Might as well toss in a link to a real astrolabe: here's the Electric Astrolabe, a very capable program that turns your computer into possibly the wolrd's most functional astrolabe.
This is one of the last DOS programs I run. It makes an interesting screensaver, too...
England England and Wales Great Britain The United Kingdom of Great Britain and Northern Ireland
My understanding (I'm a fifth generation Texan myself) is that there is still some considerable disagreement on whether "Great Britian" really and truly includes Scotland, a.k.a. Caledonia. And the current "devolution" actions, including the relatively recent acceptance of a Scottish Parliament, would tend to indicate that there's something to that argument...
X is all wrong for devices like this. Qtopia, on the other hand, which is what Sharp was smart enoug to put in them to start with, is a very good fit. This is simply a hacking tour-de-force. Sure, you can do it, and run X on an X-scale PDA, but it's *stupid* to do so. Especially since Qtopia lets you port Qt applications with minimal fuss. I suppose this gives the rabid QT haters somethign to do with their spare time, though, so it's not all bad...
That said, for machining where the precision is more important meteric makes a good system, although Imperial/English knew this and measurses precision measurements in thousandths or hundreths of inches to gain the advantage of decimal conversions and the increased precision, but metric is quite good for these types of fabrication.
Actually, inches are far superior for ordinary machining tasks, too. Most machine tools can hold a tolerance (determined by physics, not thier designers) of a little less than a thousandth of an inch. A hundredth of a millimenter (a centimiliimeter) is way too sloppy - more than two and a half times that size. A millimillimeter (a micrometer) is really too small though. The thousandth of an inch is extremely useful, and one of the reasons why we in the US are not going to give up useful units just because the rest of you decided to follow the French into a system guaranteed to create floating point math errors. Like pretty much all French ideas, SI can't stand any sort of scrutiny in the real world...
(See my other post here about the deca/deka-da/dk cockup for more information - the SI isn't even consistent with *itself*!)
The Chinese used 16 as the base for their measurement systems. (I know this is true of their system for weights used in commerce, is it true for other units as well? Anybody know?)
This can be easily demonstrated by traditional abacuses, which have 2 five-beads and 5 one-beads, as opposed to "modern" abacuses (a fairly recent invention) with one five-bead and 4 one-beads.
But one thousanth of an inch is a perfect measuring stick for cutting. Not one millimeter, not one micron, but a thousanths of an inch. Other measurements work out well with an inch too. Its just convience of scale, and the inch works well.
What's the matter with you? Are you trying to tell me that it's easier NOT to work in units of a some odd amount more than 2 and a half centi-millimeters?:-)
Seriously, the thousandth of an inch, like the degree F, is one of those units that is just a lot more inherently matched to the what people use them for in the real world than the metric alternatives.
Your cars don't have km/h on the speedometer as well as mi/h? Just about every car I've seen in the US has both units on the speedometer. Older cars, obviously, may not have this feature.
Actually, I hadn't thought of it before, but a "MPH-only" speedometer is not a bad criteria for buying a car. And you avoid all that spyware/monitoring crap that's built in to all the new ones. (Not to mention, if you've got any sense, you'll make sure you get something with that unmistakable big V-8 rumble...)
I much prefer units I can relate to, personally. If the metric system has given us nothing else (which it hasn't), at least we have the Centigrade scale. I'm all for keeping the old-fashioned units alive, but really! Who thought water freezing at 32 and boiling at 212 was a sensible scale?
This is one place where English units *definetely* make much more sense than Metric/SI units, *especially* for everyday use:
Tha problem is that a cEntigrade/Celsius degree is just too damn big to really be useful, especially for temperatures that really matter to *people*.
For instance: I used to own an Alfa Romeo 164S that, like many "metric" cars, had a climate control system that "thought" in Celsius degrees. The problem is, they're just too big to allow fine enough control for comfort: setting an AC system at 72 degrees and letting it try to stay within a degree of that works fine with Fahrenheit degrees, but doing the same thing, with a one-degree tolerance in Celsius degrees results in a temperature swing that is almost *twice* (9/5, actually) as large, so the system cannot really keep things comfortable.
Any sort of reasonable thermostat in Celsius has to resort to using half-degrees.
Who the heck cares where water boils? Especially since pretty much everyone (even metric bigots) knows that it boils at 212 degrees F.
Fahrenheit degrees are just more useful in the real world. (And they also make having to deal with negative degrees fairly rare, unless you live someplace unsuited for human habitation anyway...)
All you need to know, it's the universal prefixes (vocabulary, not math skills): Mega-, kilo-, deca-, centi-, milli-, etc (and really for everyday life you only really ever use 3 of them: kilo-, centi-, milli-).
Your post hits the nail on the head - the Metric system (SI for the pedantic) is anything but consistent!
You cite "deca" - there's NO SUCH PREFIX. In fact, the entire deci/deka mess turned into such a morass of ISO incompetency that the official prefixes are now wildly inconsistent even within the SI system. (I discovered this recently when I looked it up and realized they'd "changed" the metric system since I learned it in the late 1970s):
The prefix for "deka" or "deca" (even the spelling is inconsistent!) is deplorable - it's listed as either da or dk, depending on where you look, and I'd argue *both* of these are wrong, since they break the otherwise inviolable rule of SI prefixes being a single character.
Thus the poor Metric-loving product of European public school education who sees 3.24 dkm may (quite reasonably) wonder what other metric-loving fool would specify a distance in "decikilometers"! (Oddly, some metric preferences would make this an acceptable if quite confusing usage, since they discourage the use of less-commonly used prefixes like "hecto" in favor of more common ones like "deci" and "kilo". Only if the user understands that magnitude prefixes should never be compounded can he avoid this mistake.)
When I learned the Metric system, "deka" was recommended to be represented with an upper-case "D" to distinguish it from the lower-case "d" representing "deci". Apparently, European outcomes-based educators thought this was too confusing to teach, and so lobbied to have it changed, although apparently they failed to expunge the far-too-commonly-used prefix overloading of using the letter "M" to represent both "mega" and "milli" based on case-sensitivity. (The formerly upper case "K" for "kilo, apparently also fell at this time, replaced by the now-standard lower-case "k", breaking the at-least-sort-of-sensible former property of SI prefix symmetry, which at that time held that upper case prefixes represented a positive exponent of the 10, while lower-case ones represented a negative exponent.
