Sorry, but I have to disagree with your conclusion that this represents exponential growth.
The effect you speak of (doubling the number of processors giving less than double the final "power") is due to additional overhead - various processors coordinating their work with each other, deciding things like "Should I split this 2 ways or 4?" and so on - and that sort of stuff inevitably increases with the number of processors.
You can use improved algorithms, special-purpose hardware, etc, etc, to minimize this "friction", but it will always exist, and the percentage of processing that is "overhead" will inevitably climb as you increase the number of processors.
It's far more likely that either the earlier number resulted from some inefficiencies that existed then (due to it not being built as designed yet, perhaps), or there have been improvements in the algorithms or infrastructure which give greater efficiencies.
If it's the latter case, if you unplugged the 2nd half of the CPUs and made the measurement again, you'd probably get 150 GFlops or so.
Basically, you could write the equation for total power something like:
X - O - i**x, where X is the number of processors, O is the basic overhead (for doing things like I/O, for example), and c is the incremental cost of adding each processor.
To have what you describe would require that i**x be a negative number, which is like saying that you can have 10 individual conversations in less time than you can have five. Ain't gonna happen.
People seem to be having a lot of fun on this thread, but just in case anyone is confused:
The article speaks of "super-oxygenated water", which means water with a great deal of oxygen dissolved in it. H2O2 (hydrogen peroxide) is a totally different chemical with totally different properties.
I suspect the parent was joking. (AC, sorry if I spoiled your joke...)
I used to use it on IBM mainframes, and loved it. Of course, the language I was using before was "clist", which I've typically described as a retarded assembly language. (That's actually a bad description, because it's nothing like assembly language. My point was that I would far rather have written in assembly language (for any processor) than write clists.)
Rexx is a scripting language. I used it for things that in the Un*x world you would use a shell (ksh, bash, etc) for, or in some cases, what I would use awk for on Un*x.
This is the object-oriented version, which I've never played with. I strongly suspect it is comparable to things like Python and Ruby.
The syntax is reasonably clean (one reason I compare it to Python and Ruby). If you hate Python and Ruby and love Perl, you won't like Rexx. If you're on the opposite end of the spectrum (I am), you may like it. (Although if you already know Python or Ruby, I don't know that there would be any great advantage to doing Rexx.)
Rexx had one feature which I absolutely loved, and I've never seen another language with anything nearly as good. It had a trace function which showed you what the script was doing, first as the actual code, then the results of that code. (A similar function would be "bash -vx script".)
What made Rexx's debug facility so much better than everything else was that if turned all the way on (you could turn on and off various things, like whether you wanted to see variable values), it would show you not only the original code and the result, but all the intermediate results, as well.
For example, say you had a line that said something like:
x = ((( a + b ) * ( c / d )) * e) ^ f
To make the example simple, a=1, b=2, etc.
Rexx would show you:
x = ((( a + b ) * ( c / d )) * e) ^ f
x = ((( 1 + 2 ) * ( 3 / 4 )) * 5) ^ 6
x = (( 3 *.75 ) * 5) ^ 6
x = (2.25 * 5) ^ 6
x = 11.25 ^ 6
x = 2027286.5295... (I'm not bothering with the whole thing, ok?)
I had one program that I had written in clist language. I found a bug, and past experience led me to think it would take about two days to find and fix. Since I wanted to learn Rexx, I ported the program over to Rexx (which wasn't trivial, but it's also not important to my point). I still had the bug. Using the trace facility in Rexx, it took me 15 minutes to find and fix the bug. I was in love with the language from that point on.
A sphere you walk on top of would probably be easier to construct, but unfortunately, either way has the same problem, because you're wrong about one thing.
Assume a 30 inch step.
That makes the short side of the triangle 15 inches.
Start off with a sphere 10 feet in radius (20 feet in diameter).
15/120 =.0125, which is the sine of the triangle.
Cosine(arcsine(short/hypotenuse))=0.94 inches.
A 1 inch height difference would certainly be noticed by me.
Assuming a 0.1 inch difference as small enough to be ignored, your sphere would have to be about 94 feet in radius. (And remember, that's radius. It's almost 200 feet in diameter.
Considering that's what would be required for each person in the game, I think what they've got is definite improvement.
I'm not impressed by the photo, though. It doesn't look like you could (safely) take a step forward, unless those blocks are really fast.
To anyone who complains that I should have done that in metric:
A) I'm a Merkin. (See alt.fan.pratchett on Usenet) We're allowed.
B) I'm at work and trying to be reasonably honest with my employer's time...
I found a copy of the Fortran code for Adventure where I worked in the early 80s. I was learning PL/1 at the time, and needed a project of some sort to learn it with, so I ported Adventure over to it.
Time went by (about 8 years) and I changed jobs a couple of times. One day, I was talking with a daughter site in Boston, and somehow the subject of Adventure came up. He was trying to do something with it, and happened to mention the name of the userid that created it.
It was DTSO403, which was my old userid at the original company.
I have no CLUE how it made its way from Dallas to Boston...
Being watched and tracked (and having "privacy" essentially disappear) is pretty near inevitable, for the same reasons that patents (both hard and soft) are increasingly a bad idea, and open-source software is inevitable.
Technology has marched on, and the world has changed (again).
All the trends in technology over the last 10 years say that privacy as we have known it, is headed for extinction. Cameras that get smaller and smaller, remote controlled robots, hacking into wireless LANs, PLUS all the electronic interactions (like RFID) that are coming, PLUS computers getting cheaper by the day... This all adds up to privacy basically being impossible.
