doing so badly, by their own standards, but they're still only seeing about $10 of the $50 I paid for my last jacket)
It's all a pack of lies: The workmanship on Vietnamese made products is the equal of the same products made in the U.S. and the Vietnamese workers are getting paid less than a dollar and hour. On top of that, the level of education in Vietnam is at least as good as it is in the U.S. (maybe better) so we aren't talking about unskilled labor. Finally, the Vietnamese are no slouches: they do things fast . Obviosly the productivity of the Vietnamese workers is five or six times that of U.S. workers, by any reasonable measure (output/dollar, output/time, etc.).
Again, it's all becuase the cost of living in Viet Nam is an order of magnitude less than it is in the U.S. and there is no way that the U.S. workforce is ten times faster, or ten times smarter, or ten times more accurate than almost any other workforce in the world. Tarriffs won't change that, they will just ensure that our own cost of living will rise even further, making it even more attractive to send jobs off shore!
So go right ahead, get them tarriffs imposed. I might as well sell my house and move to Viet Nam (a couple hundred thousand dollars will go a long way there, and the scenery can't be beat!). And, heck, it can't be too much worse than living in Bush's Amerika, at least the government in Hanoi doesn't make any pretense to democracy or freedom.
My developer cohorts and I have a running joke about Wheel v2 (actually, we're probably on Wheel v 98.2). Developers love to reinvent the wheel. Good programming practice always needs to start with taking the specification and asking "Is there a product out there that solves this" Often, you find a set of libraries that solve 90% of your problem - costing much less than the development time it'd take to roll your own. Often you'll find that the developer next door has already built that. But it is just so much more fun to write your own than learn to use someone elses code!
My experience has been that, in many organizations, rolling your own code is often faster, easier, and less error prone than trying to make use of something someone else wrote (either internally or from an outside source). Often, the internall code is undocumented and too tailored to the specific task to be applied to another job without significant adaptation (writing support code around the library calls to make up for the differences in application) or refactoring (actually rewriting the library so it is general enough for both tasks). Code from external sources must be vetted by higher-ups (either to get a P.O. for proprietary code, or an Ok to use open source/free software) which is a death warrant once you are pushing a deadline.
I have seen many code reuse initiatives that focus either on replacing the current tools with new tools (this is the equivalent of the HLL vs. C/C++ argument) or on browbeating developers who don't use the existing (undocumented, error-prone) code base. I have rarely seen an organization that is willing to put the effort into both design and documentation that would be required to support code reuse.
The real culprits, both in preventing wide-spread code reuse and in generating masses of insecure code, are cultural, not technical. Organizations just aren't willing to pay for anything other than coding: everything else gets short changed (inadequate specifications, inept design and analysis, no code reuse, and laughable, pro-forma, testing).
When I was in Viet Nam we saw these short strings of digits painted on almost every vertical surface in HCMC. Each number was accompanied by a very short message (one or two words in vietnamese) but no other information whatsoever. When I asked our guide about them he told me that those were 'advertisements' and that the numbers were telephone numbers. He also said that the advertisments were highly illegal (because they were eyesores) and that the police would have any number found thus posted shut off by the telephone company (government run, of course).
While this policy didn't seem to be having a discernable effect in HCMC, we didn't see the advertisements (at least not to the same degree) in other large cities (specifically Da Nang, Hue and Hanoi).
Yes, and that was my original reply to my wife, "I'll just use pencil and paper," and that was, essentially, what I was forced to do anyway. However, I wasn't trying to do any large-scale design work, I just needed to try out a few, small-scale, algorithms that popped into my head. Pencil and paper won't compile or run my code, and I can't run anything of significant size or complexity by hand (a single run of a good test harness can easily exceed anything anyone could do by hand).
Unfortunately, the Fujitsu fails at least two of my requirements: very long battery life (>6 hours) and low cost (I just wouldn't be comfortable dropping a $1200 investment off a cliff. If I could be Ok with that, I'd just get an iBook and a couple of extra batteries). I'm also not too certain how durable the Fujitsu is: It looks a bit flimsy to be dragged around in a backpack for a month or so.
