Sigh. So you wanna talk GPL. Fine, you seem like a reasonable person, and I could use a little feedback. Let's do that.
CDDL did not exist at the time, so it was not even a choice. Supposing it did, it is an open source license rather than a free software license, so comparing choice of CDDL vs GPL is a philosophical choice rather than an 'implementation' choice. So to answer your question, no we should not.
I'm afraid I'm not following you. You seem to be saying that we shouldn't blame Linus for choosing a license that later became incompatible with the CDDL, but we should blame him for choosing a license that later became incompatible with the GPLv3. So I guess the point is we should blame Linus only for not preemptively aligning himself with the FSF's philosophy, without reservations.
If that's what you're saying, and I'm honestly not sure, then I'm afraid we'll have to agree to disagree. I think it's become painfully obvious that the Linux "philosophy", to the extent there is one, and the FSF's, are two distinct points of view, not entirely compatible (see for instance Linux exception for proprietary modules).
Maybe it's that GPLv2 happens to be an adequate common ground what seems to have made people like you believe, mistakenly, that you shared the Linux point of view. Just like it made people like me believe, also mistakenly, that I shared the FSF's point of view. For my part, I now realize that this is not the case. I also believe my point of view is the better one, of course.
But never mind that. Linus' choices were and are consistent with my point of view, which I believe is shared by the Linux developers. GPLv3 is not.
Well would you reasonably expect GPLv2 to be the last word on free software licenses? Would you expect people to never start using later version of GPL? You can in essence say 'well we want to have ultimate control over our code', which is what I said in my previous post, but not having an upgrade path is just poor planning.
You know what? You're absolutely right. Of course, I still think your preferred upgrade path (following the FSF's agenda) is completely unacceptable. But maybe I should have another look at that MPL.
I find this ironic because if you judge Linux by its rough drafts (early versions) then a lot of the code is simply a joke. It certainly did not come out of the ether fully formed. Many would have called the process used to develop it all manner of derogatory words, but what matters is what the results are.
Agreed. Can't see the irony, though. You mean that "early drafts" thing? I say that not because of the draft's rough quality, but because they made me aware of the direction the FSF intended to take, the things they consider important. As I said, I mistakenly believed that I shared their convictions. That was a rude awakening of sorts. And yes, it was entirely my mistake. Like, for instance, somehow I misunderstood or disregarded Stallman's writings. My bad, won't happen again.
And GPLv3 looks to be much better than GPLv2 in a lot of ways; the process works.
Disagree. GPLv3 goes too far in its attempt to "protect the rights" of everyone. If someone wants to take my code and build a machine that doesn't boot modified software images, say a phone or some specialized portable device, as long as they release the source code I'm cool with that, but GPLv3 forbids it. If they want to write a client for SecondLife or some such, where the software bars the user from copying graphics or models or whatever, then they can't use my code either, because then they couldn't pursue patches disabling that restriction as "circunvention devices" under WIPO. Or so I think. Tell me if I'm wrong, because section 3 of the current draft is barely comprehensible.
(And yes, that last one would be a very stupid thing, but the point is that as long as the code
So the conspiracy theory regarding that bit of your argument sound implausible frankly.
Yes, I know. And to think I just complained (elsewhere) about the current paranoid climate in open source:-S
Also, I'm complaining about something Sun is giving to some people, just because they seem to have deliberately chosen to exclude Linux. Complaining about people not giving you something you want is lame, I'm fully aware of that.
Still, I'm not trying to belittle or scorn Sun. I acknowledge Sun's contributions and I'm duly grateful for their generosity. I'm arguing because this dude with the hex handle said Linux is being "harmed" because the license its authors prefer does not conform to what Sun (or the FSF) would like, and I wanted to disagree on the grounds that Sun motives may not be in Linux best interests after all. I also have strong opinions concerning his advocacy of the GPLv3, but I really don't want to get into that.
Anyway, my point is, don't get too infatuated with Sun. Sun is a corporation, and one notoriously fickle at that. Schwartz primary responsibility is to Sun's shareholders, not the public in general and certainly not any open source community. This is not wrong or unfair or anything, it's just the way it is. They are not threatening patent litigation, thankfully, but that's now. What about next year? Five years from now? Patents last for 20 years, you know.
Consider that Caldera was once a Linux company. It doesn't get more Linux-friendly than that. If you'd tell me back then that the swines would backstab us all, I probably would have laughed at you, and now I'd be wondering how could I ever be so naïve. Being trustful and optimistic is a great way to go through life and still come out sane, but I think people shoult at least try to keep their feet on the ground.
