Granted, the poster you are replying to is giving an intuitive demonstration rather than a rigorous proof, but you are proposing mathematical nonsense.
Well, I sure as heck am not trying to prove anything, so I suppose that throws us both in that camp. As for nonsense, I think you're dismissing me a little out of hand.
I'm not going to beat around the bush with you here - I think you're misunderstanding because of a notational misrepresentation. What I think we both agree on is that
0.99 ~ 1 (that is, "is approximately equal to") and that
lim (1-x) = 0 x->1
and even that, in pseudocode, makeone() { real x = 0; real i; for(i=0.9;0;i/10) x += i; return x; }
and makeone() ~ 1
But you won't convince me short of a real proof that 1 = 0.999... - I would be delighted for you to show me wrong, but please keep nonsense out of this until you can, Mr. Rinker. I mean, come on - if 0.999... was 1, we could just call it 1, couldn't we? Oh, WAIT, that's true by definition. Odd, that.
Eh!? Infinity + 1? Index on infinity? There is no 1 at that position. The zeros stretch on interminably to the horizon and beyond.
Ah yes, the road goes ever, ever on. I was trying to make a point, to a crowd who's used to dealing with arrays. Since I've seen a few of these comments popping up I'll handle it here:
Infinity + 1 is easy. Let's say x is an integer and y=x+1 and just for illustration let's say z=x. Now imagine the graph of y vs x imposed on z vs x - you'll have two lines (one for y, one for z) and everywhere on the graph, y will be higher than x, for any x. So if we send x out to infinity, y will still be bigger by exactly 1. y=infinity + 1. Although, just *try* getting a address space that size in any language but Haskell!
If you're still having trouble, try this thought exercise: Let's construct our infinitely repeating number like this: write 0.01, then insert 0's just to the right of the decimal forever. You send the 1 off "infinitely" to the right, but it is still there... see, I'm really describing a limit - you know that for all intents and purposes, it's close enough but the hair-splitting truth is that unless you *explicitly* neglect that 1 you cannot say that 0.0000...001 = 0. It is never true, no matter how many (even infinitely many) 0's you put in front of it. All you can say is that it's so similar nobody but Zeno cares.
What does it mean for one number to be equal to another number? Well, if x = y, then x - y = 0. In this case, what's 1 - 0.9999999999999...? The answer is 0.000000000000....(infinite number of 0's). Since 0.000000000.... = 0, they are indeed the same number.
I beg to differ. After that infinite number of 0's is the digit "1" at position infinity+1 (or -infinity-1 depending on how you index) - that is, perhaps a more helpful notation would be 0.00000...001, which != 0. More true would be to say that if x=1 and as y approaches 1, x-y approaches 0 so we can select a y to get x-y arbitrarily close to 0. Hence, as we all know, we can get close enough so that nobody cares any more, but mathematically speaking I'm afraid what you've been telling people has been subject to the underflow error! Ever work for Intel?:)
how can you take a 'pill' of nanites and, 1st, have them survive the digestive system - it can get really acidic in there...
Easy, you wrap the pill in a bunch of starch and instruct the taker not to suck on it. The starch won't get digested by the enzymes in the stomach, but will be by those in the duodenum (intestine nearest the stomach).
Nonsentient devices are programmed by a human. A nonsentient device can evolve (like a virus) but I see no short term reason why this independant evolution would be faster then viruses evolution. We are not talking the virsus used for gene therapy here. We are talking fundamntal improvments to the evolutionary mechinism of a virus which "good ol' mother nature" has ben working on for many millions of years.
Natural virii reproduce asexually, and hence do not have a lot of genetic diversity. They mutate randomly and thus changes in their genetic code are not linked to the success of the previous version (unlike sexual reproduction, in which the two parents must have survived to maturity). Changes in their genetic code, like in other living things, happen between generations, when they reproduce. This would not need to happen with nanotechnology.
I suppose you might design a nanobot cold virus killer which used group processing to evolved new attacks as the cold virus evolved new defences. This group processing would take the form of units 1 to 1000 try this and tell us if you live or kill viruses. I agree that something like this could have more potential for killing humans, but we are a long way away from something like this. Also, I suspect the communications channel betwen the nanobots would presuppose the ability to include a self destruct.
Now here I see that you agree with me. Yes, we are a long way away. Can you guess if we are more or less than 30 years away from it? I cannot, and I suggest that we should consider the case now, in order to make good decisions in the future. However, don't allow the self-destruct to let you feel secure. After all, the self-destruct is part of the code on the robot, and therefore can be selected against. Which it would be, if survival is a preference we ask the robots to select - and why wouldn't we want our tools to keep working?
I never said we should not predict.. just that Bill Joy dose not know what he's talking about since his predictions are based on sci-fi instead of real theory. We predict the progress of the fiels you listed since we have a theory for them. The summery essentially said that the speakers who understand any present theory of biology and nanotechnology dismissed Bill Joy as a luddite. They were nicer then I was, but that's because they were only concerned with how Bill Joy was wrong.. where I think the things Bill Joy advocates are themselves far more dangerous.
You suggested that since technological predictions rarely pan out after 30 years, why bother? I don't think Bill Joy is a luddite. I don't think the summary pointed that way either. Yes, some of the things Bill suggested seem dangerous, however you're still missing the point that his caution is not one of them. Caution, in these cases, is good. How can you limit the "smartness" of a nanobot? How can you keep the nanobot, or rather the nano-swarm, from evolving around any limits you place on it? These issues should be dealt with, and we do that by increasing research in the area, not decreasing it. This doesn't mean we don't need to deal with the issues - that would be throwing the baby out with the bathwater.
Summery of the summery.. Bil Joy is a moron or an autocrat who wants corperations and big government to rule the future. Joy advocates restricting access to the technologies which will shape the future. This is morally wrong in soooo many way, but I will try to explain a few.
Now, it's rarely a good idea to resort to name-calling unless you have no ground to stand on and must resort to attacking the messenger rather than the message. You actually do have some ground to stand on, so let's get to the point.
I think you may have slightly missed the point. It is a common reaction that if a person can harm you using an area of knowledge, then that knowledge should be kept from that person. If the identity of the particular person is not known, then keep it from everyone!