Proprietary software is doomed, because the Internet made the level of interactivity that open-source software needs possible. For exactly the same reasons that the medieval guilds (with their proprietary methods for things like ironsmithing and glassblowing) were doomed once the movable-type printing press was invented, proprietary software cannot compete. In the near term (5-10 years), it will still have a solid space in niche markets, but I'm not even sure that will last. It certainly isn't going to last in mainstream software arenas like OSes and databases.
But that same increase in processing power and decrease in communication delay means that doing things like examining every electronic transaction that someone performs (and building a detailed profile of their life from it), is not only beginning to be possible, it's very nearly inevitable. Even the most paranoid of you out there (and on Slashdot, the percentage of paranoids is a good bit higher than average) would not want the sort of draconian methods that would be required to prevent it. (No computers and no networks, for instance.)
The proper solution, I think, is to change our culture, so that it doesn't matter that someone knows the kinks in my soul.
I am mostly connected to reality, so I'm not holding my breath on this cultural shift, but I really only see three possibilities:
We turn Luddite and roll back the clock technologically. (Not likely to happen voluntarily by most of this audience, but some of the non-technical types turning Luddite IS all too possible.)
Privacy gets moved to the same status as apprenticeship - it's something that existed historically, and it's occasionally useful for analogies, but it's not part of anybody's life anymore. This could either go the Japanese route (I believe the usual phrase is something like "Nakedness is frequently seen and never noticed." In other words, commenting on someone's quirks is far more shameful than having said quirks to begin with.), or simply an open acceptance that other people do things differently than you.
The third possibility is the one that worries me. That's a totalitarian society (probably theocratic) that uses this information to control people to a degree that has heretofore been unbelievable. I don't think such a state would last very long at all, but the creation and destruction of it would get really, really ugly.
The US is the only culture I have extensive first-hand experience with. I would strongly prefer to see us go to option 2B (taking the attitude that you can live your life any way you want as long as you don't hurt me).
That fits wonderfully with our stated national beliefs. It's an absolutely lousy fit with what our behavior says we believe. The behavior (IMO) says we urgently want #3.
That's the big reason the 3rd option worries me. I can very easily see a theocratic state as an intermediate step to the live-and-let-live one. If anyone has any practical, pragmatic suggestions for how to create such a cultural shift (one suitable for a total absence of privacy), speak up now, because the situation could get critical within 10 years, and is almost guaranteed to get critcal in 20.
I suspect this is directly related all of the noise that's been made lately about Linux-on-the-desktop.
Combine a lot of people NOT changing off of 98 with a virus/worm that cripples a company and cannot be fixed and anyone trying to sell Linux as a desktop OS would have a field day. It would be a horrendous PR mess for Microsoft.
I'm pretty sure MS did NOT want to do this. The fact that very large numbers of people (both businesses and individual consumers) refused to make the switch forced their hand.
At my company, we were using the lack of support for 98 as an argument for why we should upgrade a significant number of machines to XP. Our CIO was very resistant (he didn't like the licensing terms). What it boils down to is that Microsoft blinked first.
I had a bit of an epiphany last night, thinking about McBride's comments about the GPL.
I started out thinking about all the ways that I could demolish his argument (not terribly hard), when the thought suddenly occurred to me that, in a sense, that was exactly what SCO wanted.
Let's use a little common sense on the whole SCO thing.
First off, toss out a couple of emotionally popular but rather unlikely theories:
1) Darl McBride is insane, in the medical, legally-incompetent sense of the word. If he was, the rest of SCO and the Canopy Group would have shut him down a long time ago. Similar reasoning applies for "unbelievably stupid", "totally ignorant" and similar epithets.
2) #1 is true, but for everybody in SCO and Canopy, not just Darl, which is why he's still there. Ok, c'mon. If you're that out of touch with reality, talk to your doctor about upping the dosage on the little green pills.
A common allegation is that Darl McBride is a greedy money-grubber with the morals of an advertising exec. This is probably true, but in America, at least, being greedy on behalf of your company is not only not frowned on, it is (somewhat) required by law. (There is a SEC regulation that requires all publicly-owned corporations to try and maximize the value of their stock. Since it says nothing about HOW that's to be done, the range of tactics is pretty wide. You won't find any American corporations saying "We gave away all our profits, 'cause we felt like being nice guys", though.)
SCO's basic strategy is obvious: Fire the shotgun everywhere possible, as often as possible, and see what sticks. I saw an article the other day that said the software business in 2004 was predicted to be about $230 billion. If SCO can get 1/10 of 1% of that, they'd be ecstatic.
Another characteristic that McBride has, he shares with lawyers, politicians, and most high-powered types in business: a thick skin. I very much doubt he has been bothered in the least by the various vilifications called down on him by his detractors.
I think he was a bit surprised at it when it first started, but since then, it's been more of a weapon in his arsenal than anything else. If he is not bothered by name-calling and accusations, but the other side (the open source community, in general terms) is, then the more furious the argument, the better his odds of being able to find a weak point and exploit it somehow. At the very least, he (and SCO) can point at all the ranters and ravers and claim "With enemies that act like THAT, doesn't it make sense that we're the ones in the right?" (Something along these lines may be what got the money out of Baystar.)