Yes, I was on vacation (in an exotic and beautiful locale) for a month and I wanted to be able to code from time to time. That may sound absolutely nuts but there is one thing most folks here probably don't understand: There was a considerable amount of 'down time' on the vacation, most of it during travel.
First (and last) there is the 18 hour flight over the Pacific ocean. I managed to watch all the in-flight movies about half-way through (including the Hindi ones) and was almost tempted to listen to the religious audio channel (readings from the Quaran, no less). Second, most travel within Viet Nam isn't all that speedy, especially when travelling along the coast (and the scenery gets a bit monotonous after the first couple of days: something like eastern-shore Maryland if you replace corn/tobacco with rice). Third, after trekking around the city all day, Lina was pooped and I was ready for some relaxation from the relaxation: something less exotic than everything I'd been seeing and doing all day, something that streatched other (more familliar) parts of my brain.
So I'm a geek, I need to unwind by writing some recreational code every thirty or forty hours. At least I was able to pack a fewgoodbooks that could tide me over.
Just for the record, the wattage I quoted includes the backlight. And, yes, I would like to see more use of reflective or transflective LCDs, especially since they are much more readable in high light conditions (outdoors, for instance).
Windows NT will exist in the Smithsonian, next to your own favorite operating system, but it is unlikely that either one will be used for projects of this type.
I just don't think this stands up to even the slightest reflection. Fifteen years ago (1988) You had a few choices for large scale computing: you could run an and IBM system (MVS or VM), you could run a DEC system (VAX/VMS or RSX-11), you could run an HP based system (MPE), or you could run some form of UNIX (some offered by IBM, DEC or HP). For medium scale computing you had a number of systems for microcomputers (Novel Netware, Banyan Vines, etc.) most of which are now defunct.
In 1988 UNIX had alread been around for 15 years. The IBM and HP products could claim equal longevity, and the DEC products could trace their lineage back at least as far (though not as the same product). The HP product is now officially discontinued, and the DEC products are likely to follow (if they haven't already). The IBM products show no sign of disappearing, though they tend to change their names periodically. The only one still standing is UNIX (or UNIX-like systems).
Given this history, I don't doubt that UNIX will be alive and kicking in 2018, in many recognizable forms, and will still be used as the basis for mission-ciritical systems of all kinds.
I've been doing some research on just this topic, and you are right about the LCD, but wrong about the peripherals. Even small LCDs (I've been looking at 6"-10" models with resolutions of 640x480 or 800x600) consume 8W-10W. Disk drives consume only a few Watts in normal operation, most of the models I've looked at (2.5" laptop HDs and the IBM microdrive) consume 1W-2W in normal operation and less than 1W in standby.
The real killer is going to be the CPU: Intel (and Intel compatible) devices tend to consume anywhere from 10W-25W in full operation. Their standby modes may be much lower, but what do you care how much power the thing draws when you aren't doing anything? PowerPCs are much better (5W-12W) and ARMs are just astonishing (one of the ARM chips I'm looking at draws less than 1W at full speed)
For must purposes, you can consider that the CPU and LCD consume 80%-90% of the power in your laptop, pretty much evenly divded between the two. If we are talking about a handheld device, the LCD probably eats 60%-70% of the power all on its own.
Another book you sould consider, though it is more geared toward personal computer interfaces, is Bruce Tognazinni's Tog on Interface also from Addison-Wesley (ISBN: 0201608421). Tog is now part of the Nielsen-Norman Group, which can be hired for HCI work, if you have the moolah.
Apples really can chug along happily for many years, often performing
the same tasks that they did when new, many years after they are, in
all other way, obsolete.
Until very recently (the last three or four years) I have had to
upgrade my PC on an annual basis (I had set price marks at which I
would buy new components: a hard drive at about $150, a video card
at about $100, a new motherboard and processor at about $80 each, etc.).
In general, this meant that my PC didn't go more than about 2 years
without a complete replacement, when you count both outright failure
and forced obsolescence (often, the slightly aged motherboard wasn't
able to handle the larger hard drive, or any more memory. Less frequently,
I actaully needed to run software that wouldn't recognize some aging
peripheral or card).