As you've astutely noticed, this reaction is not productive. It's yet another incarnation of security through obscurity, and it doesn't work. There are fabulous examples in speculative fiction of this concept carried to its logical conclusion - Larry Niven's ARM springs to mind. Someone, somewhere, will come up with the same idea and then you have to either convince that person to keep quiet too, force them to, or allow the cat to leave the bag.
It doesn't matter who develops the knowledge. It doesn't matter how risky the knowledge is. Everyone should have access to it so everyone knows what to do to stop someone from harming someone else with the knowledge.
You seem to have missed what may be a valid point though, because I'm not sure he made it strongly enough - computers have the ability to change the information that fundamentally dictates their behaviour at a rate orders of magnitude higher than humans, and hence can implement "evolutionary change" much faster. It is not impossible that these nonsentient devices could become an "enemy" in and of themselves. No person would be needed to cause harm to another using the machines, they would do it themselves.
It is from this enemy that we must keep the information. It must always be well within our grasp to rein in our machines before they enter the phase of their "evolution" in which they begin to compete for real-world niches. They don't need to be intelligent for this, they just need to require something that they can get at our expense.
Plus, you should consider that no virus or bug has managed to be 100% lethal to humans. Shure, a very rare few wipe out 80% of the population, but we are not talking extinction as a wore case senario here people. Actually, your probabllity of being killed from a metior smashing into this planet is MUCH higher then any risks from bio/nano-technology. (Note: I suppose biology research reduces you chance of death as a result of metiors far far more then it will increase you chance of ding as a result of terrorism)
Also, non-science people really seem to have no understanding of the way these sorts of things progresses. The situation is summed up perfectly by John Holland's quote "Predictions 30 years ahead have always proven to be wrong, except in the cases where there is a strong theory behind it," i.e. without a theory it's just luck (which can take a LONG time).
In the end, your quick dismissal of this rational and intelligent person's reasoned opinion lends the impression that you have jumped to conclusions. The largest point you've missed seems to be that, while non-science (and by this I assume you mean applied sciences such as computing science) people may not seem to you to understand science, by the same token "science people" may not understand applied sciences.
I will not argue historical points about the effect of plagues on humans. However, I will claim that in many cases, the survival of a human attacked by some infectious disease is completely dependent on medical intervention. In order for that intervention to occur, we must have had time to adapt to the disease's method of attacking and surviving. When we can't, people die (see AIDS). If the attacking disease were sufficiently different or mutated fast enough, we would die. If the disease were widespread enough, we would All die, to the man, woman, and child. Imagine if colds were lethal - have you ever had a cold? Do you know anyone who hasn't? And colds need hosts to proliferate, unlike nanobots. If we cannot limit the rate of "evolution", we should not create that agent.
Your assertion that because predictions have failed to pan out seems valid, but the conclusion you draw does not follow from it - that we should not predict is ludicrous. I propose to you some predictions: 30 years from now, computers will be faster, some people will be greedy and some will be lazy. Manufacturing will be more advanced. I think those predictions will pan out. Now, if that continues, at some point we may need to worry about these things. I for one propose that we discuss them now, so in the event that the technology is developed to end the world by a non-human hand, we know what to do and not to do in order to keep that hand from striking.
Why the hell is this marked "Troll"? Signal 11 has some really good points here, that more people should consider, carefully, before they speak to decision makers in their company, community, or family. You know, I can't actually find a single point that he makes that isn't backed up by reliable sources (including my opinion, which is reliable to me:) ).
Where's my moderator status when I need it?
In fact, the only problem I have about breaking up MS is that it does not happen according to their terms. Personally, I'd like to see their Office division left with the hardware division, and OS broken up by product line (take that!) but then, I'm vendictive...
It's a carbon copy of the Hitler Youth program used so successfully in World War II Germany...
I wondered how long it would take for Godwin's Law to take effect in The Katz Papers. Whoever said it was limited to USENET postings?
For those who don't know, Godwin's Law is (according to the Jargon File):
"'As a Usenet discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.' There is a tradition in many groups that, once this occurs, that thread is over, and whoever mentioned the Nazis has automatically lost whatever argument was in progress. Godwin's Law thus practically guarantees the existence of an upper bound on thread length in those groups."
The idea is by the time people start bandying about comparisons to the holocaust (and let's face it, are/were there any gas chambers at *your* high school?), any rational discussion has long since been lost in the growing noise of buzzwords and purely emotional appeals.
That being said, it's too bad Katz threads seem to almost immediately degrade into analysis of the Katz article in question, for often (as in this example), he has some interesting things to say.
Actually, Photons have a zero mass all the time. What they have is momentum... it's kind of counter-intuitive to someone used to thinking about a Newtonian universe where momentum = mass * velocity.
Turns out that that's only part of the equation, you see. Bottom line, photons have no mass. They do have momentum. If you really want, I'l calculate it for you.:)
To put it simply: "the source code, the whole windows source code, and nothing but buildable source code"
Nothing? Forget that! I want the source to Solitaire and Minesweeper - without those integrated tools(tm), my Windows experience would be but a pale shadow...
They say they didn't contest it after a while because, put simply, they ran out of money. Kind of like what would happen if you tried to go up agains the MPAA in court by yourself... they'd stall, you'd lose because they have all the cash. There was a/. article about this... Ah yes, here it is.
That's kind of ironic don't you think, considering this is all the fault of the US DMCA? Shouldn't Canadians be complaining about the greedy, corrupt, freedom-stealing Americans?
I'm not, mind you. I understand that the DMCA wasn't something that most Americans wanted. I do kind of resent the fact that Canadians are being held liable in their own country for American laws (and stupid ones, too). Then again, we have a long history of letting ourselves get pushed around from the south. I suppose I should just get used to it... manifest destiny and all.
Re:You really cannot do this with the current TCP/
on
Geographic Screening
·
· Score: 2
Actually, that doesn't work because of the inefficient routing that happens all the time. For example, if I want to get to my friend's ADSL-connected machine from mine, just across town here in Vancouver, BC, my packets travel south through Tacoma, WA before coming back up the border. Your method would have me placed somewhere in the USA, but I'm not.
What would probably happen is that it would just be driven underground more
I thought of this, and perhaps it was naive of me to think that most companies would not feel comfortable offering under-the-table bribes to people for their domains. Even so, a policy could be created (not a law) that forces a domain holder to give up the domain as soon as they offer to do so in exchange for money. A company could then go to ICANN and say that the person attempted to sell them the domain, and presto, no more domain ownership (and perhaps no refund either). This causes a problem with false accusations though, which means that more complicated rules are needed to prevent this from happening. Basically to defend from an accusation like this would mean that you could not get rid of your domain (to do so would be to admit guilt).