If you're caught up in a strong emotion, you're not entirely sane. If you're angry, all kinds of little things you would ordinarily blow off make you even angrier. If you're ecstatically happy, you can find a silver lining in a mushroom cloud.
The reason I brought up politicians, lawyers, and CEOs earlier is that they all have one thing in common. I called it a "thick skin" earlier. Another way to describe it is that they have the ability to climb out of their emotions and think rationally again about whatever the subject is. That ability is what gets them paid the big bucks.
So what I think SCO is doing with a lot of the more unbelievable claims they've made (like the attack on the GPL last night) is not to seriously convince anyone of that position, it's to stir up trouble. The more emotional the opposition gets, the better the odds that something, anything, will happen that he (they) can exploit. It goes with the shotgun approach: the more you get things stirred up, the more targets of opportunity there are.
Fortunately for IBM, they have good lawyers, who haven't been influenced in any way they shouldn't be by the public furor. They simply stuck to the facts and the law. My favorite element of what they've been saying is that it's mostly in plain English. When one side speaks English and the other sid
Keep in mind that we do not sell licenses. We sell subscriptions where the value of the bits are integrated with service levels.
Then somebody at Red Hat is lying. My account rep (who I'll be nice and not name for now)told me yesterday, when discussing this exact subject, that we WERE buying (well, renting) a license, and that we most emphatically could NOT install it on multiple machines. (Which is 180 degrees from what a different account exec told me 6 months ago.)
This is the major gripe I have with the new scheme. If I want to put Linux on an unimportant box (say, an X station for the operators), or a temporary box, I must pay Red Hat $350 (absolute minimum) AND I must wait a week to ten days for them to process my order and send me the boxed set. (That statement is per the aforementioned account rep.)
Here's the way I'd like it to work:
You could install it as many times as desired, but get no support at all (not even RHN) for free
For a reasonable price (say $100), you get RHN only -- no support at all.
THEN get to the $350 if you actually want any kind of support
What we've got (per the account exec I mentioned) is item #3 as a option, and ONLY number 3.
Are you kidding? painehope wanted RedHat to support a 4000 CPU cluster for the price of a few lousy RedHat Linux boxed sets. That's just insane.
I think you misread what painehope said. He did not say he wanted it for $50 for 512 servers -- he wanted it for $50 each for 512 servers, $25,600 total. (And $35,840 for the 1024-node example.)
Since almost every server in the cluster is identical (by design), this is going to be about as hard for Red Hat to support as one box. (They will strongly tend to all have exactly the same problems.) In a few cases, you're going to have issues between boxes, and a few boxes will be different from all the rest (admin server, control point, managing node, that kind of stuff.)
I would assume they have a small test cluster (8-10 boxes) for working out bugs, especially the between-systems types. When the problem is solved, you install exactly the same thing on all 512 (or 1024, or 18 gazillion) machines.
He was even willing (even eager) to provide a caching RHN server (or perhaps he meant buy one, since RH does sell such a beast.)
So from Red Hat's perspective, this is going to be like supporting 10 or 15 different boxes (and I'm being pessimistic -- the equivalent of 4 or 5 is probably closer) and the RHN bandwith of ONE, for $25,600. That's actually a pretty sweet deal for Red Hat.
Painehope's problem is that Red Hat isn't interested in this. They say he should pay them $179,200 for 512 nodes. (Or $358,400 for 1024.)
The upshot of this is almost certainly going to be that people using the really big clusters (all the ones that make the "New supercomuter record!" headlines) are NOT going to be Red Hat.
And that $140 might be each, which would double it.
The article made the comment that people didn't think of copying software as stealing.
That's for the very good reason that it's not. If I steal something from you, I deprive you of the further use of that item. If I copy a CD, you haven't lost anything.
The argument Microsoft (and other proprietary digital-media makers, including the RIAA people) make is that I'm depriving them of the money they would have gotten if I'd purchased it.
That holds water if, and only if, I would in fact have spent the money if copying were unavailable to me.
At the equivalent of $10,000 a copy, I will guarantee that very, very few people would buy a copy if that were the only way to get Windows. They would simply do without. If Windows were the only operating system in the world and uncopyable (get that glazed look off your face, Bill. I said "if",) then the result in Vietnam would not be 2,000,000 people * $140 = $280,000,000 more in Microsoft's pocket, it would be 200 people * $140 = $28,000, which is probably about what they really got. Therefore, since MS has no less money than they would have had anyway, nobody is stealing, not, at least, by the argument of "what we would have had".
In the case of someone like me (an American, making well over the average US income), and a $15 CD, the argument is not as clear (and obviously self-serving), but in a case like the average Vietnamese (or Thai, or anywhere else the per capita income is drastically less than developed countries) I would say the argument has some weight.
In real life, if everbody in the world took "piracy" seriously and paid Microsoft full price, they would get 10%-15% more income (almost all of it from developed countries), which is probably exactly what they expect to get out of it. All the fuss and fury about hundreds of millions of dollars is just another marketing tactic.
There were two quotes in that article I really liked, both by the Microsoft rep:
With regards to the Open Source community:
"They give away innovation."
I like it. Not strictly true, but I rather like the idea.
"The battle on spam must be fought on all available fronts..."
I disagree. There is only one front that spam can be fought on successfully, and that is economic.
When spam is no longer profitable, spammers will give it up voluntarily. As long as it is profitable (especially as profitable as it is right now) people will continue to do it, regardless of one law or a thousand.