My Macs have been much less upgrade hungry, in general: I haven't maxed
out a system in terms of memory since 1992, and I've never maxed out a
system in terms of disks or removable media. Most of the video or CPU
upgrades have been due to desire more than need, and half of the hard
drive upgrade have been due to age related failure of my current device
(that hasn't happened, actually, since 1997, but was common before that
time).
As for price, I've never spent more than $2500 for a new Mac, and my
used machines tended to cost around $400 (except for the first used
machine, a Mac 512ke, which was about $1500 in 1987). My PCs have all
run around $600 when new, but the replacement cycle is three or four
times faster than the Macs, so it evens out.
My Mac history looks something like this:
1985 new Lisa 2 (8MHz 68K, 512KB, 5MB HD)
upgrade memory (1MB)
add 2nd hard drive (5MB HD)
upgrade memory (2MB)
1987 used Mac 512ke (512KB, 400K FD)
1988 new Mac Plus (8MHz 68K, 1MB, 800K FD)
add hard drive (20MB)
upgrade memory (2MB)
replace hard drive (100MB)
upgrade memory (4MB)
1993 used Mac II (16MHz 68020, 4MB, 40MB HD)
upgrade memory (8MB)
replace hard drive (250MB)
upgrade CPU (33MHz 68030)
upgrade video card (Apple accelerated video)
1996 new PowerMac 8500 (132MHz, 1GB HD, 16MB)
replace hard drive (2GB)
upgrade memory (+16MB)
add Zip Drive
upgrade memory (+32MB)
add CD-RW
add hard drive (4GB)
upgrade video card (Rage Pro 3D)
upgrade memory (+64MB)
upgrade video card (Rage 128 VR)
upgrade CPU (G3/350)
upgrade memory (+128MB)
2002 used PowerMac B&W G3 (350MHz, 6GB HD, 64MB)
upgrade video card (Radeon 7500)
upgrade memory (+256MB)
replace hard drive (40GB)
replace CD with CD-RW
replace CD-RW with DVD/CD-RW
As you can see, the new machines last for about 5 years, the used machines for about 2. This is a pretty good lifespan for machines that serve as my everyday workhorses. I have used these machines for word processing/desktop publishing, graphics, telecommunications (bulletin boards prior to 1995, web and internet after), and programming (BASIC, then Pascal, then C/C++) for each of their entire lifetimes.
Slashdot Mirror
For a moment there I was afraid that this would kill my own pet project, but after seeing what the power consumption is like, I am not worried.
Next, you will need to rig a larger ink reservior for the print head. This should be pretty easy if you have any mechanical aptitude and access to a machine shop. You also need to secure a source of compatible ink.
I'm tempted to say you should skip the print head altogether and use a simple spray nozzle (disassemble a cheap airbrush with a fine nozzle and actuate the needle with a solenoid), but this would mean that you would be working at very low resolution (maybe 10 or 20 dpi). You will still need to maintain a reasonably constant distance from the media, but you tolerances will be less demanding and you will have a wider choice of inks.
Finally, you need a three axis robot to position the head over the media. You can sense the distance from the media optically (a couple photodiodes, some lenses and a diode laser should do the trick). Getting exact positioning of the head in the X-Y direction is a bit harder (good mechanical positioning over 2 meters will be expensive) but you can probably use an ultrasonic system and get reasonable results (there used to be digitizers for CAD use that had an ultrasonic clicker in the pen/puck and a pair of microphones on perpendicular edges of the tablet. Either the delay from click to reception, or the amplitude at the mics, was used to determine the pen/puck position). Just put a ultrasonic clicker on the print head, and click a couple of times on each X-Y positioning operation to fix the head location.
Assemble the above parts, along with some custom hardware and software for interface and control, and you should be good to go.
Oh, only a few thousand, but we don't know the real number because if the government had to reveal who it has secretly arrested THEN THE TERRORISTS HAVE WON!
Many of the current MCUs are ball grid array (BGA) devices, which make them pretty hard to work with if you're not a professional, but a few can be had in PLCC or QFP packages, which means you can get an adapter board or socket.