Alternatively, it could be made a point of policy not to award domains to entities who paid for their release from scalpers. This would mean that companies could not buy domains from people and the trade would dry up. While fly-by-night domain scalpers don't care about under-the-table deals being hidden from view, don't companies have to make their accounts public to some degree? Here's my naivete again, I have no idea.
These arguments are based on the assumption that companies wouldn't want a back-room deal to smear their public image and that the risk of doing so would keep them from doing truly shady deals for a domain. Of course, having some law passed might make this easier, just like ticket scalping.
I suppose that where they do control who the registrars are, that they could "blacklist" non-compliant registrars.
True, they don't control all the geographical domains like.ca,.to,.tv, whatever. I don't think these domains are hotly contested over like.com,.net, and.org - and these are completely controlled by the US as far as I know. Perhaps there's more leverage there. I do agree that it's to the registrars' advantage to have people registering hundreds of domains for no good reason though.
I've got to agree with your sentiment here. Recently I went shopping for a shiny new domain name for a domain I and some friends are putting together. In the 100 or so domain names I happened to come up with, I found about 4 that were not taken. While I kept no good statistics, I'd be willing to bet that 60 of those were taken by squatters like noname, inc.
Well, this method that the ICANN is chosing won't help me at all, because I don't have any trademark rights to anything and if I want to register some domain that these guys own (or will soon own) then I'll have to go buy it (which I will NOT!). Here's a question for you: why don't we just remove the incentive to squat on domains by making it impossible to sell them?
I'm serious, is there any reason besides greed that would motivate someone to register a domain and sell it to someone? I honestly can't think of any. A quick pick:
A company that starts a project, then cancels it might want to recoup their costs, but we're talking like $70 here and most companies that I know (even the ones run out of basements) wouldn't be too worried about that since it would cost more to pay someone to try to sell it.
Someone like myself who gets a domain then wants to change it could simply register the other domain and abandon the first one. Perhaps it would be possible to sell the domain back to the registrar (say to get back the cost for the second year or something) on some kind of pro-rated basis. Why would I want to try to sell it, do I look like a registrar?
A person or corporation who is trying to protect their name, trademark, or whatever wouldn't want to give up the domain.
So why don't we do this? Make domains only brokerable between the registrar and the "owner", with no possibility of profit? I think that would stop domain scalping. IANAL, I have no MBA, and I'd love to hear from people who are more enlightened about this than I am.
UCITA would be pretty bad if passed in the USA. I'm from Canada, and as far as I know there is no UCITA equivalent up here. Theoretically I shouldn't be worried too much, but I am since Canada often follows the US' lead on policies like these. In the meanwhile though, I'm confused by one thing, derived from the fact that companies do not make different software for Canada. We even use the same US-crypto software as the US does. What happens when somebody doesn't like something my company does and turns off my software?
Well, I suppose that my employer would get upset and launch a suit against the vendor if the vendor didn't turn our software back on after we complained nicely... but would they win?
IANAL, so I'm not sure but it strikes me that Canadians and Canadian companies aren't covered by the same legislation, excepting international copyright law. Wouldn't this make UCITA a dangerous tool for vendors to use? Any time software ends up outside the US they'll be liable for damages from their time bombs, etc. Does anyone with more knowledge have a take on this?
This discussion strikes a particular note with me because I believe the Linux UI to be the Wild West of current computing - it's not really formed, and it's no place for the uninitiated. Because of this, I'm pointing my career towards UI development and Human/Computer Interaction. In a couple years I hope to be able to do some really neat work here, and the prospect of being able to work with a large Open community effort on good Linux HCI is really exciting. That said, to the point!:)
it seems like an awefully formidible task to try to be all things to everyone
Hey, everyone, why is this such a big deal? Let's look at this like a programming problem (surprise!), because it maps really really well.
Our problem as I see it: There are features in Linux which are not needed by some people, and as such they tend to "clutter" the features they do need. There is no easy model they can apply to let their real-world trained conventions make this problem easier, so they are stumped.
Allow an analogy: A moderately complex C library will probably have features that say a first-year computing student won't need, and as such they tend to "clutter" the features they do need. One easy answer is to get the 101 student to use C++ and slap the C library behind a wrapper C++ class which hides the features they don't need . Now, here's the kicker: If they want to use those features, they just stop peel back the wrapper and access the library.
What we seem to be talking about here is UI, not OS complexity for truly, a user sees the UI and only kernel hackers and other coders (ie, not users but developers) see the real guts of the OS. So why can't we do the same with UI complexity as we can with that C library? Hide complexity in layer upon layer of black box abstractions.
Now a lot of you are going to say "That's slow!", and you've run into my poor analogy there. I'm not advocating acually using layer upon layer of classes or real programming abstractions, I'm not talking about one UI running over another UI etc etc. I'm talking about layers of perceived complexity. Top is the playskool desktop, with three or four big buttons on it ("Browse Web", "Write Email", "Type Letter", "Go to the Window Manager mode"). The last option lets you use the window manager, and from there you can get to something else like a command shell or better...
At each level you can have different complexities of configuration tools. In Windows, there are two levels: Control Panel, and registry editing / INI file editing. For many people, the first is too simplistic and the second is too unfriendly to dig through. In Linux, there's really one level for most things: Edit config files. This is the "all the features, all the time" problem we had with that C library and the 101 student. Surely we can do better than Windows' 2 layers, and I'd love to see it.
We don't have to make an OS that is everything to everyone, because that's not possible. We do have to make it easy enough for people to configure it themselves so they can customize it to their needs - ideally they won't have to but we can make it easy for them. The key is that *our* OS won't make the next step prohibitively complicated, but rather ease the user into more and more configurable options. It won't tell you where you're going to go today unless you want it to.
Linux is useful. Only some of us don't realize that (Hi, Bill!). Let's make it useable, and put that up as one of our major issues, not just a distraction from more features in grep or a prettier desktop environment (pretty != useable).
Is this the only way? Probably not, but it sounds good to me. Is this doable? Yes, but we'll have to junk a whole bunch of stuff, probably including the X windowing system as we know it today. More importantly, what do you think? Is this the way to go? Would you like to use it? Would you like to work on it, because that's how things happen in Open projects? Do you know about projects that do this kind of thing (please post them so we can all bandwagon and make it a success!) or know someone who does?