Thank you for your answer. I think we are in the same page, as to what will happen if Sun releases ZFS under GPLv3. Minor nitpick: when I said Sun would be "forfeiting" the patents, I was just thinking about Sun losing that advantage over Linux, their main competitor. You know, what you referred to as the "main differentiator" (which btw I think it's a bit of an overstatement of the benefits of ZFS, but that's beside the point).
GPLv3 actually helps Sun to have their codebase as the defacto standard, because any implementation would have to be based on Sun's code.
Right. But that's no different than the CDDL. Right now, to use ZFS you have to be able to glue a chunk of CDDL code into your project. Linux can't use it because of the GPL (any version). Shouldn't we, by your reasoning, say that that's Linus' fault for choosing the GPL instead of something else?
Anyway, my point is, it was Sun who chose to license it in a way that Linux couldn't touch it. And if they release it under GPLv3 (big if, IMO), that will also be Sun's choice. Now, you may take the charitative view and say that they had to do it that way, or that being incompatible with Linux is an unfortunate, unintended consequence. Me? Well... I guess I've just become too cynical in my old age.
Linus is now preemptively blaming Sun for releasing ZFS under GPLv3, if they do, which would mean that Linux can't use it. What I am saying is that, like the bitkeeper fiasco, choosing to use *only* GPLv2 now looks like a big short-sighted mistake. And that this was Linus & Co's decision not Sun's.
Well, that's your perception. The short-sighted mistake, I mean. To me, it was a sensible choice. If you care about how people use your code, then you don't leave a backdoor for third parties to relicense your code as they see fit. Not even the FSF.
In fact, and this will sound trollish, and I do apologise for that, but after reading the GPLv3, particularly the early drafts, I have to qualify: especially not the FSF.
Oh and btw, I think BK was also a sensible choice at the time. And the offspring of that "fiasco", namely git, more than compensates for everything. IMO of course.
Is anybody seriously saying Sun should release their code under the soon to be out-of-date GPLv2 simply because Linus 'likes' GPLv2 better? That's pretty absurd, and I assume you would agree.
I'm really sorry for being difficult, but no, I would not agree. C'mon, I keep reading GPLv3 advocates saying as much: GPLv2 is not going anywhere, you are as free to use it as you are to choose GPLv3. Now, regardless of how things got the way they are, or whose "fault" was it, Linux is GPLv2, and it doesn't seem likely that that will change anytime soon. Sun can choose GPLv2 if they want. In fact, I think there's a good chance they'll do precisely that, just so they can use the drivers.
If Sun actually does release ZFS under GPLv3 Linux developers and users won't be enjoying that crow because Linux will still not have it in the kernel. It'll have a FUSE only slow version or have a legitimate patent cloud hanging over it. The existing FUSE ZFS would also have a patent cloud over it until they base it on Sun's implementation and thus release it under GPLv3 license.
I just read notamisfit's reply to my post. He's right, BSD is pretty much free to use it under the CDDL, as is that guy doing the FUSE port. My mistake.
But you're right too, if Sun does that, Linux will not benefit. In fact, unless Sun allows GPLv2, Linux is pretty much out of luck, and the patents ensure that it will stay out of luck. Another unintended consequence, perhaps.
Finally, you poo-poo software patents. I don't actually have a problem with software patents, what I have a problem with is bogus ones (one click, FAT32, etc) and patent trolls (patents owned by
Apparently ZFS is so much better that Linux must adapt in whatever way practical in order to get it.
It would be highly desirable to have it on Linux, yes. If Sun were to license ZFS under GPLv3, then that would be a good reason for considering moving Linux to GPLv3. In fact, this may be the very first actual justification for such a huge undertaking. But that doesn't mean Linux must get it that way. Switching to a compatible license may not be possible at all, and even if it were, whether it would be desirable is questionable. There are always alternatives.
ZFS is looking to be the defacto standard for filesystems for the next decade and even if Linus tries to change the kernel to GPLv3 it will still be years behind Solaris on this.
Yeah, about two years behing would be my WAG. I'm thinking along the lines that rewriting the whole thing would be easier than changing the license of the kernel, and that it took Sun some 6 years to implement ZFS but cloning something is much easier than writing it the first time. But see below.
ZFS is already in Solaris, Mac OS X (rsn), and FreeBSD.
And Linux, via FUSE. It's slow, but getting better.