So instead, they decided that "deka" would stand alone in all of SI nomenclature, and use a two-character prefix, but apparently, in true ISO fashion, they couldn't agree on even what those two characters should be, "da" or "dk"? (Look around, and you'll find both, even in "official" publications from a number of countries that have adopted the SI as "standard".
(And then of course, there's the pathologically broken use of the lower-case Greek mu as the prefix for "micro". This results in a unit that cannot be typed directly on the vast majority of the world's keyboards, and which is at the mercy of extended character maps that often do not successfully carry the character across different applications, operating systems, or displays, thus rendering yet another prefix at least halfway useless.)
Now Metric-bigots will respond, "But no one uses deci much or deka at all, so who cares?" Those same metric-bigots should, because the bureaucratic bungling that has surrounded SI politics since the beginning is what has has rendered them unusable. If you can't actually use all those decimal prefixes, because doing so creates more confusion than it resolves, then how can you possibly claim that the Metric system is easier, more logical, or more consistent than the English system. You can't - the Metric system is every bit as arbitrary and capricious as the English system, but much harder to use in the real world, because 10 is such a difficult number.
Why do you think hours are 60 minutes, or circles 360 degrees? Because 60 is evenly divisible by 2,3,4,5,6,10,12,15,20, and 30! Try plotting your ship's course in radians sometime and see how long you can sail before running aground - talk about a stupid measurement!
Re:Imperial, not English... Re:Imperial, not English... (Score:1) by Bob Uhl (30977) on Monday January 26, @12:03PM (#8089873) (http://latakia.dyndns.org/blosxom/blo g) Of course I do. Let's say I wish to cut a foot-wide piece of wood into inch-width segments: I cut in half, then in half again, then in thirds. How would I cut a decimetre-wide piece into centimetres? In half, then into fifths? Ever try eyeballing a fifth? There are very easy ways to estimate a half or third quite accurately, but I am aware of none to estimate a fifth.
Let's say I have 1 gallon of beer and need to serve it in 1 pint measurements. Dead simple: I cut in half (yielding pottles), in half again (yielding quarts) and in half one last time, yielding pints. Doubling and halving are extremely easy with liquids and masses; anything else is a right royal pain in the ass.
Try dividing a decilitre into centilitres or millilitres. Good luck.
The standard system, used as it was for millennia, was optimised for use, for manipulation of concrete amounts. The French system was optimised for conversion between units. The one is needed daily; the other almost never (now that we have computers, practically never). Which would an unbiased observer prefer?
The standard system is, of course, imperfect and could use certain improvements. Its basis, though, is sound. The French system is also imperfect, but its sole basis is a silly attachment to 10 (a mathematically ugly number anyway). I'd rather spend effort on the system with a future.
Someone with mod points please mod up parent. He raises some *very* valid points that are far too often overlooked. This is one of the most important reasons why the engineering community her in the US will continue using Englinsh units for the foreseeable future. (I fully expect the US to remain primarily on English units through this century, and possibly far beyond...)
My company just designed a new mechanical product, and it was decided that English rather than Metric units were preferable for many reasons including ease-of-use, marketing considerations (even though half the unit volume will be international), and availability/selection and lower cost of hardware.
Could you provide same results from a plausible source - creationists and their well-known lies don't come even close to being anything more than a good laugh.
Actually, if you'd bother to read the links and their footnotes, you'd see the original sources are published science and comments of scientists. The site itself is run by uber-programmer David Pogge, a.k.a "Do-While Jones", who has achieved the high honor of being named a fellow at the US Navy's China Lake Weapons Center as one of the world's leading experts on the hardware and software design of guided missile systems.
Most of the quotes in the article are from the likes of Science and Nature, sources not noted for their "creationist" sympathies. There is no credibility gap here, unless its yours. They said what they said, in print, and what they said clearly does not support evolution.
In any case, Pogge's site, Science Against Evolution is not at all creationist, but simply points out that science actually does anything but support the theory of evolution. In this case, it is the scientist's own words and backtracking from their unjustifiable assertion about this "Mars rock" that make the links so interesting.
I can see it now - Advent(ure): the Dilbert Edition...
You are in a board room. Its long, polished oval table and leatherette chairs are quite intimidating. A filter coffee machine bubbles quietly in the corner.
A Board Member is here. A Chairman is here. An Executive Directory is here. An Axe is on the floor.
Your PHB has entered the room and taken a seat at the table.
> Kill PHB
Kill a PHB?!! With what? Your bare hands?
> Yes
Congratulations, you have just vanquished a PHB with your bare hands!
A woman with tall triangular hair has entered the room. She nonchalantly kicks the PHB's corpse into the corner and takes his seat.
The chairman is shocked at your actions, but announces that he is granting you options on 1000 shares in this moribund subsidiary of a large but ultimately doomed computer company. If the entire corporate behemoth exceeds expectations by at least 30% for the next 12 quarters in a row (by somehow figuring out how to make money off Linux), those options won't even be underwater anymore.
> Refuse options
You are in a maze of tiny, light beige, but still somehow dark, cubicles, all alike...
There is a woman with tall triangular hair here, laughing at you in a hurtful way...
> It never ceases to amaze me that in the world of "science" that theory is most always talked of as proven fact. If the probability that object x doesn't exist is 80%, that object is talked about as if it doesn't exist conclusively, even though the possiblity that it exists is in the other 20%t. The age of the universe is a good example. Scientist y estimates to to being x Billion years old, and every textbook, news article, journal, documentary, etc states that the universe is x billion years old as if proven fact. The truth is that no-one currently knows exactly whether it's one billion or a billion billion years old.
That's why scientists prefer convergent evidence over one individual's opinion.
BTW, AFAICT astronomers are now nearly unanimous on an age of 13.7 billion years. This is a fairly recent result.