But the experienced sysadmins and free/open-source developers are doing precisely what you suggest on Slashdot, and the academic-style Internet old-timers with their well-reasoned logical posts likewise I expect, whatever their diverse experiences.
I'm afraid I disagree with you. I think there are many very technically literate people reading/. who post their opinions and horror stories and success stories because they know that here, they will be commended for it. Ever notice how stories which are Linux vs. MS always seem to have like three times the number of comments as other articles (and I'm talking about non-grits posts)? How many of those comments are from complete fools? Not a whole lot, or I wouldn't read them. I tend to read posts at 2 or above, because I don't have a lot of time. Most are pro-Linux. Most don't present an objective view. Most are from people who know whereof they speak. Very few I would be proud to show Bruce, my very Pro-MS friend. This is the yardstick I use: could I use this to convince Bruce? Only rarely!
I recently read in an article in the Globe and Mail about a study which showed that the skills to recognize competence were the same as the skills for competence. This could also be said, "Don't pat yourself on the back", and it means that if we are complacent that the problem is with "other people" then nobody will change, even if in your case it happens to be true. Next time you're talking about Linux, check yourself. Show some of the other side of the story. It broke my heart to recommend Win 98 SE to some people I know, but it fit the bill for them and to make them use Linux would be stretching them to fill the Procrustean bed.
Set an example! Ignore the moderators and speak the truth. The whole truth. After all, this is how Open Software works: lots of eyeballs, and all of them looking for problems to fix.
Since the zero-cost base O/S means that Linux sales will never be able to support a marketing budget of any size, it's only right that other forums take over that role, like Slashdot does.
Could be, but if you present a well-balanced, carefully considered argument you're much more likely to win a debate. Mudslinging attacks the integrity of our argument, as can be made obvious by the opinions of people such as the person you're replying to.
And that doesn't happen in pure propaganda forums. What you get here is bias, yes, but it's a clued-up bias.
I would still rather see a more balanced view, because it would allow us to address real issues with Linux so that they stop being issues. To hide flaws behind an argument that "MS obfuscates their OS flaws, therefore we should too" means that we obscure our supposedly free, open sourced OS.
Now, before you drag out your flame throwers, please bear in mind that since most people haven't the technical ability or simply the time to go through and examine the code, or even to install the OS and brave out the weeks of stigmatized clueless newbiehood. To get their impressions about what OS to install, they'll go to the "experts", meaning web sites like this one and MS' (yes, they will - how many times has/. been touted as The Voice of The Community?). If they both sound the same, said OS consumer is back to square one and will probably choose the safe bet. Presently that's Windows, sorry - more consumer applications, better UI for newbies, and the perception of support.
However, what if the experts sounded like this?
MS: "We're #1! We're the best! Everyone else is doing it, you should too!
/.: "Well, MS may have point A, B, and C, but so do we. Furthermore we have D, E, and F. Do you want to know more?"
MS: "Don't listen to those Linux hackers! We have the blessing of Corporate America on our side! One world, one web, one program! Don't you know that Linux doesn't support new tech like DVD's?"
/.: "That's a problem and we're working on it. Here is a list of all the problems we can think of. If you can solve them, we want to hear from you. If not, there are others working on the problem. However, we've had experience that suggests Windows has problems too. You might be lucky, but..."
MS: "Windows is perfect!"
/.: "Obviously we haven't seen that Windows. Great, show us!"
MS: "Uh, it's not ready yet... but it's perfect!"
Which argument do you find most convincing, the totally biased quasi-ethical self promotion, or the honest consideration of the fully disclosed facts? Stating that we have a clue and are therefore worthy is flattering, but not particularly convincing since both sides are doing it. If you could provide some real and convincing evidence that we have more clue then people might buy it.
MS says that IE4 is required for NT's Y2K compliance. Furthermore, since when have you been able to stroll down to your local shop and purchase a Linux copy of IE? IE isn't tied into NT, but NT (or some other Windows) is tied into IE. Using small words: You can have Windows with no IE, but not IE with no Windows. Hence, browser choice to a great degree influences OS preference. Hence, *since* the browser is tied to the OS (as in, cannot escape the OS), you can't have the one without the other. QED, glad we agree.
Anyway, it has nothing to do with the tying, it has to do with the fact that IE works and Netscape doesn't. Who the hell could serious use Linux as a desktop machine when it doesn't have a working webbrowser?
Precisely my point. The plain fact is, IE doesn't exist on Linux, and it never will. It doesn't exist on any Windows competitors (Macs don't compete on the same hardware). If you like IE (and let's face it, there's not a whole lot of difference, it's mainly personal preference) then hmmm.. which OS should you choose for that Athlon, sparky?
For the record, Netscape seems to be the most unstable program on my linux box. It also crashes most often on my Windows machine, but that could just be because it's always running. I don't use IE4 because while it does seem to render HTML properly, there are so many little edges which remain tied to Windows. MS calls it "integration" and I call it "coupling" - as in that principle they taught us all was BAD in data abstraction. Since I don't want the same type associations for my browser and my file manager, I choose Netscape.
Besides, Netscape seems to "work" on my machines. If you're going to define "work" as "does exactly what I want all the time", we could extend your question a little: Who the hell could seriously use Windows as a desktop machine when it doesn't have a working memory manager, file system, multitasking manager, GUI shell, command line shell, or logging facility? Heck it doesn't even have a working web browser!
But then again, if you're stupid enough to buy something on the basis of its version number, then you're buying the hype and not the product, so you've got what you paid for.
I don't think so. I usually avoid *.0 versions of most commercial software (especially MS "software") because of bugginess. There's usually a *.0.1 or *.1 version released a couple of months later anyhow and by then it's cheaper. Windows 98, however, is very different from 95. It is larger, slower and less robust in areas, and it has IE4 installed.
Besides, remember that people are dumb. Individuals are smart, but people in general usually act dumb. That's why it's such a big deal that IE is a 5.0 browser. I've had people swear up and down that this "proves" MS has left Netscape in the dust. Imagine my fun trying to explain what a "Mozilla milestone" was to them.:)
By default, you should not execute code on your box, if that code came from someone else (e.g. a web site, an email, etc), unless you examine the code first. That's just common sense.