You can also buy preassembled demo/development boards (this is the route I'm taking) and wire-up anything the board doesn't include by hand. Most of the MCUs on the market will have 32-bit memory busses (though they may not support more that 25 or 26 address lines), so you can attach just about anything you want to them.
This is exactly the kind of thing that Steve Ciarcia (of Ciarcia's Circuit Cellar) used to do: building personal computers from microcontrollers. Most of his designs used Z80 based devices, which was fine back in the mid- to late-eighties. Now, however, you can do a fair bit better.
As for speed, I don't know exactly what you're looking for, but the ARM devices can be had in speeds from 50MHz to 400MHz, and the same is true of the PPC and MIPS devices. That may not seem like much, compared to a 2GHz Pentium, but it's really quite nice.
Some good resources: Digi-Key is a reasonable source for all sorts of parts, Atmel makes some nice MCUs, programmable logic, and Flash RAM, Cirrus Logic makes some ARM MCUs and networking chips (amoung other things), Sharp, Samsung, Motorola, and AMD all make nice MCUs, Cogent Computers builds some nice development boards, and EarthLCD has good prices on LCDs and has an ARM based board in the works.
If you check out my journal you will get a detailed description of what I've been doing and why. In a nutshell: I'm trying to build a handheld computer that delivers adequate performance with maximum flexibility and battery life. The target design will have a small (7-8 inch) VGA screen, a moderately slow processor (100-200 MHz ARM), moderately sized hard disk (10-20 GB, 2" IDE), a moderate amount of RAM (32-256 MB), and run off of conventional rechargable batteries (8 AA NiMH cells) for at least 8-hours on a single charge.
Just to build the prototype will take several months and cost at least $1000 (probably several times that, when all is said and done). If I go to production, I can probably get the cost down to $300 per unit (not counting NRE costs), but the selling price will still be up in the $500 range.
The only way I am able to do any of this is that I am abandoning any kind of PC compatability: This device will never be able to run Windows or play flashy games (though it will run a regular, non-embedded, version of Linux). I've had to give up on all kinds of features that seemed like they should be simple: I don't have a clamshell case because the engineering is too complex for me to do myself and I don't have a keybaord because I couldn't find a source for laptop-style keyboards. I'm still holding out for polymer-LiION batteries, but I haven't found anywhere that builds them on contract, yet.
This is not the same sort of task as building a desktop computer from white-box parts (which is only slightly harder than plugging together lego blocks). This is much closer to actual engineering: I'll be constructing some of my own breadboards to connect the SBC to the LCD and the IDE hard disk, as well as custom hardware to charge and monitor the batteries.
Second, some books:
Third, don't skimp on the methematics! Mathematics, especially calculus, underlies all of modern science. You can't really understand most of the science without understanding calculus, and if you understand calculus, much of the science will simply fall into place.
Finally, don't be too proud or stubborn to actually go back to school. You can enroll in a night course at the local community college for less than the cost of a bare-bones PC-clone. A little actual instruction goes a lot farther than a whole lot of unaided reading. You might also have some fun. (it's amazing how much fun learning is when you're not worried about getting the piece-of-paper)
If you can learn the basics of physics and math, you will be able to cope with (if not master) just about anything. Anything you weren't actually taught, you will be able to get a reasonable grasp of after a few days (sometimes only a few hours) of reading and thinking. If something takes longer than that, you have hit upon a trully difficult subject and may need to look for a course to take.
The marketing-speak "Industry Leading Processors" is awfully suspicious. The sad part is, SGI doesn't have any good options:
SGI has tried just about every dumb trick in the book (most pioneered by DEC) to find some way to move from thier ever shrinking niche (data visualization and computer animation) to something broader and more profitable. At each step along the way they have annoyed and alienated their loyal customers.
As for laying the blame for this attitude at Microsoft's feet: I don't think that's entirely fair. Microsoft certainly benefits from frequent replacement of IT equipment, but there are plenty of other benficiaries who have contributed: Intel is a obvious culprit, since they only stay in business if they can convince people to replace existing systems with brand-spanking-new kit. Another culprit, however, is not nearly so malign: Moore's Law has reduced the cost of all computing hardware to the point that many systems are cheaper to replace than to repair or, even, maintain.