Your "convergent evidence", though, is not really any such thing, but just a general agreement on what is politically acceptable, not scientifically justified.
It's vitally important to understand that these dates are in reality based solely on conjecture, since every radioactive dating method rests on unfounded assumptions - assumptions that produce results that are known to be grossly and wildly inconsistent in the few cases we have where the ages of rocks *are* known precisely. The assumptions are "tweaked" to produce the desired age, not the other way around...
From the article: The team that found it was led by experienced meteorite hunters Carine Bidaut and Bruno Fectay, who have now found six rocks from Mars - a record.
Interesting that they seem to know *just* where to find Martian rocks.
It's also really interesting that the last big hoo-rah about finding "a rock from Mars" here on Earth coincided with Bush Sr's proposal for a mission to Mars. What's really amazing is that these discoveries are so strongly correlated to Congressional consideration of tens or hundreds of billions of dollars for Mars research! (Perhaps we should fund a $40 million NASA/CBO research study to determine if this correlation can be linked to any sort of causality? Hmmm?)
There is absolutely no real evidence that these rocks are from Mars, and in fact, NASA very quickly backed away from defending the Mars origin theory when pressed to provide scientific backup. There is absolutely no substantial reason to believe that ALH84001 came from Mars, other than that it sure seems to help get funding.
Of course, if we go to Mars, it should be without NASA - how we could think that a government agency with a worse operational safety and efficiency record than the Post Office should run a Mars mission is beyond me. I've worked at NASA, and belive me, the only real soultion is to bulldoze the entire organization and start over - even thenm you'd need to wait ten years before starting over to avoid the revolving door effect...
Someone with mode points mod pls mod parent up. He makes some very good points, although he'll probably get modded to the basement because he (correctly) points out that CE is indeed winning the OS battle in embedded devices.
This is very likely caused at least in part by the fact that the embedded Linux market is extremely fragmented, with no real agreement or standards, especially w.r.t the visual environment: There are literally dozens of incompatible embedded GUI choices, none of which have enough momentum to succeed. (And all of which rely on wildly inconsistent programming toolsets and methods.)
As a result, CE, which is almost as fragmented across platforms, but at least has centralized and coordinated developer support, wins without a battle.
In reality, Linux is ill-suited today for use in embedded devices that requires a GUI, since getting comaprable functionality requires GUI libraries and tools that are just as proprietary as MS, but without the benefit of mind and market share or a big pot of MS developer assistance dollars. Another case of the open source community shooting itself in the foot...
I am beginning to think that some of the trouble is not just windows, but also tied to poor hardware. All said, I am glad that my machine is more stable.
As a former program manager responsible for all software on the laptops of a large Texas computer manufacturer, I can tell you that Microsoft has its share of bugs, but by far the majority of the serious ones are indeed due to incredibly poor driver code - not Microsoft's responsibility at all. Even really big companies fall prey to this: for the past week, I've been trying to get Xerox to get me a driver for my multifunction printer that will actually load and run under XP - their own documentation makes it clear their driver only works sometimes, and they have no workaround, this for a driver released broken neraly two years ago. So much for service and support...
Driver certification probably is a racket for MS, but the testing process does at least ensure that someone who has a clue about writing drivers has checked to see that things are done correctly. Surprisingly, that is almost never the case if the drivers are "developed" by the hardware device vendor. (In reality, they are more likely just hacked together until they sorta work, at which point they cut their losses by deciding not to add any more destabilizing changes, so they just ship the crap and hope for the best.)
I've learned my lesson - for any hardware for a box running XP, I'm going to insist on signed and certified drivers from now on, to avoid having another several hundred dollar paperweight like the Xerox...
Anythign that Can Run Windows CE can Run Linux... I am pretty sure You would be pretty hard pressed to find a Platform that Runs Windows CE that allready doesn't have a good working port of linux.
How about an Epods One (a.k.a. Ezex Polaris)? Nice little CE box, color LCD touchscreen, decent battery life, even CF and PCMCIA slots. Available for less than a $100 barely used from lots of us who thought there might one day be Linux or BSD for it. If there was a candidate worth pursuing, you'd think this would be it.
Sadly, there's no version of either Linux or BSD that actualy runs on this device, although there each of the above has had an incomplete attempt made...
"and thus quite often any machine running a commercial Unix variant would stock GNU versions of utilities to take advantage of these features and to have a standardized version of them."
Only true in a tiny portion of the cases. At Chevron in the early 1990's we added the GNU utilities to those available natively on the system, but most people favored the SysV or BSD originals, as they were more functional and more stable. The only thing the GNU utilities had going for them was that they were at least consistent. But with GNU's rejection of man pages for documentation, their famously gratuitously incompatible command line options, and the hideous double-dash/minus-minus syntax that no one likes to type or deal with, it's not surprising people preferred the originals. At least the GNU utilities made a uniformly bad implementation available across all platforms...
Your history is a little mixed up, as it was AT&T's code (not Berkeley's) that was initially circulating in academia. This is how Berkeley got a hold of it, and started tweaking it and eventually got into a lawsuit over it. Here's a timeline to help you out.
Yes, AT&T's code was out there first, but it was clearly the BSD code that really took fire and spread. After all, it was BSD that added real networking support, and we all should now realize how much more valuable a networked Unix box is than one that is just standalone...
i>Sadly, I don't think I could convince you how bad it is to lock away innovation from the community as the private commercial Unix companies did. But, if you need proof as to the abusive result, take a trip to Redmond and ask them about the countless man hours lost to the Blue Screen of Death.
First, they did anything but lock it away. As I pointed out, Sun opened up Unix, and was single handedly reponsible for proving that it was a viable alternative to the IBM and VAX machines that ruled in those days. And NFS was an open spec from day one, and always available for licensing even to competitors on commercailly attractive terms - are you telling me that the implementaiotn of the first open network-accessible file system wasn't a big deal?
Umm, how old are you? Please refer to the links I give above for Unix history and how not all Unix variants descend from BSD.