What we see here is Linux already starting to be held back by requiring what will be an older out-of-date license and Linus already starting to blame others for it, when with his stewardship it is his choice that caused it. Linux can't use a GPLv3 ZFS because Linus is against using the GPLv3 license. Linus removed the "or later version" clause because he wanted more control over the kernel.
Whoa dude, hold on a minute. Three points here. First, the kernel devs (not only Linus) chose GPLv2 because they liked it. They don't seem to like GPLv3, so removing the "later version" language was evidently a prescient and wise move.
Second, whether GPLv3 is a better license for Linux is highly debatable. I happen to dislike it for somewhat amorphous ethical and aesthetic reasons... but doesn't matter, that's irrelevant. There are practical concerns: embedded Linux is huge, for instance, and GPLv3 may be poison to that. ZFS alone is probably not worth the risk of losing that "market share" (God, I hate that term).
Third, and most important, you seem to be granting as fact that Sun will license ZFS under GPL. I think that's unlikely. Linus makes some reasonable points in that article, and here's another: Sun itself has got some 50 patents around ZFS. Licensing under GPLv3 would amount to forfeiting these for all practical purposes. In fact, licensing under GPLv2 will probably do that as well.
Now, if Sun actually steps ahead and does it, I will gladly take back all this, apologise, and run to the kitchen to get some Tabasco for my nice plate of crow. And I will enjoy chewing on it, because it would be great to have ZFS. But, like Linus, I won't hold my breath waiting for that to happen.
And while we're talking patents, I don't think you should feel that safe using ZFS on BSD. The fact that there's an implementation available does not release you from Sun's patents, which are still hanging over your head, just the same as if you choose FUSE ZFS. As long as those patents are there, it may be in everyone's best interests (meaning BSD and Linux people) to start working in a free, non-encumbered, superior alternative. I mean, merely cloning ZFS may be not enough.
(*sigh* what a waste of resources... fuckin' evil software patents)
Oh, and Apple may have a patent license from Sun, yes. But somehow I don't think you're advocating switching to a proprietary OS just to get ZFS.
Funny, I've been considering switching from Emacs, too, but for different reasons. In my case, I'm worried by the GPLv3 and all that jazz. I guess I'm having second thoughts about the FSF---all that "forcing evil tivoizers" to "preserve people's freedoms" is starting to rub me the wrong way.
Or maybe I'm just getting paranoid and peculiar in my old age. I hope that's it.
But anyway, politics aside, I've been thinking, hypothetically, how hard would it be to go without GNU. Not only switching to BSD or OSX or something, but actual life without GNU at all. Worst case scenario. Oh and, as you say, I've not switched yet, I don't think things are that bad, and anyway I probably couldn't switch anytime soon, even if I really wanted to. So please don't kill me either.
Right now, there are two pieces of GNU software I absolutely can't live without: Emacs and GCC. To a lesser extent, Bash, GDB and GNU Make. Everything else is negotiable, I think.
Currently GCC can't be replaced, of course, no need to waste time on that one. But Emacs should be replaceable. My problem is nxml. Just thinking of giving up nxml gives me the creeps.
So I guess this rant is to ask, how do people edit XML in vim, or something else? Is there an alternative at all? Would I need a dedicated XML editor instead, besides the one I'd use to write code and plain text?
Oh for the love of root. Nobody is about to be sued by Microsoft. That's just a bluff, a good old fashioned piece of FUD, nothing more.
But hey, even if someone were, what makes you so sure Linus could do something about it? You're probably thinking about him switching Linux to GPLv3, but I have to say, even if that were feasible, I'm not convinced at all that it would be a change for the better. In fact, if the new GPL is designed to do this sort of thing, trying to bind and control people who don't even distribute, then it's starting to look quite evil to me.
Now I'm not a lawyer, and I certainly wouldn't dream of outsmarting Moglen at his game. So I won't even guess if this is actually proper and legal and not an abuse of copyright. That's how it feels, though, and I want no part of that. I'm glad Linus has rejected it.
Btw, that BK "fiasco" was a lot noise, indeed, but look at the bright side: nobody got sued, nobody even acted unfairly (not even Larry McVoy, who was entirely within his right to be an asshole). And in fact the whole episode gave us git, a wonderful piece of software I use everyday now. Not much of a tragedy if you look at it this way, no?
Dude... I'm hoping there's some sarcasm there that I missed. You can't be serious. C'mon, you can't even take Iraq!