That's right! I never execute any code on my precious Windows machine unless I've examined... the... oh, wait. Never mind.
Slashdot Mirror
Interesting....
Granted, the poster you are replying to is giving an intuitive demonstration rather than a rigorous proof, but you are proposing mathematical nonsense.
Well, I sure as heck am not trying to prove anything, so I suppose that throws us both in that camp. As for nonsense, I think you're dismissing me a little out of hand.
I'm not going to beat around the bush with you here - I think you're misunderstanding because of a notational misrepresentation. What I think we both agree on is that
0.99 ~ 1 (that is, "is approximately equal to")
and that
lim (1-x) = 0
x->1
and even that, in pseudocode,
makeone()
{
real x = 0;
real i;
for(i=0.9;0;i/10)
x += i;
return x;
}
and makeone() ~ 1
But you won't convince me short of a real proof that 1 = 0.999... - I would be delighted for you to show me wrong, but please keep nonsense out of this until you can, Mr. Rinker. I mean, come on - if 0.999... was 1, we could just call it 1, couldn't we? Oh, WAIT, that's true by definition. Odd, that.
Eh!? Infinity + 1? Index on infinity? There is no 1 at that position. The zeros stretch on interminably to the horizon and beyond.
Ah yes, the road goes ever, ever on. I was trying to make a point, to a crowd who's used to dealing with arrays. Since I've seen a few of these comments popping up I'll handle it here:
Infinity + 1 is easy. Let's say x is an integer and y=x+1 and just for illustration let's say z=x. Now imagine the graph of y vs x imposed on z vs x - you'll have two lines (one for y, one for z) and everywhere on the graph, y will be higher than x, for any x. So if we send x out to infinity, y will still be bigger by exactly 1. y=infinity + 1. Although, just *try* getting a address space that size in any language but Haskell!
If you're still having trouble, try this thought exercise: Let's construct our infinitely repeating number like this: write 0.01, then insert 0's just to the right of the decimal forever. You send the 1 off "infinitely" to the right, but it is still there... see, I'm really describing a limit - you know that for all intents and purposes, it's close enough but the hair-splitting truth is that unless you *explicitly* neglect that 1 you cannot say that 0.0000...001 = 0. It is never true, no matter how many (even infinitely many) 0's you put in front of it. All you can say is that it's so similar nobody but Zeno cares.
Hmm...
:)
What does it mean for one number to be equal to another number? Well, if x = y, then x - y = 0. In this case, what's 1 - 0.9999999999999...? The answer is 0.000000000000....(infinite number of 0's). Since 0.000000000.... = 0, they are indeed the same number.
I beg to differ. After that infinite number of 0's is the digit "1" at position infinity+1 (or -infinity-1 depending on how you index) - that is, perhaps a more helpful notation would be 0.00000...001, which != 0. More true would be to say that if x=1 and as y approaches 1, x-y approaches 0 so we can select a y to get x-y arbitrarily close to 0. Hence, as we all know, we can get close enough so that nobody cares any more, but mathematically speaking I'm afraid what you've been telling people has been subject to the underflow error! Ever work for Intel?
Hmm...
Pentium 3 Vs. Athlon - Which Is Right For You?
[ Technology ] Posted by Hemos on Friday April 21, @10:49AM
Time passes...
Athlons Sold Out
[ AMD ] Posted by emmett on Saturday April 22, @03:44PM
Coincidence? Maybe...
how can you take a 'pill' of nanites and, 1st, have them survive the digestive system - it can get really acidic in there...
Easy, you wrap the pill in a bunch of starch and instruct the taker not to suck on it. The starch won't get digested by the enzymes in the stomach, but will be by those in the duodenum (intestine nearest the stomach).
Nonsentient devices are programmed by a human. A nonsentient device can evolve (like a virus) but I see no short term reason why this independant evolution would be faster then viruses evolution. We are not talking the virsus used for gene therapy here. We are talking fundamntal improvments to the evolutionary mechinism of a virus which "good ol' mother nature" has ben working on for many millions of years.
Natural virii reproduce asexually, and hence do not have a lot of genetic diversity. They mutate randomly and thus changes in their genetic code are not linked to the success of the previous version (unlike sexual reproduction, in which the two parents must have survived to maturity). Changes in their genetic code, like in other living things, happen between generations, when they reproduce. This would not need to happen with nanotechnology.
I suppose you might design a nanobot cold virus killer which used group processing to evolved new attacks as the cold virus evolved new defences. This group processing would take the form of units 1 to 1000 try this and tell us if you live or kill viruses. I agree that something like this could have more potential for killing humans, but we are a long way away from something like this. Also, I suspect the communications channel betwen the nanobots would presuppose the ability to include a self destruct.
Now here I see that you agree with me. Yes, we are a long way away. Can you guess if we are more or less than 30 years away from it? I cannot, and I suggest that we should consider the case now, in order to make good decisions in the future. However, don't allow the self-destruct to let you feel secure. After all, the self-destruct is part of the code on the robot, and therefore can be selected against. Which it would be, if survival is a preference we ask the robots to select - and why wouldn't we want our tools to keep working?
I never said we should not predict.. just that Bill Joy dose not know what he's talking about since his predictions are based on sci-fi instead of real theory. We predict the progress of the fiels you listed since we have a theory for them. The summery essentially said that the speakers who understand any present theory of biology and nanotechnology dismissed Bill Joy as a luddite. They were nicer then I was, but that's because they were only concerned with how Bill Joy was wrong.. where I think the things Bill Joy advocates are themselves far more dangerous.
You suggested that since technological predictions rarely pan out after 30 years, why bother? I don't think Bill Joy is a luddite. I don't think the summary pointed that way either. Yes, some of the things Bill suggested seem dangerous, however you're still missing the point that his caution is not one of them. Caution, in these cases, is good. How can you limit the "smartness" of a nanobot? How can you keep the nanobot, or rather the nano-swarm, from evolving around any limits you place on it? These issues should be dealt with, and we do that by increasing research in the area, not decreasing it. This doesn't mean we don't need to deal with the issues - that would be throwing the baby out with the bathwater.
Summery of the summery.. Bil Joy is a moron or an autocrat who wants corperations and big government to rule the future. Joy advocates restricting access to the technologies which will shape the future. This is morally wrong in soooo many way, but I will try to explain a few.