I'm old enough to have lived and worked in the Unix field for almost all of the timeline you link to. I realize that not all Unices trace their ancestry directly to the BSDs, but ALL of them that survive today are *very* heavily influenced by them - the Internet itslef would not exist if it were not for BSD. You obviously were not there yourself, or you'd kjnow just how close the world came to the very closed and proprietary ISO/OSI networks rabidly advocated by the European computer companies - networks that would have had concentrated central controls on everything. It very nearly happened, and we have only the beachead that BSD established to thank for today's Internet. I know. I was there, fighting for the Internet way.
>>I really wonder what great opportunities we're missing now simply because som much good code is rendered useless for commerce by the GPL.
>I understand how a certain perspective can make the mistake, but I believe you meant "rendered useless" for theft.
No, I meant what I said. Until and unless there is some way to let innovators recover the very substantial cost of non-trivial improvements in software (especially those that don't "scratch an itch" felt by programmers), then GPL code cannot be reasonably used in any commercial app that really does innovate. Generally, this is not much of a loss, since every critical function is available under a license that allows investors to recover the costs of developing their inprovements, and GPL programs can be reimplemented to avoid the constraints of that intentionally poisonous license. Let's face it, the *entire* purpose of the GPL is to be as anti-commerce as poss
BSD is great. BSD developers are great. But BSD fanboys are arrogant snobs.
No, the BSD developers are (mostly) very serious software professionals that have a far deeper understanding of the implications of their design decisions than most Linux developers do. They're often interpreted as "arrogant" by pimply-faced kids that refuse to admit that experience is valuable, and those folks know things they won't understand for years yet, not matter what kind of hot-shot programmer they think they are today.
Making good architecture decisions is *hard*, and the engineering process the the BSDs use to determinet the correct answer to such questions has real advantages in the real world. Consider that the BSDs are still very viable alternatives, despite having several orders of magnitude fewer developers than Linux.
The article draws this distinction quite clearly, but since when did those involved in a/. flamefest ever bother to read the article?...:-)
I've seen people who download, compile, reboot, and repeat for EVERY single Linux kernel every released. They upgrade on Monday, but if a new minor version comes out and Wednesday they upgrade all over again.
These people are different from those of us that understand that the computer is a tool, and it's here to serve us, rather than the other way around. Whether you want to stay on the bleeding edge or in a snug secure harbor, BSD makes it easier than Linux to do so safely.
In fact, I've only ever encountered two OSes I think are truly upgradeable - QNX, and BSD. Both can be safely upgraded in place for years across many versions, and in QNX, you don't even have to reboot...
If Berkeley had licensed their version of Unix under something similar to the GPL, the Unix Wars never would've happened and Microsoft wouldn't have become the monstrous monopoly it is today.
The "Unix Wars" wouldn't have happened for the very simple reason that there could have been no commercial versions of Unix - the GPL is inherently and immutably anti-commerce. The BSD license (even the old, "bad" one) was the catalyst that ignited the entire Unix philosophy and mindset, initially in academia, but soon after throughour industry. The BSD license allowed Sun to build the world's first computer deisgned to run an open OS, and Sun's competitors were all forced to follow suit, making the Unix "idea" the single most dominant way of thinking about computing throughout the world.
Damn straight that couldn't have happened under the GPL: all the incentive would be gone. Linux itself owes its very existence to BSD, for without it, that young Finnish hacker would never have heard of Unix and thus to want to do somethinng so daft as to get a copy of it running on a PC. Instead, Linus would have been forced to attempt to copy Wang VS...:-)
In fact, Linux would just be another hobby OS and not the juggernaught it's growing up to be.
Damn straight the modern world couldn't have happened under the GPL: all the incentive would be gone. Linux itself owes its very existence to BSD, for without it, that young Finnish hacker would never have heard of Unix and thus to want to do somethinng so daft as to get a copy of it running on a PC. Instead, Linus would have been forced to attempt to copy Wang VS...:-)
I really wonder what great opportunities we're missing now simply because som much good code is rendered useless for commerce by the GPL. I know it happens, because I jsut made the call on an embedded OS for a proposed product at out new startup: it will be BSD-based rather than Linux based. Partly for BSD's greatly superior stability, but mostly so that we can freely modify the system and offer those improvements as a real benefit to our customers. That's simply not possible with the GPL...
I have an old 9" Heathkit amber monitor (NTSC video) I built for an Altair kit computer in the '70s. I think the monitor was a 1980 kit.... I also have two 1982 Magnavox separated video monitors...
OK, somebody give this guy the prize - I think he has us all beat...
You'll have to make a cable to connect your 1541 discdrive to your PC. See this page for more info.
Sadly, making the cable is quicker than loading a file on the 1541. Was this the most pathologically glacial disk drive ever built? I've got to believe it is. It's hard to imagine a slower disk, although it seemed like a huge step up from Kansas City format cassette tapes...
I have and use a 14" TVM monitor from 1992. Does 640x480@70, 800x600@56, and 1024x768@43.5 *interlaced*. Attached to a 486 DX/50 w/ 8mb of ram running Gentoo linux. I need all the compiler flags I can get ! This is not a joke.
I have two small 486sx33 "utility servers" running an old Mandrake (old enough to run on a 486!) and strippo BSD. They only support 32 MB of RAM. The "monitor" for one is an old Lear-Seigler ADM3a terminal (with the upgrade to support mixed-case letters by displaying characters with "descenders" raised up to fit on the line.) the other uses an old but reliable Compaq 151S, at one time a respectable monitor.
8 MB! I feel for you - one of the reasons I've left mine alone is that they work, and installing *anything* is a major PITA, since these are embeded 486's and cannot accept a CD drive at all. You get a 2.5" laptop hard disk (no more than 2 GB, please), a floppy, and an NE-2000 clone. If you can't install it with that, you're toast. They're almost as hard as my old Toshiba Libretto 50, which is really tough, since neither a floppy nor a CD is possible on it. Ugh. But it sure is cool;-) - It's been my portable file server for years...
Slashdot Mirror
But.. but... the pentium bug had nothing do to with normal floating point rounding error. Why does everyone keep making this joke?