First: Feh! on the business-speak drivel.
Second: The author is just plain looney. IT is not an investment any more than sticky notes are, and for precisely the same reason: Most of your IT hardware will need replacement in very short order (only a few years).
If any part of the IT department is an investment it is the expertise of the people you employ to administer your IT, but either you (as a company) already know this (by treating all your employees as investments) or you don't (which, it appears, is the far more common attitude). Even the employess are not a very good investment, since there is damn little you can do to make sure they don't leave (you can't own your employees, at least not in most places) and they require a huge amount of maintenance (salaries, benefits, training, etc.).
Everything else about IT is obviosly a simple cost (possibly amortized over a short period of time): All the hardware will be obsolete within only a few years. All the data you collect has an almost equally short shelf life (how long are individual customer demographics good for? how about marketing materials? maybe financial data has a reasonable lifespan, but it's not much more than 5 or 7 years) and have a continuing cost of maintenance in order to keep them up-to-date.
The longest lived part of your IT infrastructure is probably the network cables, which can last 10 years or so. None of these items have the lifespan of a building, however, which is the canonical example of an investment item.
In fact, even if none of the electronics were designed by Apple, the most important part "originated in Steve's back yard:" What other company has the guts and determination to mass-produce non-x86 (and, for that matter, non-Windows) hardware targeted at the consumer desktop? Apple is the only company left standing from the great microcomputer ecology of the seventies and eighties, and the only one that was willing to stay the course, no matter how difficult.
So, sure, Apple farms out huge parts of their engineering effort, but the only way this sets them apart from any other consumer manufacturer, is that they do more custom engineering than any of their competition! (Does anyone really think that Dell or Gateway are doing anything more than throwing together a bunch of white-box parts and sticking them in a fancy box?)
I'll admit that calling this the culminatio of Apple's technology may be gilting the rose, but it is certainly the culmination of Apple's vision and dedication. Without Apple this sort of technology would never have been brought to a desktop near you, or, at best, it would still be years in the future.
Because hydrogen molecules are really small, much smaller than any of the other gasses you mentioned (methane, butane, propane and acetylene are all hydrocarbons, which means they are fairly large: at leat one carbon and four hydrogens. Even the noble gasses are pretty big, with the exception of helium) and have a tendancy to leak through solid metal pipes! See this site for some details.
That said, I can think of a few ways to build storage containers and pipelines that would ensure that leaked hydrogen re-entered the atmosphere as water rather than pure hydrogen. An obvious solution is to surrond the storage container or pipeline in a bed of powdered catalyst and pump oxygen through the bed. Any hydrogen that escapes the container or pipe will combine with oxygen in the catalytic bed, producing water.
This wouldn't solve the leakage problem (you would still lose a fair percentage of your product) but it would easily take care of environmental effects: the qantity of waste hydrogen (emerging as water vapor) would never be more than a small fraction of the total water vapor released into the atmosphere by natural action (plant respiration, evaporation, volcanism, etc.)
On Topic, when I was still wasting my youth as a painter, I spent about 6 months writing a simple drawing package for the Apple IIe, complete with a simple vector font library so that I could place text at any size or orientation into my images. This was around 1985-86 and I was using an Apple Lisa 2, at home, to prototype my paintings before committing to canvas (much more expensive than fan-fold paper) but my high-school only had Apple IIes.
I'd already been programming for a number of years, but this was the first time I actually tried to write something big and of my own devising. I stumbled upon the concept of top-down design and step-wise refinement almost immediately, which could probably be taken as an indication that Art was not my real calling.
The IIe was a nice machine, but was awfully slow, even in '85. I remember that, even with a fairly well optimized circle drawing algorithm (a lookup table for sines and cosines and reflection along two axes), you actually had to wait for the figure to draw on screen.