Now, it's rarely a good idea to resort to name-calling unless you have no ground to stand on and must resort to attacking the messenger rather than the message. You actually do have some ground to stand on, so let's get to the point.
I think you may have slightly missed the point. It is a common reaction that if a person can harm you using an area of knowledge, then that knowledge should be kept from that person. If the identity of the particular person is not known, then keep it from everyone!
As you've astutely noticed, this reaction is not productive. It's yet another incarnation of security through obscurity, and it doesn't work. There are fabulous examples in speculative fiction of this concept carried to its logical conclusion - Larry Niven's ARM springs to mind. Someone, somewhere, will come up with the same idea and then you have to either convince that person to keep quiet too, force them to, or allow the cat to leave the bag.
It doesn't matter who develops the knowledge. It doesn't matter how risky the knowledge is. Everyone should have access to it so everyone knows what to do to stop someone from harming someone else with the knowledge.
You seem to have missed what may be a valid point though, because I'm not sure he made it strongly enough - computers have the ability to change the information that fundamentally dictates their behaviour at a rate orders of magnitude higher than humans, and hence can implement "evolutionary change" much faster. It is not impossible that these nonsentient devices could become an "enemy" in and of themselves. No person would be needed to cause harm to another using the machines, they would do it themselves.
It is from this enemy that we must keep the information. It must always be well within our grasp to rein in our machines before they enter the phase of their "evolution" in which they begin to compete for real-world niches. They don't need to be intelligent for this, they just need to require something that they can get at our expense.
Plus, you should consider that no virus or bug has managed to be 100% lethal to humans. Shure, a very rare few wipe out 80% of the population, but we are not talking extinction as a wore case senario here people. Actually, your probabllity of being killed from a metior smashing into this planet is MUCH higher then any risks from bio/nano-technology. (Note: I suppose biology research reduces you chance of death as a result of metiors far far more then it will increase you chance of ding as a result of terrorism)
Also, non-science people really seem to have no understanding of the way these sorts of things progresses. The situation is summed up perfectly by John Holland's quote "Predictions 30 years ahead have always proven to be wrong, except in the cases where there is a strong theory behind it," i.e. without a theory it's just luck (which can take a LONG time).
In the end, your quick dismissal of this rational and intelligent person's reasoned opinion lends the impression that you have jumped to conclusions. The largest point you've missed seems to be that, while non-science (and by this I assume you mean applied sciences such as computing science) people may not seem to you to understand science, by the same token "science people" may not understand applied sciences.
I will not argue historical points about the effect of plagues on humans. However, I will claim that in many cases, the survival of a human attacked by some infectious disease is completely dependent on medical intervention. In order for that intervention to occur, we must have had time to adapt to the disease's method of attacking and surviving. When we can't, people die (see AIDS). If the attacking disease were sufficiently different or mutated fast enough, we would die. If the disease were widespread enough, we would All die, to the man, woman, and child. Imagine if colds were lethal - have you ever had a cold? Do you know anyone who hasn't? And colds need hosts to proliferate, unlike nanobots. If we cannot limit the rate of "evolution", we should not create that agent.
Your assertion that because predictions have failed to pan out seems valid, but the conclusion you draw does not follow from it - that we should not predict is ludicrous. I propose to you some predictions: 30 years from now, computers will be faster, some people will be greedy and some will be lazy. Manufacturing will be more advanced. I think those predictions will pan out. Now, if that continues, at some point we may need to worry about these things. I for one propose that we discuss them now, so in the event that the technology is developed to end the world by a non-human hand, we know what to do and not to do in order to keep that hand from striking.
Why the hell is this marked "Troll"? Signal 11 has some really good points here, that more people should consider, carefully, before they speak to decision makers in their company, community, or family. You know, I can't actually find a single point that he makes that isn't backed up by reliable sources (including my opinion, which is reliable to me :) ).
Where's my moderator status when I need it?
In fact, the only problem I have about breaking up MS is that it does not happen according to their terms. Personally, I'd like to see their Office division left with the hardware division, and OS broken up by product line (take that!) but then, I'm vendictive...
It's a carbon copy of the Hitler Youth program used so successfully in World War II Germany...
I wondered how long it would take for Godwin's Law to take effect in The Katz Papers. Whoever said it was limited to USENET postings?
For those who don't know, Godwin's Law is (according to the Jargon File):
"'As a Usenet discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches one.' There is a tradition in many groups that, once this occurs, that thread is over, and whoever mentioned the Nazis has automatically lost whatever argument was in progress. Godwin's Law thus practically guarantees the existence of an upper bound on thread length in those groups."
The idea is by the time people start bandying about comparisons to the holocaust (and let's face it, are/were there any gas chambers at *your* high school?), any rational discussion has long since been lost in the growing noise of buzzwords and purely emotional appeals.
That being said, it's too bad Katz threads seem to almost immediately degrade into analysis of the Katz article in question, for often (as in this example), he has some interesting things to say.
You are smarter than that.
;)
And there's another for the "Crackpot Theories" department.
Seriously, he's probably just bored and wanted to watch all the knees jerking.
Actually, Photons have a zero mass all the time. What they have is momentum... it's kind of counter-intuitive to someone used to thinking about a Newtonian universe where momentum = mass * velocity.
:)
Turns out that that's only part of the equation, you see. Bottom line, photons have no mass. They do have momentum. If you really want, I'l calculate it for you.
To put it simply: "the source code, the whole windows source code, and nothing but buildable source code"
Nothing? Forget that! I want the source to Solitaire and Minesweeper - without those integrated tools(tm), my Windows experience would be but a pale shadow...
They say they didn't contest it after a while because, put simply, they ran out of money. Kind of like what would happen if you tried to go up agains the MPAA in court by yourself... they'd stall, you'd lose because they have all the cash. There was a /. article about this... Ah yes, here it is.
That's kind of ironic don't you think, considering this is all the fault of the US DMCA? Shouldn't Canadians be complaining about the greedy, corrupt, freedom-stealing Americans?
I'm not, mind you. I understand that the DMCA wasn't something that most Americans wanted. I do kind of resent the fact that Canadians are being held liable in their own country for American laws (and stupid ones, too). Then again, we have a long history of letting ourselves get pushed around from the south. I suppose I should just get used to it... manifest destiny and all.