I don't know, Andy. Maybe you're just paranoid...
Might as well toss in a link to a real astrolabe: here's the Electric Astrolabe, a very capable program that turns your computer into possibly the wolrd's most functional astrolabe.
This is one of the last DOS programs I run. It makes an interesting screensaver, too...
Well, there's:
England
England and Wales
Great Britain
The United Kingdom of Great Britain and Northern Ireland
My understanding (I'm a fifth generation Texan myself) is that there is still some considerable disagreement on whether "Great Britian" really and truly includes Scotland, a.k.a. Caledonia. And the current "devolution" actions, including the relatively recent acceptance of a Scottish Parliament, would tend to indicate that there's something to that argument...
X is all wrong for devices like this. Qtopia, on the other hand, which is what Sharp was smart enoug to put in them to start with, is a very good fit. This is simply a hacking tour-de-force. Sure, you can do it, and run X on an X-scale PDA, but it's *stupid* to do so. Especially since Qtopia lets you port Qt applications with minimal fuss. I suppose this gives the rabid QT haters somethign to do with their spare time, though, so it's not all bad...
That said, for machining where the precision is more important meteric makes a good system, although Imperial/English knew this and measurses precision measurements in thousandths or hundreths of inches to gain the advantage of decimal conversions and the increased precision, but metric is quite good for these types of fabrication.
Actually, inches are far superior for ordinary machining tasks, too. Most machine tools can hold a tolerance (determined by physics, not thier designers) of a little less than a thousandth of an inch. A hundredth of a millimenter (a centimiliimeter) is way too sloppy - more than two and a half times that size. A millimillimeter (a micrometer) is really too small though. The thousandth of an inch is extremely useful, and one of the reasons why we in the US are not going to give up useful units just because the rest of you decided to follow the French into a system guaranteed to create floating point math errors. Like pretty much all French ideas, SI can't stand any sort of scrutiny in the real world...
(See my other post here about the deca/deka-da/dk cockup for more information - the SI isn't even consistent with *itself*!)
The Chinese used 16 as the base for their measurement systems. (I know this is true of their system for weights used in commerce, is it true for other units as well? Anybody know?)
This can be easily demonstrated by traditional abacuses, which have 2 five-beads and 5 one-beads, as opposed to "modern" abacuses (a fairly recent invention) with one five-bead and 4 one-beads.
But one thousanth of an inch is a perfect measuring stick for cutting. Not one millimeter, not one micron, but a thousanths of an inch. Other measurements work out well with an inch too. Its just convience of scale, and the inch works well.
:-)
What's the matter with you? Are you trying to tell me that it's easier NOT to work in units of a some odd amount more than 2 and a half centi-millimeters?
Seriously, the thousandth of an inch, like the degree F, is one of those units that is just a lot more inherently matched to the what people use them for in the real world than the metric alternatives.
Your cars don't have km/h on the speedometer as well as mi/h? Just about every car I've seen in the US has both units on the speedometer. Older cars, obviously, may not have this feature.
Actually, I hadn't thought of it before, but a "MPH-only" speedometer is not a bad criteria for buying a car. And you avoid all that spyware/monitoring crap that's built in to all the new ones. (Not to mention, if you've got any sense, you'll make sure you get something with that unmistakable big V-8 rumble...)
I much prefer units I can relate to, personally. If the metric system has given us nothing else (which it hasn't), at least we have the Centigrade scale. I'm all for keeping the old-fashioned units alive, but really! Who thought water freezing at 32 and boiling at 212 was a sensible scale?
This is one place where English units *definetely* make much more sense than Metric/SI units, *especially* for everyday use:
Tha problem is that a cEntigrade/Celsius degree is just too damn big to really be useful, especially for temperatures that really matter to *people*.
For instance: I used to own an Alfa Romeo 164S that, like many "metric" cars, had a climate control system that "thought" in Celsius degrees. The problem is, they're just too big to allow fine enough control for comfort: setting an AC system at 72 degrees and letting it try to stay within a degree of that works fine with Fahrenheit degrees, but doing the same thing, with a one-degree tolerance in Celsius degrees results in a temperature swing that is almost *twice* (9/5, actually) as large, so the system cannot really keep things comfortable.
Any sort of reasonable thermostat in Celsius has to resort to using half-degrees.
Who the heck cares where water boils? Especially since pretty much everyone (even metric bigots) knows that it boils at 212 degrees F.
Fahrenheit degrees are just more useful in the real world. (And they also make having to deal with negative degrees fairly rare, unless you live someplace unsuited for human habitation anyway...)
All you need to know, it's the universal prefixes (vocabulary, not math skills): Mega-, kilo-, deca-, centi-, milli-, etc (and really for everyday life you only really ever use 3 of them: kilo-, centi-, milli-).
Your post hits the nail on the head - the Metric system (SI for the pedantic) is anything but consistent!
You cite "deca" - there's NO SUCH PREFIX. In fact, the entire deci/deka mess turned into such a morass of ISO incompetency that the official prefixes are now wildly inconsistent even within the SI system. (I discovered this recently when I looked it up and realized they'd "changed" the metric system since I learned it in the late 1970s):
The prefix for "deka" or "deca" (even the spelling is inconsistent!) is deplorable - it's listed as either da or dk, depending on where you look, and I'd argue *both* of these are wrong, since they break the otherwise inviolable rule of SI prefixes being a single character.
Thus the poor Metric-loving product of European public school education who sees 3.24 dkm may (quite reasonably) wonder what other metric-loving fool would specify a distance in "decikilometers"! (Oddly, some metric preferences would make this an acceptable if quite confusing usage, since they discourage the use of less-commonly used prefixes like "hecto" in favor of more common ones like "deci" and "kilo". Only if the user understands that magnitude prefixes should never be compounded can he avoid this mistake.)