No, really. Taking on China is out of the question. That would be Dr. Strangelove-grade insanity.
Besides, I'm pretty sure nobody who matters wants confrontation. The Chinese could screw your economy quite thoroughly, yes, but I don't think they'd want to lose their biggest customer---not to mention seeing the 400 billion they hold in US treasury securities become toilet paper. And American corporations (who seem to run your country anyway) would hate to lose dirt-cheap, profit-maximizing Chinese manufacturing.
So I don't see any sort of confrontation any time soon (thankfully). What I see is American and Chinese billionaries getting even richer, while the general standard of living in China raises marginally, and in America goes to the dogs.
But what do I know, really.
Oh, and as for the pollution thing, that's nasty, indeed, shame on the Chinese. But they're not the world's most polluting country---that would be the assholes who emit 24% of the world's greenhouse gases, and yet refuse to sign up to Kyoto.
Yeah, well, it's kinda sad if a constructive user with some coding skills and no patience for bureaucrats must resort to something like forking, in order to get a useful feature in the GNOME code base. Come to think of it, wasn't something like that attitude that killed XFree86?
But hey, in this case, with Linus being a sacred cow and all, his changes will most likely be taken in, and I'll see that small improvement in a future version. Maybe I'll even find it useful, since I also use the focus-follow-mouse mode, and tend not to maximize windows. Still, a bit worrying if you have to be a sacred cow to get someone to listen at all, at GNOME Towers.
That said, I don't think it's that bad with GNOME, really. As I said, I think most of the screaming is being done neither by Linus nor most of the GNOME coders; just a few people hypersensible to harsh truths and direct language.
Geez. Did you read the thread in that mailing list? I did. Linus is right. Not only right, what he actually did is much more than what I probably would have done in his place (I probably would have sent just the "fuck you", sans patch).
Now I use GNOME, which to me is much prettier than KDE, has a nice shell where I spend most of the time, and lets me run Emacs. So I do appreciate the work the GNOME guys do and all. And never even noticed that right-clicking on the border of windows brings up a useless menu, and that there's no way to change this behavior. So Linus noticed, and it seems he talked to some maintainer and the guys in that list and eventually got fed up. Why, I don't exactly know. In fact, it does surprise me a bit, since I know Linus is a no-nonsense kind of guy, and kinda smart too, and the people who coded this beautiful user interface just can't be morons.
So I think that this brouhaha was caused by, sorry, people like you, getting all worked up in the name of "the norms of human interaction in a public forum" or (more likely in that case) the "norms for talking sweetly to the commitee so a change they didn't thought of is even considered". I'm sorry, but you sound exasperating. *Fuck* the norms of human interaction yadda-yadda, the change is an improvement, it does not affect usability at all (as I said, I didn't even noticed before), and if someone told Linus that they were not going to do it because GNOME is better off without it, then that someone was being a bureaucratic prick and deserved a sound fuck-you.
And neither do we have any fossil record of the millions of animals that would had to have appeared in iterations between one species and the next they supposedly evolved to.
You do realize how extremely unlikely is that an organism fossilizes after it dies, right? Most of them just decompose and dissolve back into the environment. It is a freak occurrence for a fossil to form: the organism has to be covered with sediment quickly after it dies. Think about it. Then ask yourself if it is really surprising that there are gaps in the fossil record.
Nevertheless, the continuous evolution of life has been going on for so long, and produced so many variations, that transitional fossils are actually quite common. Google for them.
What we do have is "Lucy", the supposed half-ape/half-woman, whose pelvic bone was sawn apart by its discoverers and glued back together differently so it would look like it belonged to a biped. We have dinosaur models in museums with feathers glued on them. We have highly-educated scientists saying "Well gee, we don't know how genetic code can be modified to include extra instructions so animals can evolve from one species to another, but we're assuming it did somehow, at some time." Yeah, that's some really convincing "science" you got there.
I am not familiar with neither of the "sawed bone" or "glued feathers" anecdotes. Any references to those? In any case, "Lucy" is not the only Australopithecus fossil that has been found. And the dinosaur model with feathers glued onto it... well, I am sorry, but I just can not see the relevance of such a thing existing, if it even does.
As for the genetic code, I do not know who those scientists you are talking about are. But anyway, what is surprising to me is that mutations are so rare. I mean, I have seen bytes changed in tapes and disks, for no apparent reason (exposure to some magnetic field, no doubt). Intuitively I would think the DNA mechanism would be even flakier, what with all the million ways it could go wrong, from simple physical trauma to cosmic rays and so. It obviously is not. Intuition does have its limits.