Actually, that doesn't work because of the inefficient routing that happens all the time. For example, if I want to get to my friend's ADSL-connected machine from mine, just across town here in Vancouver, BC, my packets travel south through Tacoma, WA before coming back up the border. Your method would have me placed somewhere in the USA, but I'm not.
First of all, thanks for your input.
.ca, .to, .tv, whatever. I don't think these domains are hotly contested over like .com, .net, and .org - and these are completely controlled by the US as far as I know. Perhaps there's more leverage there. I do agree that it's to the registrars' advantage to have people registering hundreds of domains for no good reason though.
What would probably happen is that it would just be driven underground more
I thought of this, and perhaps it was naive of me to think that most companies would not feel comfortable offering under-the-table bribes to people for their domains. Even so, a policy could be created (not a law) that forces a domain holder to give up the domain as soon as they offer to do so in exchange for money. A company could then go to ICANN and say that the person attempted to sell them the domain, and presto, no more domain ownership (and perhaps no refund either). This causes a problem with false accusations though, which means that more complicated rules are needed to prevent this from happening. Basically to defend from an accusation like this would mean that you could not get rid of your domain (to do so would be to admit guilt).
Alternatively, it could be made a point of policy not to award domains to entities who paid for their release from scalpers. This would mean that companies could not buy domains from people and the trade would dry up. While fly-by-night domain scalpers don't care about under-the-table deals being hidden from view, don't companies have to make their accounts public to some degree? Here's my naivete again, I have no idea.
These arguments are based on the assumption that companies wouldn't want a back-room deal to smear their public image and that the risk of doing so would keep them from doing truly shady deals for a domain. Of course, having some law passed might make this easier, just like ticket scalping.
I suppose that where they do control who the registrars are, that they could "blacklist" non-compliant registrars.
True, they don't control all the geographical domains like
Well, this method that the ICANN is chosing won't help me at all, because I don't have any trademark rights to anything and if I want to register some domain that these guys own (or will soon own) then I'll have to go buy it (which I will NOT!). Here's a question for you: why don't we just remove the incentive to squat on domains by making it impossible to sell them?
I'm serious, is there any reason besides greed that would motivate someone to register a domain and sell it to someone? I honestly can't think of any. A quick pick:
So why don't we do this? Make domains only brokerable between the registrar and the "owner", with no possibility of profit? I think that would stop domain scalping. IANAL, I have no MBA, and I'd love to hear from people who are more enlightened about this than I am.
UCITA would be pretty bad if passed in the USA. I'm from Canada, and as far as I know there is no UCITA equivalent up here. Theoretically I shouldn't be worried too much, but I am since Canada often follows the US' lead on policies like these. In the meanwhile though, I'm confused by one thing, derived from the fact that companies do not make different software for Canada. We even use the same US-crypto software as the US does. What happens when somebody doesn't like something my company does and turns off my software?
Well, I suppose that my employer would get upset and launch a suit against the vendor if the vendor didn't turn our software back on after we complained nicely... but would they win?
IANAL, so I'm not sure but it strikes me that Canadians and Canadian companies aren't covered by the same legislation, excepting international copyright law. Wouldn't this make UCITA a dangerous tool for vendors to use? Any time software ends up outside the US they'll be liable for damages from their time bombs, etc. Does anyone with more knowledge have a take on this?
This discussion strikes a particular note with me because I believe the Linux UI to be the Wild West of current computing - it's not really formed, and it's no place for the uninitiated. Because of this, I'm pointing my career towards UI development and Human/Computer Interaction. In a couple years I hope to be able to do some really neat work here, and the prospect of being able to work with a large Open community effort on good Linux HCI is really exciting. That said, to the point! :)
it seems like an awefully formidible task to try to be all things to everyone
Hey, everyone, why is this such a big deal? Let's look at this like a programming problem (surprise!), because it maps really really well.
Our problem as I see it: There are features in Linux which are not needed by some people, and as such they tend to "clutter" the features they do need. There is no easy model they can apply to let their real-world trained conventions make this problem easier, so they are stumped.
Allow an analogy: A moderately complex C library will probably have features that say a first-year computing student won't need, and as such they tend to "clutter" the features they do need. One easy answer is to get the 101 student to use C++ and slap the C library behind a wrapper C++ class which hides the features they don't need . Now, here's the kicker: If they want to use those features, they just stop peel back the wrapper and access the library.
What we seem to be talking about here is UI, not OS complexity for truly, a user sees the UI and only kernel hackers and other coders (ie, not users but developers) see the real guts of the OS. So why can't we do the same with UI complexity as we can with that C library? Hide complexity in layer upon layer of black box abstractions.
Now a lot of you are going to say "That's slow!", and you've run into my poor analogy there. I'm not advocating acually using layer upon layer of classes or real programming abstractions, I'm not talking about one UI running over another UI etc etc. I'm talking about layers of perceived complexity. Top is the playskool desktop, with three or four big buttons on it ("Browse Web", "Write Email", "Type Letter", "Go to the Window Manager mode"). The last option lets you use the window manager, and from there you can get to something else like a command shell or better...
At each level you can have different complexities of configuration tools. In Windows, there are two levels: Control Panel, and registry editing / INI file editing. For many people, the first is too simplistic and the second is too unfriendly to dig through. In Linux, there's really one level for most things: Edit config files. This is the "all the features, all the time" problem we had with that C library and the 101 student. Surely we can do better than Windows' 2 layers, and I'd love to see it.
We don't have to make an OS that is everything to everyone, because that's not possible. We do have to make it easy enough for people to configure it themselves so they can customize it to their needs - ideally they won't have to but we can make it easy for them. The key is that *our* OS won't make the next step prohibitively complicated, but rather ease the user into more and more configurable options. It won't tell you where you're going to go today unless you want it to.
Linux is useful. Only some of us don't realize that (Hi, Bill!). Let's make it useable, and put that up as one of our major issues, not just a distraction from more features in grep or a prettier desktop environment (pretty != useable).
Is this the only way? Probably not, but it sounds good to me. Is this doable? Yes, but we'll have to junk a whole bunch of stuff, probably including the X windowing system as we know it today. More importantly, what do you think? Is this the way to go? Would you like to use it? Would you like to work on it, because that's how things happen in Open projects? Do you know about projects that do this kind of thing (please post them so we can all bandwagon and make it a success!) or know someone who does?