While I don't think that VoD is going to take off, much less erradicate DVDs (or any other long-term storage du jour), I don't think that it's reasonable to expect that new and original movies, by any delivery medium, will be essentially free (I'm calling anything under $5.00 "free"). If you will bear with me, this is my back-of-the-envelope anaylsis:
One movie probably requires the creative skills of several hundred people (writers, editors, costume designers, set designers, make-up artists, carpenters, tailors/seamstresses, camera operators, sound technicians, lighting technicians, special-effects technicians, and human resource administrators, not to mention the actors and directors) for, at least, the better part of a year or two. The cost of equipping and employing those people for that time is likely to be in the range of $100,000,000 (yes, I know that indie-folk can turn out a compelling and original story in under a year for less than $100,000, but that kind of production only goes so far. Every once in a while I'd like to watch a real Hollywood blockbuster with aliens blowing up the Whitehouse and explosions chasing people down hallways instead of Clerks or Smoke). That money has to be recouped relatively quickly in order to ensure that the studio can go on to the next project without laying off it's staff (all the techs and admins, not the actors and directors).
Let's assume that a reasonable size domestic (U.S.) audience is on the order of 25,000,000 people (about 1/8 of the U.S. population). These are the people who want to see the movie and will bother to shell out for a theater ticket. Let's also assume that the theatrical run will be about 2 months. At $8.00/ticket the movie will gross $200,000,000 over it's run. Since some of that money goes back to the theater owner, and some to other middle men (distributers, promoters, etc.) let's assume that we only see half of the ticket-counter gross: $100,000,000. We just manage to break-even.
Now we need to bank on long-term sales for home-viewing (mass-media broadcast, rental and direct sale) to get any profit out of this turkey. We can assume that there is another 25,000,000 people who wouldn't have watched the movie in the theaters but will see it on TV, or rent/buy it from the local video store. The TV audience is limited by the amount of time we can hold the movie back from the rental/retail sales channel, so we can only expect to get a few showings. I have no idea what the big networks will pay for the rights to debut a movie on TV, but we can assume that it's not all that much money, and that it's pretty much a one-shot deal (After the movie has made it's TV debut, the amount that the networks are willing to pay to show it again will plummet). We can assume that a fair amount of the money we get from TV will be eaten up in the effort needed to make the sale in the first place. Now we are left with only the rental/retail market to make our profit.
As I said, our rental/retail market is probably about 25,000,000 people. Some fraction of these folk will probably have already seen the movie in either the theater or on TV, but let's just neglect them for the moment. The rental/retail market has a much longer life-span that either the theatrical release or the TV release (virtually infinite) but the rate of return is correspondingly lower. Lets assume that we will move only 5,000,000 units per year. At the end of a 5 year period we will have saturated the remaining market and our income will have dropped to a trickle (as close to zero as makes no odds).
We need to put out some amount of effort to market the movie to the rental/retail channel (there is limited shelf space in any bricks and mortar store) so there is a non-zero cost to every movie that we sell. We also need to allow the retail outlets to make some profit, so our return is diminised even further. I will assume, purely for the sake of argument, that reproduction, packag
Don't you mean D/A (also known as a DAC or digital to analog converter)? The data on the CD is already in digital (D) format and needs to be converted to an analog (A) signal for your speakers. I don't see any place for an analog to digital converter in a CD player.
This might just be my EE bias, but my understanding of the definition of the term amplifier is that it refers to an analog defice. Sending digital signals (two values signals, usually V+ and V- or V+ and 0) to an amplifier doesn't do anything, since the amplitude of the signal doesn't carry the information (the information is carried in the pattern of high and low signals: 0V,+5V,0V,+5V,+5V has the same meaning as +3V,+20V,+3V,+20V,+20V).
I suppose, if you wanted to run you digital signal over a very long cable, you might feed it through an amplifier to get sufficient current and voltage to get to the other end, but it won't change the amplitude of the output sound one whit (or Watt, for that matter). More likely, however, you would re-encode the digital data in a format that could detect and correct multi-bit errors (using a Hamming code, for instance, though I'd bet that data on CDs is already encoded in just such a way).