Each law passed to try to stop it restricts someone's freedoms, and it's pretty much unavoidable that the law will restrict some who do not deserve such restriction. If that would eliminate spam, that might be considered a reasonable tradeoff. Since it will not, each and every law passed solely on the basis of "We've got to DO something!" is a bad idea.
The best thing I've seen so far is A Plan for Spam. I tend to agree with the author's assesment of how likely it is to work, specifically because it destroys the economics of spam.
I think most of our collective energy should be going to integrating things like "A plan for Spam" into common email programs, possibly extending it to allow people to join anti-spam clubs, so they don't have to label the thousand or so emails the thing needs to train with themselves.
Simply having a button that says "Spam!" will help a lot of people deal with the frustration involved with spam, and if it is pitched to the general public as "Every time you kill an email as spam (instead of just deleting it) you're helping to put spammers out of business," you're going to have to beat them off with a stick. People would love to have a way to get back at spammers.
Nothing will stop them; in fact, that one can complain anonymously will tend to encourage that.
For precisely that reason, they won't investigate every complaint, or even a large fraction of them.
Like all complaint-takers for decades (if not centuries) they will investigate things based on:
Frequency of complaint
Precision of complaint
If 500 different people complain about the same thing, it's going to be taken more seriously than if one person does.
A complaint like "Microsoft screwed me blind! I want you to GET those evil bastards!" will probably be handled by a spam filter.
(Warning to those who are overly literal-minded: the following example is completely fictional!)
A complaint like "On Friday, Jun 14th, Jahfhs Ffjl of the Microsoft Licensing group told me that I would have to pay full retail price for Windows XP Home, rather than the usual OEM price, because I sell a line of computers that run Linux," will be much more effective.
Specific dates, names, statements, amounts, etc. will get far more attention than vague hate mail.
Finally, whatever they think should be done will get choked back to what can be done, given their budget.
End result: The most flagrant violations will get curbed quickly (or may never happen because they would have been caught so quickly).
Borderline cases will break down into two groups:
The ones Microsoft doesn't care about will get "cured", and used as public-relations fodder by MS.
The ones Microsoft decidedly does not want to get stuck with will be challenged in court, or similar stalling tactics.
In a few odd cases here and there, Microsoft will get their hand slapped and have to legitimately clean up their act some.
Bottom line: As usual, it's between the two extremes mentioned. It will do some good, but there's a lot of stuff that will still fall through the cracks.
My only complaint about it is that most of the good will come in the next year or two, and it will probably eat up money for then next couple of decades...
I'm not a lawyer either, so take this explanation with a grain of salt...
Peter Veeck wrote a bunch of codes and standards for buildings and utilities. As best I can tell, he had them up on a Web site, available at no charge to anyone who wanted them. (Being available free like that does not affect his copyright - he would still own it.)
He then proposed these standards to some of the nearby towns as proposals for laws.
At least two of the cities took him up on it, and passed the codes into law in their municipalities.
SBCCI made a reference to the codes as being "public domain". Veeck objected to this.
As best I can determine, he was not trying to prevent anyone from accessing the information, he was just saying "This is my text, I wrote it, I have the copyright on it".
The court said, in effect, "Not after those cities turned it into law!".
There seem to be two key elements: First, Veeck offered his writeup of the codes to the towns to be made into law (i.e., they didn't steal it from him). Second, they incorporated it by reference, i.e. saying something to the effect of "The codes that Peter Veeck wrote up are now law." (I didn't dig deep enough to see whether that was his idea or theirs.)
All laws are public domain (having to pay a fee to see a law that you are bound by might start another revolution...). What the court said here is that by that specific code (the one he wrote and (I think) had on his web site) becoming law, the "public domainness" of law overrode his (ordinarily perfectly good) copyright, so SBCCI is right, and what he wrote is now in the public domain.
I saw at least one post that lamented the lack of an active pen, and several that asked "What are you going to about command line?".
The typing issue is trivial. I'm nearly certain you can plug in a standard keyboard (it's basically still a laptop, after all), and for those occasions when you can't, a simple app could put a keyboard on the screen and you can type on that.
Also, there's a neat gadget (
http://www.time.com/time/2002/inventions/rob_keybo ard.html )
that could be incorporated pretty easily, although I'm sure it's not in the machine under discussion. (Sorry if the URL is broken in half -- I couldn't figure out how to fix it.)
That's just one reason I consider a passive touchscreen (anything which can depress it can activate it) to an active one (has a special pen that is the only thing that will activate it.)
Here's a few others:
* Losing your stylus doesn't require a $15-$20 (at least) replacement, and render your machine completely dysfunctional until you get it.
* It allows software to become much more intuitive. The original concept of a mouse was to simulate being able to touch the objects on the "desktop" and move them around. The objects will still be a simulation, but the touching will be reality.
This has one medium serious drawback, of course. It means Apple was right to stick with a one-button mouse all this time...
* It opens up a whole bunch of new interface interaction possibilities. For example, to address the one-button,two-button,three-button issue, you could tap with one, two, or three fingers. (Some of the stuff that's been done with right-clicking I actually like. I'd rather not give it up.)
Even more, it makes possible interactions that could not possibly be done with a mouse, such as moving a window around with one hand and resizing it with the other at the same time (or resizing with two fingers on the same hand); or moving two windows (maybe even three or four if you've got big hands and coordination) at once.