When I learned the Metric system, "deka" was recommended to be represented with an upper-case "D" to distinguish it from the lower-case "d" representing "deci". Apparently, European outcomes-based educators thought this was too confusing to teach, and so lobbied to have it changed, although apparently they failed to expunge the far-too-commonly-used prefix overloading of using the letter "M" to represent both "mega" and "milli" based on case-sensitivity. (The formerly upper case "K" for "kilo, apparently also fell at this time, replaced by the now-standard lower-case "k", breaking the at-least-sort-of-sensible former property of SI prefix symmetry, which at that time held that upper case prefixes represented a positive exponent of the 10, while lower-case ones represented a negative exponent.
So instead, they decided that "deka" would stand alone in all of SI nomenclature, and use a two-character prefix, but apparently, in true ISO fashion, they couldn't agree on even what those two characters should be, "da" or "dk"? (Look around, and you'll find both, even in "official" publications from a number of countries that have adopted the SI as "standard".
(And then of course, there's the pathologically broken use of the lower-case Greek mu as the prefix for "micro". This results in a unit that cannot be typed directly on the vast majority of the world's keyboards, and which is at the mercy of extended character maps that often do not successfully carry the character across different applications, operating systems, or displays, thus rendering yet another prefix at least halfway useless.)
Now Metric-bigots will respond, "But no one uses deci much or deka at all, so who cares?" Those same metric-bigots should, because the bureaucratic bungling that has surrounded SI politics since the beginning is what has has rendered them unusable. If you can't actually use all those decimal prefixes, because doing so creates more confusion than it resolves, then how can you possibly claim that the Metric system is easier, more logical, or more consistent than the English system. You can't - the Metric system is every bit as arbitrary and capricious as the English system, but much harder to use in the real world, because 10 is such a difficult number.
Why do you think hours are 60 minutes, or circles 360 degrees? Because 60 is evenly divisible by 2,3,4,5,6,10,12,15,20, and 30! Try plotting your ship's course in radians sometime and see how long you can sail before running aground - talk about a stupid measurement!
Re:Imperial, not English...
Re:Imperial, not English... (Score:1)
by Bob Uhl (30977) on Monday January 26, @12:03PM (#8089873)
(http://latakia.dyndns.org/blosxom/bl
Of course I do. Let's say I wish to cut a foot-wide piece of wood into inch-width segments: I cut in half, then in half again, then in thirds. How would I cut a decimetre-wide piece into centimetres? In half, then into fifths? Ever try eyeballing a fifth? There are very easy ways to estimate a half or third quite accurately, but I am aware of none to estimate a fifth.
Let's say I have 1 gallon of beer and need to serve it in 1 pint measurements. Dead simple: I cut in half (yielding pottles), in half again (yielding quarts) and in half one last time, yielding pints. Doubling and halving are extremely easy with liquids and masses; anything else is a right royal pain in the ass.
Try dividing a decilitre into centilitres or millilitres. Good luck.
The standard system, used as it was for millennia, was optimised for use, for manipulation of concrete amounts. The French system was optimised for conversion between units. The one is needed daily; the other almost never (now that we have computers, practically never). Which would an unbiased observer prefer?
The standard system is, of course, imperfect and could use certain improvements. Its basis, though, is sound. The French system is also imperfect, but its sole basis is a silly attachment to 10 (a mathematically ugly number anyway). I'd rather spend effort on the system with a future.
Someone with mod points please mod up parent. He raises some *very* valid points that are far too often overlooked. This is one of the most important reasons why the engineering community her in the US will continue using Englinsh units for the foreseeable future. (I fully expect the US to remain primarily on English units through this century, and possibly far beyond...)
My company just designed a new mechanical product, and it was decided that English rather than Metric units were preferable for many reasons including ease-of-use, marketing considerations (even though half the unit volume will be international), and availability/selection and lower cost of hardware.
Could you provide same results from a plausible source - creationists and their well-known lies don't come even close to being anything more than a good laugh.
Actually, if you'd bother to read the links and their footnotes, you'd see the original sources are published science and comments of scientists. The site itself is run by uber-programmer David Pogge, a.k.a "Do-While Jones", who has achieved the high honor of being named a fellow at the US Navy's China Lake Weapons Center as one of the world's leading experts on the hardware and software design of guided missile systems.
Most of the quotes in the article are from the likes of Science and Nature, sources not noted for their "creationist" sympathies. There is no credibility gap here, unless its yours. They said what they said, in print, and what they said clearly does not support evolution.
In any case, Pogge's site, Science Against Evolution is not at all creationist, but simply points out that science actually does anything but support the theory of evolution. In this case, it is the scientist's own words and backtracking from their unjustifiable assertion about this "Mars rock" that make the links so interesting.
I can see it now - Advent(ure): the Dilbert Edition...
You are in a board room. Its long, polished oval table and leatherette chairs are quite intimidating. A filter coffee machine bubbles quietly in the corner.
A Board Member is here.
A Chairman is here.
An Executive Directory is here.
An Axe is on the floor.
Your PHB has entered the room and taken a seat at the table.
> Kill PHB
Kill a PHB?!! With what? Your bare hands?
> Yes
Congratulations, you have just vanquished a PHB with your bare hands!
A woman with tall triangular hair has entered the room. She nonchalantly kicks the PHB's corpse into the corner and takes his seat.
The chairman is shocked at your actions, but announces that he is granting you options on 1000 shares in this moribund subsidiary of a large but ultimately doomed computer company. If the entire corporate behemoth exceeds expectations by at least 30% for the next 12 quarters in a row (by somehow figuring out how to make money off Linux), those options won't even be underwater anymore.
> Refuse options
You are in a maze of tiny, light beige, but still somehow dark, cubicles, all alike...
There is a woman with tall triangular hair here, laughing at you in a hurtful way...
> It never ceases to amaze me that in the world of "science" that theory is most always talked of as proven fact. If the probability that object x doesn't exist is 80%, that object is talked about as if it doesn't exist conclusively, even though the possiblity that it exists is in the other 20%t. The age of the universe is a good example. Scientist y estimates to to being x Billion years old, and every textbook, news article, journal, documentary, etc states that the universe is x billion years old as if proven fact. The truth is that no-one currently knows exactly whether it's one billion or a billion billion years old.
That's why scientists prefer convergent evidence over one individual's opinion.