The real problem is, of course, that the cost of living the U.S. is way too high. In other countries you can live comfortably on half or a third of what it costs to live in the U.S. (and if your willing to live in the developing world, you can get by on a tenth or less) Almost everything is cheaper: housing, food, clothes, the only major exception is gasoline.
The main reason that the U.S. cost of living is insane is that we have never recovered from the inflation of late-seventies/early-eighties: prices went up when comodities surged, but never went down again. Salaries have had to follow suit just to maintain parity.
And all the while there are talking heads claiming that we are the most productive and efficient work-force in the world. If that's so, why do I pay 10-times as much for a shirt here as I would in Viet Nam? It's the exact same shirt! (check the tags at your local department store, at least half will say Made in Vietname or China) Someone is making out like a bandit on this stuff, and I don't think it's the seamstress' and tailors in Hoi An! (They're not doing so badly, by their own standards, but they're still only seeing about $10 of the $50 I paid for my last jacket)
It's all a pack of lies: The workmanship on Vietnamese made products is the equal of the same products made in the U.S. and the Vietnamese workers are getting paid less than a dollar and hour. On top of that, the level of education in Vietnam is at least as good as it is in the U.S. (maybe better) so we aren't talking about unskilled labor. Finally, the Vietnamese are no slouches: they do things fast . Obviosly the productivity of the Vietnamese workers is five or six times that of U.S. workers, by any reasonable measure (output/dollar, output/time, etc.).
Again, it's all becuase the cost of living in Viet Nam is an order of magnitude less than it is in the U.S. and there is no way that the U.S. workforce is ten times faster, or ten times smarter, or ten times more accurate than almost any other workforce in the world. Tarriffs won't change that, they will just ensure that our own cost of living will rise even further, making it even more attractive to send jobs off shore!
So go right ahead, get them tarriffs imposed. I might as well sell my house and move to Viet Nam (a couple hundred thousand dollars will go a long way there, and the scenery can't be beat!). And, heck, it can't be too much worse than living in Bush's Amerika, at least the government in Hanoi doesn't make any pretense to democracy or freedom.
My experience has been that, in many organizations, rolling your own code is often faster, easier, and less error prone than trying to make use of something someone else wrote (either internally or from an outside source). Often, the internall code is undocumented and too tailored to the specific task to be applied to another job without significant adaptation (writing support code around the library calls to make up for the differences in application) or refactoring (actually rewriting the library so it is general enough for both tasks). Code from external sources must be vetted by higher-ups (either to get a P.O. for proprietary code, or an Ok to use open source/free software) which is a death warrant once you are pushing a deadline.
I have seen many code reuse initiatives that focus either on replacing the current tools with new tools (this is the equivalent of the HLL vs. C/C++ argument) or on browbeating developers who don't use the existing (undocumented, error-prone) code base. I have rarely seen an organization that is willing to put the effort into both design and documentation that would be required to support code reuse.
The real culprits, both in preventing wide-spread code reuse and in generating masses of insecure code, are cultural, not technical. Organizations just aren't willing to pay for anything other than coding: everything else gets short changed (inadequate specifications, inept design and analysis, no code reuse, and laughable, pro-forma, testing).
While this policy didn't seem to be having a discernable effect in HCMC, we didn't see the advertisements (at least not to the same degree) in other large cities (specifically Da Nang, Hue and Hanoi).
Yes, and that was my original reply to my wife, "I'll just use pencil and paper," and that was, essentially, what I was forced to do anyway. However, I wasn't trying to do any large-scale design work, I just needed to try out a few, small-scale, algorithms that popped into my head. Pencil and paper won't compile or run my code, and I can't run anything of significant size or complexity by hand (a single run of a good test harness can easily exceed anything anyone could do by hand).
Unfortunately, the Fujitsu fails at least two of my requirements: very long battery life (>6 hours) and low cost (I just wouldn't be comfortable dropping a $1200 investment off a cliff. If I could be Ok with that, I'd just get an iBook and a couple of extra batteries). I'm also not too certain how durable the Fujitsu is: It looks a bit flimsy to be dragged around in a backpack for a month or so.