How'd you like to design a web page layout by literal "hands-on" design? Eye-hand coordination is built into the species. Remote control devices like mice, trackballs, graphics tablets, etc all make it more difficult to use your mind for the creative, aesthetic side of the work. (Geeks, who usually don't have any aesthetic sense worth speaking of, and who use mice and keyboards more often than any other object, will not get as big a boost in productivity. Sorry.)
Add to that any other activities that involve multiple fingers (musical keyboards, 10-key pads, phone number pads and so on), and I consider a passive touchscreen a much better option than One Pen to Rul... Sorry, started to get carried away.
Slashdot Mirror
Sorry, but I have to disagree with your conclusion that this represents exponential growth.
The effect you speak of (doubling the number of processors giving less than double the final "power") is due to additional overhead - various processors coordinating their work with each other, deciding things like "Should I split this 2 ways or 4?" and so on - and that sort of stuff inevitably increases with the number of processors.
You can use improved algorithms, special-purpose hardware, etc, etc, to minimize this "friction", but it will always exist, and the percentage of processing that is "overhead" will inevitably climb as you increase the number of processors.
It's far more likely that either the earlier number resulted from some inefficiencies that existed then (due to it not being built as designed yet, perhaps), or there have been improvements in the algorithms or infrastructure which give greater efficiencies.
If it's the latter case, if you unplugged the 2nd half of the CPUs and made the measurement again, you'd probably get 150 GFlops or so.
Basically, you could write the equation for total power something like:
X - O - i**x, where X is the number of processors, O is the basic overhead (for doing things like I/O, for example), and c is the incremental cost of adding each processor.
To have what you describe would require that i**x be a negative number, which is like saying that you can have 10 individual conversations in less time than you can have five. Ain't gonna happen.
People seem to be having a lot of fun on this thread, but just in case anyone is confused:
The article speaks of "super-oxygenated water", which means water with a great deal of oxygen dissolved in it. H2O2 (hydrogen peroxide) is a totally different chemical with totally different properties.
I suspect the parent was joking. (AC, sorry if I spoiled your joke...)
Rexx is a scripting language. I used it for things that in the Un*x world you would use a shell (ksh, bash, etc) for, or in some cases, what I would use awk for on Un*x.
This is the object-oriented version, which I've never played with. I strongly suspect it is comparable to things like Python and Ruby.
The syntax is reasonably clean (one reason I compare it to Python and Ruby). If you hate Python and Ruby and love Perl, you won't like Rexx. If you're on the opposite end of the spectrum (I am), you may like it. (Although if you already know Python or Ruby, I don't know that there would be any great advantage to doing Rexx.)
Rexx had one feature which I absolutely loved, and I've never seen another language with anything nearly as good. It had a trace function which showed you what the script was doing, first as the actual code, then the results of that code. (A similar function would be "bash -vx script".)
What made Rexx's debug facility so much better than everything else was that if turned all the way on (you could turn on and off various things, like whether you wanted to see variable values), it would show you not only the original code and the result, but all the intermediate results, as well.
For example, say you had a line that said something like:
x = ((( a + b ) * ( c / d )) * e) ^ f
To make the example simple, a=1, b=2, etc.
Rexx would show you:
x = ((( a + b ) * ( c / d )) * e) ^ f
x = ((( 1 + 2 ) * ( 3 / 4 )) * 5) ^ 6
x = (( 3 * .75 ) * 5) ^ 6
x = (2.25 * 5) ^ 6
x = 11.25 ^ 6
x = 2027286.5295... (I'm not bothering with the whole thing, ok?)
I had one program that I had written in clist language. I found a bug, and past experience led me to think it would take about two days to find and fix. Since I wanted to learn Rexx, I ported the program over to Rexx (which wasn't trivial, but it's also not important to my point). I still had the bug. Using the trace facility in Rexx, it took me 15 minutes to find and fix the bug. I was in love with the language from that point on.
A sphere you walk on top of would probably be easier to construct, but unfortunately, either way has the same problem, because you're wrong about one thing.
m l, here's what I came up:
.0125, which is the sine of the triangle.
Cosine(arcsine(short/hypotenuse))=0.94 inches.
It would have to be quite large to seem flat.
Thanks to a helpful page on chords at http://mathforum.org/library/drmath/view/57832.ht
Assume a 30 inch step.
That makes the short side of the triangle 15 inches.
Start off with a sphere 10 feet in radius (20 feet in diameter).
15/120 =
A 1 inch height difference would certainly be noticed by me.
Assuming a 0.1 inch difference as small enough to be ignored, your sphere would have to be about 94 feet in radius. (And remember, that's radius. It's almost 200 feet in diameter.
Considering that's what would be required for each person in the game, I think what they've got is definite improvement.
I'm not impressed by the photo, though. It doesn't look like you could (safely) take a step forward, unless those blocks are really fast.
To anyone who complains that I should have done that in metric:
A) I'm a Merkin. (See alt.fan.pratchett on Usenet) We're allowed.
B) I'm at work and trying to be reasonably honest with my employer's time...
Time went by (about 8 years) and I changed jobs a couple of times. One day, I was talking with a daughter site in Boston, and somehow the subject of Adventure came up. He was trying to do something with it, and happened to mention the name of the userid that created it.
It was DTSO403, which was my old userid at the original company.
I have no CLUE how it made its way from Dallas to Boston...
Technology has marched on, and the world has changed (again).