But the experienced sysadmins and free/open-source developers are doing precisely what you suggest on Slashdot, and the academic-style Internet old-timers with their well-reasoned logical posts likewise I expect, whatever their diverse experiences.
/. who post their opinions and horror stories and success stories because they know that here, they will be commended for it. Ever notice how stories which are Linux vs. MS always seem to have like three times the number of comments as other articles (and I'm talking about non-grits posts)? How many of those comments are from complete fools? Not a whole lot, or I wouldn't read them. I tend to read posts at 2 or above, because I don't have a lot of time. Most are pro-Linux. Most don't present an objective view. Most are from people who know whereof they speak. Very few I would be proud to show Bruce, my very Pro-MS friend. This is the yardstick I use: could I use this to convince Bruce? Only rarely!
I'm afraid I disagree with you. I think there are many very technically literate people reading
I recently read in an article in the Globe and Mail about a study which showed that the skills to recognize competence were the same as the skills for competence. This could also be said, "Don't pat yourself on the back", and it means that if we are complacent that the problem is with "other people" then nobody will change, even if in your case it happens to be true. Next time you're talking about Linux, check yourself. Show some of the other side of the story. It broke my heart to recommend Win 98 SE to some people I know, but it fit the bill for them and to make them use Linux would be stretching them to fill the Procrustean bed.
Set an example! Ignore the moderators and speak the truth. The whole truth. After all, this is how Open Software works: lots of eyeballs, and all of them looking for problems to fix.
Since the zero-cost base O/S means that Linux sales will never be able to support a marketing budget of any size, it's only right that other forums take over that role, like Slashdot does.
/. been touted as The Voice of The Community?). If they both sound the same, said OS consumer is back to square one and will probably choose the safe bet. Presently that's Windows, sorry - more consumer applications, better UI for newbies, and the perception of support.
/.: "Well, MS may have point A, B, and C, but so do we. Furthermore we have D, E, and F. Do you want to know more?"
/.: "That's a problem and we're working on it. Here is a list of all the problems we can think of. If you can solve them, we want to hear from you. If not, there are others working on the problem. However, we've had experience that suggests Windows has problems too. You might be lucky, but..."
/.: "Obviously we haven't seen that Windows. Great, show us!"
Could be, but if you present a well-balanced, carefully considered argument you're much more likely to win a debate. Mudslinging attacks the integrity of our argument, as can be made obvious by the opinions of people such as the person you're replying to.
And that doesn't happen in pure propaganda forums. What you get here is bias, yes, but it's a clued-up bias.
I would still rather see a more balanced view, because it would allow us to address real issues with Linux so that they stop being issues. To hide flaws behind an argument that "MS obfuscates their OS flaws, therefore we should too" means that we obscure our supposedly free, open sourced OS.
Now, before you drag out your flame throwers, please bear in mind that since most people haven't the technical ability or simply the time to go through and examine the code, or even to install the OS and brave out the weeks of stigmatized clueless newbiehood. To get their impressions about what OS to install, they'll go to the "experts", meaning web sites like this one and MS' (yes, they will - how many times has
However, what if the experts sounded like this?
MS: "We're #1! We're the best! Everyone else is doing it, you should too!
MS: "Don't listen to those Linux hackers! We have the blessing of Corporate America on our side! One world, one web, one program! Don't you know that Linux doesn't support new tech like DVD's?"
MS: "Windows is perfect!"
MS: "Uh, it's not ready yet... but it's perfect!"
Which argument do you find most convincing, the totally biased quasi-ethical self promotion, or the honest consideration of the fully disclosed facts? Stating that we have a clue and are therefore worthy is flattering, but not particularly convincing since both sides are doing it. If you could provide some real and convincing evidence that we have more clue then people might buy it.
IE isn't tied into NT4.
MS says that IE4 is required for NT's Y2K compliance. Furthermore, since when have you been able to stroll down to your local shop and purchase a Linux copy of IE? IE isn't tied into NT, but NT (or some other Windows) is tied into IE. Using small words: You can have Windows with no IE, but not IE with no Windows. Hence, browser choice to a great degree influences OS preference. Hence, *since* the browser is tied to the OS (as in, cannot escape the OS), you can't have the one without the other. QED, glad we agree.
Anyway, it has nothing to do with the tying, it has to do with the fact that IE works and Netscape doesn't. Who the hell could serious use Linux as a desktop machine when it doesn't have a working webbrowser?
Precisely my point. The plain fact is, IE doesn't exist on Linux, and it never will. It doesn't exist on any Windows competitors (Macs don't compete on the same hardware). If you like IE (and let's face it, there's not a whole lot of difference, it's mainly personal preference) then hmmm.. which OS should you choose for that Athlon, sparky?
For the record, Netscape seems to be the most unstable program on my linux box. It also crashes most often on my Windows machine, but that could just be because it's always running. I don't use IE4 because while it does seem to render HTML properly, there are so many little edges which remain tied to Windows. MS calls it "integration" and I call it "coupling" - as in that principle they taught us all was BAD in data abstraction. Since I don't want the same type associations for my browser and my file manager, I choose Netscape.
Besides, Netscape seems to "work" on my machines. If you're going to define "work" as "does exactly what I want all the time", we could extend your question a little: Who the hell could seriously use Windows as a desktop machine when it doesn't have a working memory manager, file system, multitasking manager, GUI shell, command line shell, or logging facility? Heck it doesn't even have a working web browser!
But then again, if you're stupid enough to buy something on the basis of its version number, then you're buying the hype and not the product, so you've got what you paid for.
:)
I don't think so. I usually avoid *.0 versions of most commercial software (especially MS "software") because of bugginess. There's usually a *.0.1 or *.1 version released a couple of months later anyhow and by then it's cheaper. Windows 98, however, is very different from 95. It is larger, slower and less robust in areas, and it has IE4 installed.
Besides, remember that people are dumb. Individuals are smart, but people in general usually act dumb. That's why it's such a big deal that IE is a 5.0 browser. I've had people swear up and down that this "proves" MS has left Netscape in the dust. Imagine my fun trying to explain what a "Mozilla milestone" was to them.
I can see the MS release now: "See, having the browser tied to the operating has nothing to do with consumer OS choice...". Sheesh.
By default, you should not execute code on your box, if that code came from someone else (e.g. a web site, an email, etc), unless you examine the code first. That's just common sense.
That's right! I never execute any code on my precious Windows machine unless I've examined... the... oh, wait. Never mind.