First (and last) there is the 18 hour flight over the Pacific ocean. I managed to watch all the in-flight movies about half-way through (including the Hindi ones) and was almost tempted to listen to the religious audio channel (readings from the Quaran, no less). Second, most travel within Viet Nam isn't all that speedy, especially when travelling along the coast (and the scenery gets a bit monotonous after the first couple of days: something like eastern-shore Maryland if you replace corn/tobacco with rice). Third, after trekking around the city all day, Lina was pooped and I was ready for some relaxation from the relaxation: something less exotic than everything I'd been seeing and doing all day, something that streatched other (more familliar) parts of my brain.
So I'm a geek, I need to unwind by writing some recreational code every thirty or forty hours. At least I was able to pack a few good books that could tide me over.
Just for the record, the wattage I quoted includes the backlight. And, yes, I would like to see more use of reflective or transflective LCDs, especially since they are much more readable in high light conditions (outdoors, for instance).
I just don't think this stands up to even the slightest reflection. Fifteen years ago (1988) You had a few choices for large scale computing: you could run an and IBM system (MVS or VM), you could run a DEC system (VAX/VMS or RSX-11), you could run an HP based system (MPE), or you could run some form of UNIX (some offered by IBM, DEC or HP). For medium scale computing you had a number of systems for microcomputers (Novel Netware, Banyan Vines, etc.) most of which are now defunct.
In 1988 UNIX had alread been around for 15 years. The IBM and HP products could claim equal longevity, and the DEC products could trace their lineage back at least as far (though not as the same product). The HP product is now officially discontinued, and the DEC products are likely to follow (if they haven't already). The IBM products show no sign of disappearing, though they tend to change their names periodically. The only one still standing is UNIX (or UNIX-like systems).
Given this history, I don't doubt that UNIX will be alive and kicking in 2018, in many recognizable forms, and will still be used as the basis for mission-ciritical systems of all kinds.
The real killer is going to be the CPU: Intel (and Intel compatible) devices tend to consume anywhere from 10W-25W in full operation. Their standby modes may be much lower, but what do you care how much power the thing draws when you aren't doing anything? PowerPCs are much better (5W-12W) and ARMs are just astonishing (one of the ARM chips I'm looking at draws less than 1W at full speed)
For must purposes, you can consider that the CPU and LCD consume 80%-90% of the power in your laptop, pretty much evenly divded between the two. If we are talking about a handheld device, the LCD probably eats 60%-70% of the power all on its own.
Another book you sould consider, though it is more geared toward personal computer interfaces, is Bruce Tognazinni's Tog on Interface also from Addison-Wesley (ISBN: 0201608421). Tog is now part of the Nielsen-Norman Group, which can be hired for HCI work, if you have the moolah.
Until very recently (the last three or four years) I have had to upgrade my PC on an annual basis (I had set price marks at which I would buy new components: a hard drive at about $150, a video card at about $100, a new motherboard and processor at about $80 each, etc.). In general, this meant that my PC didn't go more than about 2 years without a complete replacement, when you count both outright failure and forced obsolescence (often, the slightly aged motherboard wasn't able to handle the larger hard drive, or any more memory. Less frequently, I actaully needed to run software that wouldn't recognize some aging peripheral or card).
My Macs have been much less upgrade hungry, in general: I haven't maxed out a system in terms of memory since 1992, and I've never maxed out a system in terms of disks or removable media. Most of the video or CPU upgrades have been due to desire more than need, and half of the hard drive upgrade have been due to age related failure of my current device (that hasn't happened, actually, since 1997, but was common before that time).
As for price, I've never spent more than $2500 for a new Mac, and my used machines tended to cost around $400 (except for the first used machine, a Mac 512ke, which was about $1500 in 1987). My PCs have all run around $600 when new, but the replacement cycle is three or four times faster than the Macs, so it evens out. My Mac history looks something like this:
As you can see, the new machines last for about 5 years, the used machines for about 2. This is a pretty good lifespan for machines that serve as my everyday workhorses. I have used these machines for word processing/desktop publishing, graphics, telecommunications (bulletin boards prior to 1995, web and internet after), and programming (BASIC, then Pascal, then C/C++) for each of their entire lifetimes.