You all talk about evolution like it's fact, so show me proof that backs your assertions up. Show me how carbon-14 dating, known to be unreliable for anything over 60K years old, can be used to prove that the earth is billions of years old. Show me the fossils of species in transition. Dig as deep into the ground as you want, dig up anything you find, and prove to me that it's more than about 6K years old.
Evolution is most likely a fact. Regardless of the available evidence, the model works: I have seen it and I have used it in practice. I have made LISP lambda expressions that evolve from gibberish to actual useful functions, using only random mutations and fitness feedback. In fact, I'm working on a pet project which has, at its heart, a genetic algorithm that relies on evolution. And it most certainly works.
Oh, and carbon-14 is a short-range dating technique, as you are plainly aware of. There are longer range radiometric dating techniques available, of course, from Uranium-Thorium to Potassium-Argon and Uranium-Lead. It is not my duty to "show you how" anything is done, but if you are genuinely curious, just search for them, just as for the transitional fossils. Detailed information is abundant.
Fact is, evolution is junk science, protected by an agenda. Creationism is unpopular with you because the idea includes the activities of an intelligent designer who created the earth and everything that appeared in it initially. If such a being exists, that being is God, and it would follow that we, as part of his creation, would all be accountable to him. Some of us are okay with that. The rest of you aren't, and that's okay with God, since he created you with a free will. Those of us who believe in our Friend In the Sky are also okay with the fact
MS should simply call their bluff and cut Europe off completely from MS products for a few months.
I can't believe you are serious about this. The notion of Microsoft pulling out of Europe is just too bizarre, even as a thought experiment. I mean, thought experiments are not as fun when they are that disconnected from reality.
I don't think Microsoft can just "cut off" Europe in the way that's been suggested. Not unless they are prepared to refund every company and individual who would be left unable to use the operating system they paid for. I don't know how many copies of Windows could be "shut down by autoupdate", but I guess the amount Microsoft would have to refund would be quite large. Good luck convincing your shareholders that diverting a big chunk of the company's cash to refund people (or to pay for litigation) is preferrable to having to document some protocols.
A little less outlandish, though not even close to realistic, would be for Microsoft to refrain from selling any more products in Europe, and convincing the court to let them skip complying with the previous order on that grounds. But I don't think that would be a very smart choice by Microsoft either. It would cause them to lose a huge market, and all that, but there's something even more important...
EU businesses would scream bloody murder and governments across the EU would be in chaos.
Microsoft has had great success imbuing in the minds of people this weird feeling of dependence of Windows. That prediction of yours is a prime example of this. Even in the face of steep prices, defective products, vulnerabilities and all, people just think they need Microsoft to use their computers---it's almost like an addiction.
Now this really puzzles Mac and Linux users, because they know for a fact that there are perfectly serviceable replacements for most of Microsoft products. Not only serviceable, but in many cases technically superior, cheaper, or both. But people at large shun those and go along with Microsoft, out of fud, herd mentality, or whatever---I don't claim to know how that happens. But I do know Microsoft would not even consider forcing people out of their addiction to its products. Force Europe to switch to Linux or whatever, and that many people will find out that there really is no need to pay Microsoft for anything. I submit that would be much worse for Microsoft than losing the European market. In fact, I think that would be the end of Microsoft's dominance worldwide.
I suspect, within a month or two, Neelie Kroes would die in a surprisingly convenient car accident, which British MI5 would call "A terribly tragic thing...Yes, very tragic, that." and the EU would back down.
That's just disgusting.
Anyway, back here in reality, I think Microsoft will "cave in" and document the stupid protocols. Really, they have no choice. It will be harder for them to keep their monopoly, having fewer ways to lock-in people and all, but at least it won't cost them their entire business. Not immediately, at least.
How much money has the DNC spent calling people? Why is this not quoted?
I don't know, but if I had to guess, I'd say probably not much. I mean, if that particular datum were newsworthy, it'd be all over the news by now.
But... why is that particular question relevant? I mean, if the Democrats were doing the same, does it somehow justify the Republicans? Would it make it right, or at least not as wrong?
Anyway. I don't really care for, or know much about, your political parties. I'm just another foreigner watching the process from home, concerned about the possibility of your neocons keeping their power. That'd be just sad.
I hope it's like you say, because in the media it came more like this:...