BTW, AFAICT astronomers are now nearly unanimous on an age of 13.7 billion years. This is a fairly recent result.
Your "convergent evidence", though, is not really any such thing, but just a general agreement on what is politically acceptable, not scientifically justified.
It's vitally important to understand that these dates are in reality based solely on conjecture, since every radioactive dating method rests on unfounded assumptions - assumptions that produce results that are known to be grossly and wildly inconsistent in the few cases we have where the ages of rocks *are* known precisely. The assumptions are "tweaked" to produce the desired age, not the other way around...
From the article: The team that found it was led by experienced meteorite hunters Carine Bidaut and Bruno Fectay, who have now found six rocks from Mars - a record.
Interesting that they seem to know *just* where to find Martian rocks.
It's also really interesting that the last big hoo-rah about finding "a rock from Mars" here on Earth coincided with Bush Sr's proposal for a mission to Mars. What's really amazing is that these discoveries are so strongly correlated to Congressional consideration of tens or hundreds of billions of dollars for Mars research! (Perhaps we should fund a $40 million NASA/CBO research study to determine if this correlation can be linked to any sort of causality? Hmmm?)
There is absolutely no real evidence that these rocks are from Mars, and in fact, NASA very quickly backed away from defending the Mars origin theory when pressed to provide scientific backup. There is absolutely no substantial reason to believe that ALH84001 came from Mars, other than that it sure seems to help get funding.
Of course, if we go to Mars, it should be without NASA - how we could think that a government agency with a worse operational safety and efficiency record than the Post Office should run a Mars mission is beyond me. I've worked at NASA, and belive me, the only real soultion is to bulldoze the entire organization and start over - even thenm you'd need to wait ten years before starting over to avoid the revolving door effect...
Someone with mode points mod pls mod parent up. He makes some very good points, although he'll probably get modded to the basement because he (correctly) points out that CE is indeed winning the OS battle in embedded devices.
This is very likely caused at least in part by the fact that the embedded Linux market is extremely fragmented, with no real agreement or standards, especially w.r.t the visual environment: There are literally dozens of incompatible embedded GUI choices, none of which have enough momentum to succeed. (And all of which rely on wildly inconsistent programming toolsets and methods.)
As a result, CE, which is almost as fragmented across platforms, but at least has centralized and coordinated developer support, wins without a battle.
In reality, Linux is ill-suited today for use in embedded devices that requires a GUI, since getting comaprable functionality requires GUI libraries and tools that are just as proprietary as MS, but without the benefit of mind and market share or a big pot of MS developer assistance dollars. Another case of the open source community shooting itself in the foot...
I am beginning to think that some of the trouble is not just windows, but also tied to poor hardware. All said, I am glad that my machine is more stable.
As a former program manager responsible for all software on the laptops of a large Texas computer manufacturer, I can tell you that Microsoft has its share of bugs, but by far the majority of the serious ones are indeed due to incredibly poor driver code - not Microsoft's responsibility at all. Even really big companies fall prey to this: for the past week, I've been trying to get Xerox to get me a driver for my multifunction printer that will actually load and run under XP - their own documentation makes it clear their driver only works sometimes, and they have no workaround, this for a driver released broken neraly two years ago. So much for service and support...
Driver certification probably is a racket for MS, but the testing process does at least ensure that someone who has a clue about writing drivers has checked to see that things are done correctly. Surprisingly, that is almost never the case if the drivers are "developed" by the hardware device vendor. (In reality, they are more likely just hacked together until they sorta work, at which point they cut their losses by deciding not to add any more destabilizing changes, so they just ship the crap and hope for the best.)
I've learned my lesson - for any hardware for a box running XP, I'm going to insist on signed and certified drivers from now on, to avoid having another several hundred dollar paperweight like the Xerox...
Anythign that Can Run Windows CE can Run Linux... I am pretty sure You would be pretty hard pressed to find a Platform that Runs Windows CE that allready doesn't have a good working port of linux.
How about an Epods One (a.k.a. Ezex Polaris)? Nice little CE box, color LCD touchscreen, decent battery life, even CF and PCMCIA slots. Available for less than a $100 barely used from lots of us who thought there might one day be Linux or BSD for it. If there was a candidate worth pursuing, you'd think this would be it.
Sadly, there's no version of either Linux or BSD that actualy runs on this device, although there each of the above has had an incomplete attempt made...
"and thus quite often any machine running a commercial Unix variant would stock GNU versions of utilities to take advantage of these features and to have a standardized version of them."
Only true in a tiny portion of the cases. At Chevron in the early 1990's we added the GNU utilities to those available natively on the system, but most people favored the SysV or BSD originals, as they were more functional and more stable. The only thing the GNU utilities had going for them was that they were at least consistent. But with GNU's rejection of man pages for documentation, their famously gratuitously incompatible command line options, and the hideous double-dash/minus-minus syntax that no one likes to type or deal with, it's not surprising people preferred the originals. At least the GNU utilities made a uniformly bad implementation available across all platforms...
Your history is a little mixed up, as it was AT&T's code (not Berkeley's) that was initially circulating in academia. This is how Berkeley got a hold of it, and started tweaking it and eventually got into a lawsuit over it. Here's a timeline to help you out.
Yes, AT&T's code was out there first, but it was clearly the BSD code that really took fire and spread. After all, it was BSD that added real networking support, and we all should now realize how much more valuable a networked Unix box is than one that is just standalone...
i>Sadly, I don't think I could convince you how bad it is to lock away innovation from the community as the private commercial Unix companies did. But, if you need proof as to the abusive result, take a trip to Redmond and ask them about the countless man hours lost to the Blue Screen of Death.
First, they did anything but lock it away. As I pointed out, Sun opened up Unix, and was single handedly reponsible for proving that it was a viable alternative to the IBM and VAX machines that ruled in those days. And NFS was an open spec from day one, and always available for licensing even to competitors on commercailly attractive terms - are you telling me that the implementaiotn of the first open network-accessible file system wasn't a big deal?
Umm, how old are you? Please refer to the links I give above for Unix history and how not all Unix variants descend from BSD.