All the trends in technology over the last 10 years say that privacy as we have known it, is headed for extinction. Cameras that get smaller and smaller, remote controlled robots, hacking into wireless LANs, PLUS all the electronic interactions (like RFID) that are coming, PLUS computers getting cheaper by the day... This all adds up to privacy basically being impossible.
Proprietary software is doomed, because the Internet made the level of interactivity that open-source software needs possible. For exactly the same reasons that the medieval guilds (with their proprietary methods for things like ironsmithing and glassblowing) were doomed once the movable-type printing press was invented, proprietary software cannot compete. In the near term (5-10 years), it will still have a solid space in niche markets, but I'm not even sure that will last. It certainly isn't going to last in mainstream software arenas like OSes and databases.
But that same increase in processing power and decrease in communication delay means that doing things like examining every electronic transaction that someone performs (and building a detailed profile of their life from it), is not only beginning to be possible, it's very nearly inevitable. Even the most paranoid of you out there (and on Slashdot, the percentage of paranoids is a good bit higher than average) would not want the sort of draconian methods that would be required to prevent it. (No computers and no networks, for instance.)
The proper solution, I think, is to change our culture, so that it doesn't matter that someone knows the kinks in my soul.
I am mostly connected to reality, so I'm not holding my breath on this cultural shift, but I really only see three possibilities:
We turn Luddite and roll back the clock technologically. (Not likely to happen voluntarily by most of this audience, but some of the non-technical types turning Luddite IS all too possible.)
Privacy gets moved to the same status as apprenticeship - it's something that existed historically, and it's occasionally useful for analogies, but it's not part of anybody's life anymore. This could either go the Japanese route (I believe the usual phrase is something like "Nakedness is frequently seen and never noticed." In other words, commenting on someone's quirks is far more shameful than having said quirks to begin with.), or simply an open acceptance that other people do things differently than you.
The third possibility is the one that worries me. That's a totalitarian society (probably theocratic) that uses this information to control people to a degree that has heretofore been unbelievable. I don't think such a state would last very long at all, but the creation and destruction of it would get really, really ugly.
The US is the only culture I have extensive first-hand experience with. I would strongly prefer to see us go to option 2B (taking the attitude that you can live your life any way you want as long as you don't hurt me).
That fits wonderfully with our stated national beliefs. It's an absolutely lousy fit with what our behavior says we believe. The behavior (IMO) says we urgently want #3.
That's the big reason the 3rd option worries me. I can very easily see a theocratic state as an intermediate step to the live-and-let-live one. If anyone has any practical, pragmatic suggestions for how to create such a cultural shift (one suitable for a total absence of privacy), speak up now, because the situation could get critical within 10 years, and is almost guaranteed to get critcal in 20.
Combine a lot of people NOT changing off of 98 with a virus/worm that cripples a company and cannot be fixed and anyone trying to sell Linux as a desktop OS would have a field day. It would be a horrendous PR mess for Microsoft.
I'm pretty sure MS did NOT want to do this. The fact that very large numbers of people (both businesses and individual consumers) refused to make the switch forced their hand.
At my company, we were using the lack of support for 98 as an argument for why we should upgrade a significant number of machines to XP. Our CIO was very resistant (he didn't like the licensing terms). What it boils down to is that Microsoft blinked first.
I started out thinking about all the ways that I could demolish his argument (not terribly hard), when the thought suddenly occurred to me that, in a sense, that was exactly what SCO wanted.
Let's use a little common sense on the whole SCO thing.
First off, toss out a couple of emotionally popular but rather unlikely theories:
1) Darl McBride is insane, in the medical, legally-incompetent sense of the word. If he was, the rest of SCO and the Canopy Group would have shut him down a long time ago. Similar reasoning applies for "unbelievably stupid", "totally ignorant" and similar epithets.
2) #1 is true, but for everybody in SCO and Canopy, not just Darl, which is why he's still there. Ok, c'mon. If you're that out of touch with reality, talk to your doctor about upping the dosage on the little green pills.
A common allegation is that Darl McBride is a greedy money-grubber with the morals of an advertising exec. This is probably true, but in America, at least, being greedy on behalf of your company is not only not frowned on, it is (somewhat) required by law. (There is a SEC regulation that requires all publicly-owned corporations to try and maximize the value of their stock. Since it says nothing about HOW that's to be done, the range of tactics is pretty wide. You won't find any American corporations saying "We gave away all our profits, 'cause we felt like being nice guys", though.)
SCO's basic strategy is obvious: Fire the shotgun everywhere possible, as often as possible, and see what sticks. I saw an article the other day that said the software business in 2004 was predicted to be about $230 billion. If SCO can get 1/10 of 1% of that, they'd be ecstatic.
Another characteristic that McBride has, he shares with lawyers, politicians, and most high-powered types in business: a thick skin. I very much doubt he has been bothered in the least by the various vilifications called down on him by his detractors.
I think he was a bit surprised at it when it first started, but since then, it's been more of a weapon in his arsenal than anything else. If he is not bothered by name-calling and accusations, but the other side (the open source community, in general terms) is, then the more furious the argument, the better his odds of being able to find a weak point and exploit it somehow. At the very least, he (and SCO) can point at all the ranters and ravers and claim "With enemies that act like THAT, doesn't it make sense that we're the ones in the right?" (Something along these lines may be what got the money out of Baystar.)
If you're caught up in a strong emotion, you're not entirely sane. If you're angry, all kinds of little things you would ordinarily blow off make you even angrier. If you're ecstatically happy, you can find a silver lining in a mushroom cloud.
The reason I brought up politicians, lawyers, and CEOs earlier is that they all have one thing in common. I called it a "thick skin" earlier. Another way to describe it is that they have the ability to climb out of their emotions and think rationally again about whatever the subject is. That ability is what gets them paid the big bucks.
So what I think SCO is doing with a lot of the more unbelievable claims they've made (like the attack on the GPL last night) is not to seriously convince anyone of that position, it's to stir up trouble. The more emotional the opposition gets, the better the odds that something, anything, will happen that he (they) can exploit. It goes with the shotgun approach: the more you get things stirred up, the more targets of opportunity there are.
Fortunately for IBM, they have good lawyers, who haven't been influenced in any way they shouldn't be by the public furor. They simply stuck to the facts and the law. My favorite element of what they've been saying is that it's mostly in plain English. When one side speaks English and the other sid
Then somebody at Red Hat is lying. My account rep (who I'll be nice and not name for now)told me yesterday, when discussing this exact subject, that we WERE buying (well, renting) a license, and that we most emphatically could NOT install it on multiple machines. (Which is 180 degrees from what a different account exec told me 6 months ago.)
This is the major gripe I have with the new scheme. If I want to put Linux on an unimportant box (say, an X station for the operators), or a temporary box, I must pay Red Hat $350 (absolute minimum) AND I must wait a week to ten days for them to process my order and send me the boxed set. (That statement is per the aforementioned account rep.)
Here's the way I'd like it to work:
You could install it as many times as desired, but get no support at all (not even RHN) for free
For a reasonable price (say $100), you get RHN only -- no support at all.
THEN get to the $350 if you actually want any kind of support
What we've got (per the account exec I mentioned) is item #3 as a option, and ONLY number 3.
I think you misread what painehope said. He did not say he wanted it for $50 for 512 servers -- he wanted it for $50 each for 512 servers, $25,600 total. (And $35,840 for the 1024-node example.)
Since almost every server in the cluster is identical (by design), this is going to be about as hard for Red Hat to support as one box. (They will strongly tend to all have exactly the same problems.) In a few cases, you're going to have issues between boxes, and a few boxes will be different from all the rest (admin server, control point, managing node, that kind of stuff.)
I would assume they have a small test cluster (8-10 boxes) for working out bugs, especially the between-systems types. When the problem is solved, you install exactly the same thing on all 512 (or 1024, or 18 gazillion) machines.
He was even willing (even eager) to provide a caching RHN server (or perhaps he meant buy one, since RH does sell such a beast.)
So from Red Hat's perspective, this is going to be like supporting 10 or 15 different boxes (and I'm being pessimistic -- the equivalent of 4 or 5 is probably closer) and the RHN bandwith of ONE, for $25,600. That's actually a pretty sweet deal for Red Hat.
Painehope's problem is that Red Hat isn't interested in this. They say he should pay them $179,200 for 512 nodes. (Or $358,400 for 1024.)
The upshot of this is almost certainly going to be that people using the really big clusters (all the ones that make the "New supercomuter record!" headlines) are NOT going to be Red Hat.
The article made the comment that people didn't think of copying software as stealing.
That's for the very good reason that it's not. If I steal something from you, I deprive you of the further use of that item. If I copy a CD, you haven't lost anything.
The argument Microsoft (and other proprietary digital-media makers, including the RIAA people) make is that I'm depriving them of the money they would have gotten if I'd purchased it.
That holds water if, and only if, I would in fact have spent the money if copying were unavailable to me.
At the equivalent of $10,000 a copy, I will guarantee that very, very few people would buy a copy if that were the only way to get Windows. They would simply do without. If Windows were the only operating system in the world and uncopyable (get that glazed look off your face, Bill. I said "if",) then the result in Vietnam would not be 2,000,000 people * $140 = $280,000,000 more in Microsoft's pocket, it would be 200 people * $140 = $28,000, which is probably about what they really got. Therefore, since MS has no less money than they would have had anyway, nobody is stealing, not, at least, by the argument of "what we would have had".
In the case of someone like me (an American, making well over the average US income), and a $15 CD, the argument is not as clear (and obviously self-serving), but in a case like the average Vietnamese (or Thai, or anywhere else the per capita income is drastically less than developed countries) I would say the argument has some weight.
In real life, if everbody in the world took "piracy" seriously and paid Microsoft full price, they would get 10%-15% more income (almost all of it from developed countries), which is probably exactly what they expect to get out of it. All the fuss and fury about hundreds of millions of dollars is just another marketing tactic.
There were two quotes in that article I really liked, both by the Microsoft rep:
With regards to the Open Source community: "They give away innovation."
I like it. Not strictly true, but I rather like the idea.
"We encourage the government to lead by example."
I have bad news for you, guy. They ARE.
I disagree. There is only one front that spam can be fought on successfully, and that is economic.
When spam is no longer profitable, spammers will give it up voluntarily. As long as it is profitable (especially as profitable as it is right now) people will continue to do it, regardless of one law or a thousand.
Each law passed to try to stop it restricts someone's freedoms, and it's pretty much unavoidable that the law will restrict some who do not deserve such restriction. If that would eliminate spam, that might be considered a reasonable tradeoff. Since it will not, each and every law passed solely on the basis of "We've got to DO something!" is a bad idea.
The best thing I've seen so far is A Plan for Spam. I tend to agree with the author's assesment of how likely it is to work, specifically because it destroys the economics of spam.