I'm old enough to have lived and worked in the Unix field for almost all of the timeline you link to. I realize that not all Unices trace their ancestry directly to the BSDs, but ALL of them that survive today are *very* heavily influenced by them - the Internet itslef would not exist if it were not for BSD. You obviously were not there yourself, or you'd kjnow just how close the world came to the very closed and proprietary ISO/OSI networks rabidly advocated by the European computer companies - networks that would have had concentrated central controls on everything. It very nearly happened, and we have only the beachead that BSD established to thank for today's Internet. I know. I was there, fighting for the Internet way.
>>I really wonder what great opportunities we're missing now simply because som much good code is rendered useless for commerce by the GPL.
>I understand how a certain perspective can make the mistake, but I believe you meant "rendered useless" for theft.
No, I meant what I said. Until and unless there is some way to let innovators recover the very substantial cost of non-trivial improvements in software (especially those that don't "scratch an itch" felt by programmers), then GPL code cannot be reasonably used in any commercial app that really does innovate. Generally, this is not much of a loss, since every critical function is available under a license that allows investors to recover the costs of developing their inprovements, and GPL programs can be reimplemented to avoid the constraints of that intentionally poisonous license. Let's face it, the *entire* purpose of the GPL is to be as anti-commerce as poss
BSD is great. BSD developers are great. But BSD fanboys are arrogant snobs.
/. flamefest ever bother to read the article?... :-)
No, the BSD developers are (mostly) very serious software professionals that have a far deeper understanding of the implications of their design decisions than most Linux developers do. They're often interpreted as "arrogant" by pimply-faced kids that refuse to admit that experience is valuable, and those folks know things they won't understand for years yet, not matter what kind of hot-shot programmer they think they are today.
Making good architecture decisions is *hard*, and the engineering process the the BSDs use to determinet the correct answer to such questions has real advantages in the real world. Consider that the BSDs are still very viable alternatives, despite having several orders of magnitude fewer developers than Linux.
The article draws this distinction quite clearly, but since when did those involved in a
I've seen people who download, compile, reboot, and repeat for EVERY single Linux kernel every released. They upgrade on Monday, but if a new minor version comes out and Wednesday they upgrade all over again.
These people are different from those of us that understand that the computer is a tool, and it's here to serve us, rather than the other way around. Whether you want to stay on the bleeding edge or in a snug secure harbor, BSD makes it easier than Linux to do so safely.
In fact, I've only ever encountered two OSes I think are truly upgradeable - QNX, and BSD. Both can be safely upgraded in place for years across many versions, and in QNX, you don't even have to reboot...
If Berkeley had licensed their version of Unix under something similar to the GPL, the Unix Wars never would've happened and Microsoft wouldn't have become the monstrous monopoly it is today.
:-)
:-)
The "Unix Wars" wouldn't have happened for the very simple reason that there could have been no commercial versions of Unix - the GPL is inherently and immutably anti-commerce. The BSD license (even the old, "bad" one) was the catalyst that ignited the entire Unix philosophy and mindset, initially in academia, but soon after throughour industry. The BSD license allowed Sun to build the world's first computer deisgned to run an open OS, and Sun's competitors were all forced to follow suit, making the Unix "idea" the single most dominant way of thinking about computing throughout the world.
Damn straight that couldn't have happened under the GPL: all the incentive would be gone. Linux itself owes its very existence to BSD, for without it, that young Finnish hacker would never have heard of Unix and thus to want to do somethinng so daft as to get a copy of it running on a PC. Instead, Linus would have been forced to attempt to copy Wang VS...
In fact, Linux would just be another hobby OS and not the juggernaught it's growing up to be.
Damn straight the modern world couldn't have happened under the GPL: all the incentive would be gone. Linux itself owes its very existence to BSD, for without it, that young Finnish hacker would never have heard of Unix and thus to want to do somethinng so daft as to get a copy of it running on a PC. Instead, Linus would have been forced to attempt to copy Wang VS...
I really wonder what great opportunities we're missing now simply because som much good code is rendered useless for commerce by the GPL. I know it happens, because I jsut made the call on an embedded OS for a proposed product at out new startup: it will be BSD-based rather than Linux based. Partly for BSD's greatly superior stability, but mostly so that we can freely modify the system and offer those improvements as a real benefit to our customers. That's simply not possible with the GPL...
I have an old 9" Heathkit amber monitor (NTSC video) I built for an Altair kit computer in the '70s. I think the monitor was a 1980 kit.... I also have two 1982 Magnavox separated video monitors...
OK, somebody give this guy the prize - I think he has us all beat...
You'll have to make a cable to connect your 1541 discdrive to your PC. See this page for more info.
Sadly, making the cable is quicker than loading a file on the 1541. Was this the most pathologically glacial disk drive ever built? I've got to believe it is. It's hard to imagine a slower disk, although it seemed like a huge step up from Kansas City format cassette tapes...
I have and use a 14" TVM monitor from 1992. Does 640x480@70, 800x600@56, and 1024x768@43.5 *interlaced*. Attached to a 486 DX/50 w/ 8mb of ram running Gentoo linux. I need all the compiler flags I can get ! This is not a joke.
;-) - It's been my portable file server for years...
I have two small 486sx33 "utility servers" running an old Mandrake (old enough to run on a 486!) and strippo BSD. They only support 32 MB of RAM. The "monitor" for one is an old Lear-Seigler ADM3a terminal (with the upgrade to support mixed-case letters by displaying characters with "descenders" raised up to fit on the line.) the other uses an old but reliable Compaq 151S, at one time a respectable monitor.
8 MB! I feel for you - one of the reasons I've left mine alone is that they work, and installing *anything* is a major PITA, since these are embeded 486's and cannot accept a CD drive at all. You get a 2.5" laptop hard disk (no more than 2 GB, please), a floppy, and an NE-2000 clone. If you can't install it with that, you're toast. They're almost as hard as my old Toshiba Libretto 50, which is really tough, since neither a floppy nor a CD is possible on it. Ugh. But it sure is cool