I think most of our collective energy should be going to integrating things like "A plan for Spam" into common email programs, possibly extending it to allow people to join anti-spam clubs, so they don't have to label the thousand or so emails the thing needs to train with themselves.
Simply having a button that says "Spam!" will help a lot of people deal with the frustration involved with spam, and if it is pitched to the general public as "Every time you kill an email as spam (instead of just deleting it) you're helping to put spammers out of business," you're going to have to beat them off with a stick. People would love to have a way to get back at spammers.
For precisely that reason, they won't investigate every complaint, or even a large fraction of them.
Like all complaint-takers for decades (if not centuries) they will investigate things based on:
Frequency of complaint
Precision of complaint
If 500 different people complain about the same thing, it's going to be taken more seriously than if one person does.
A complaint like "Microsoft screwed me blind! I want you to GET those evil bastards!" will probably be handled by a spam filter.
(Warning to those who are overly literal-minded: the following example is completely fictional!)
A complaint like "On Friday, Jun 14th, Jahfhs Ffjl of the Microsoft Licensing group told me that I would have to pay full retail price for Windows XP Home, rather than the usual OEM price, because I sell a line of computers that run Linux," will be much more effective.
Specific dates, names, statements, amounts, etc. will get far more attention than vague hate mail.
Finally, whatever they think should be done will get choked back to what can be done, given their budget.
End result: The most flagrant violations will get curbed quickly (or may never happen because they would have been caught so quickly).
Borderline cases will break down into two groups:
The ones Microsoft doesn't care about will get "cured", and used as public-relations fodder by MS.
The ones Microsoft decidedly does not want to get stuck with will be challenged in court, or similar stalling tactics.
In a few odd cases here and there, Microsoft will get their hand slapped and have to legitimately clean up their act some.
Bottom line: As usual, it's between the two extremes mentioned. It will do some good, but there's a lot of stuff that will still fall through the cracks.
My only complaint about it is that most of the good will come in the next year or two, and it will probably eat up money for then next couple of decades...
Peter Veeck wrote a bunch of codes and standards for buildings and utilities. As best I can tell, he had them up on a Web site, available at no charge to anyone who wanted them. (Being available free like that does not affect his copyright - he would still own it.)
He then proposed these standards to some of the nearby towns as proposals for laws.
At least two of the cities took him up on it, and passed the codes into law in their municipalities.
SBCCI made a reference to the codes as being "public domain". Veeck objected to this.
As best I can determine, he was not trying to prevent anyone from accessing the information, he was just saying "This is my text, I wrote it, I have the copyright on it".
The court said, in effect, "Not after those cities turned it into law!".
There seem to be two key elements: First, Veeck offered his writeup of the codes to the towns to be made into law (i.e., they didn't steal it from him). Second, they incorporated it by reference, i.e. saying something to the effect of "The codes that Peter Veeck wrote up are now law." (I didn't dig deep enough to see whether that was his idea or theirs.)
All laws are public domain (having to pay a fee to see a law that you are bound by might start another revolution...). What the court said here is that by that specific code (the one he wrote and (I think) had on his web site) becoming law, the "public domainness" of law overrode his (ordinarily perfectly good) copyright, so SBCCI is right, and what he wrote is now in the public domain.
Ooogh. Double-plus ungood. Well, now we know what the next advance in touchscreens needs to be...
The typing issue is trivial. I'm nearly certain you can plug in a standard keyboard (it's basically still a laptop, after all), and for those occasions when you can't, a simple app could put a keyboard on the screen and you can type on that.
Also, there's a neat gadget ( http://www.time.com/time/2002/inventions/rob_keybo ard.html )
that could be incorporated pretty easily, although I'm sure it's not in the machine under discussion. (Sorry if the URL is broken in half -- I couldn't figure out how to fix it.)
That's just one reason I consider a passive touchscreen (anything which can depress it can activate it) to an active one (has a special pen that is the only thing that will activate it.)
Here's a few others:
* Losing your stylus doesn't require a $15-$20 (at least) replacement, and render your machine completely dysfunctional until you get it.
* It allows software to become much more intuitive. The original concept of a mouse was to simulate being able to touch the objects on the "desktop" and move them around. The objects will still be a simulation, but the touching will be reality.
This has one medium serious drawback, of course. It means Apple was right to stick with a one-button mouse all this time...
* It opens up a whole bunch of new interface interaction possibilities. For example, to address the one-button,two-button,three-button issue, you could tap with one, two, or three fingers. (Some of the stuff that's been done with right-clicking I actually like. I'd rather not give it up.)
Even more, it makes possible interactions that could not possibly be done with a mouse, such as moving a window around with one hand and resizing it with the other at the same time (or resizing with two fingers on the same hand); or moving two windows (maybe even three or four if you've got big hands and coordination) at once.
How'd you like to design a web page layout by literal "hands-on" design? Eye-hand coordination is built into the species. Remote control devices like mice, trackballs, graphics tablets, etc all make it more difficult to use your mind for the creative, aesthetic side of the work. (Geeks, who usually don't have any aesthetic sense worth speaking of, and who use mice and keyboards more often than any other object, will not get as big a boost in productivity. Sorry.)
Add to that any other activities that involve multiple fingers (musical keyboards, 10-key pads, phone number pads and so on), and I consider a passive touchscreen a much better option than One Pen to Rul... Sorry, started to get carried away.