Funny. However, I think this is probably just delay in the media picking up the Spamhaus story. For instance, the last three stories about this on Slashdot concern with reactions and commentary, by different users and reporters, on a single Spamhaus press release. Both Steve Linford and Richard Cox have been asked to comment, and to the best of my knowledge, they've just repeated what can already be found in Spamhaus' answer to the proposed order:
Spamhaus.org's Spam Advisory List, the Spamhaus Block List, blocks 50 Billion spams per day across the Internet, therefore the effect of suddenly not blocking such a large amount of spam would mean that volume of unwanted junk hitting mail server queues all over the world. That in itself has a technical effect we can not properly estimate but would certainly cause very serious problems in most countries.
Technical issues aside, the vast majority of those 50 Billion spams are highly illegal, spam for drugs, extreme pornography, scams and bank phishes. The effect of 650 million email boxes (Spamhaus' userbase) suddenly receiving such a barrage of illegal spam, scams and bank phishes is, in my opinion, extremely dangerous. For this reason alone we believe that ICANN suspending spamhaus.org is almost certainly a no-starter.
What I'm saying is, I don't really think it is Spamhaus whining over and over about impending doom. They said their piece. Now it's the media slowly picking it up, with every new article coming through as if Spamhaus made yet another dramatic warning about the fate of a world without them.
But then again, what do I know? I just can't understand how some people can take a stand against Spamhaus. IMO they are the best asset the Internet has right now for keeping spam at bay; I know my job would be considerably harder without them, and my users would waste a lot more time digging through the junk. So Spamhaus gave the finger to some judge, and that was not the smartest thing to do. Fine. They probably should apologize to that judge or something. It doesn't change the fact that the judge should not have any authority over them. And I'd also say we should be helping them, not adding to their worries right now.
I think you should wait before changing anything. I don't think spamhaus.org.uk, or any other name besides spamhaus.org, will ever resolve the Spamhaus RBLs.
From Spamhaus' response to the proposed order (proposed, people, by the spammer's counsel, no judge has ordered ICANN anything), it seems they'll intend to contest this. They mention they don't think that ICANN suspending them can actually happen, for reasons I in fact agree with (go read them at their site). They also mention that "one U.S. government agency has begun working on a response."
However, if worse comes to worst, they probably won't switch to any other domain name. They state: "... if Spamhaus gets around the court order by switching domain to maintain the blocking, the judge would very likely then rule us in criminal contempt. We don't want a criminal record for the sake of fighting spam. We normally help fit the spammers with criminal records, not the other way round."
Which I read as, if this order is enforced, and ICANN caves in and all that, there will be no more Spamhaus, period.
Which would really piss me off. The whole episode already already seems like a bad dream to me. To see Spamhaus destroyed by some spammer scum would be just depressing. One thing's for certain, though: it'll be a cold day in hell before any site I manage will exchange traffic with this spammer.
Worse than the problem? So, what you're saying is that if I disable the Spamhaus check in my servers, me and my users will be better off?
Let's see. As things are, currently, my MTA rejects any IP address flagged by either the SBL ("direct UBE sources, verified spam services and ROKSO spammers") or the XBL ("illegal 3rd party exploits, including proxies, worms and trojan exploits"). This happens right before the first SMTP command is accepted, which saves me bandwidth and CPU time I'd otherwise use accepting the headers and scanning them, and, if the message passes that, accepting the body and running clamav and bogofilter on it.
So yes, I "blindly" reject clients listed by Spamhaus. I'm perfectly aware that it is possible I could be blocking some innocent client somewhere. What I'd expect in such a case would be some friend or contact or user would complain that her mail bounced, at which point I'd probably whitelist her and be done with it. Of course, if such problems were common, I'd probably stop using Spamhaus, as their criteria for blocking would clearly be too different than mine and my users'. But I've never had a complaint, so I guess false positives are rare, or my users are really shy (yeah, right).
Anyway. If I don't rely on Spamhaus, or any other RBL because, as you say, they are "worse than the problem", then I have to accept all mail and scan it. I think I usually get over 95% accuracy with bogofilter, but keeping it that well trained is time consuming. And of course, I can't block mail with it, just tag it (a bayesian filter trained with the spam *I* receive is not necessarily accurate for my users), so everyone has to setup rules to actually use the filter rating. If I know my users, they won't do that, they'll rely on the "[SPAM]" added to the subject to filter manually. And 5% of a lot of mail a lot anyway. As for malware, it gets worse: messages are usually very large, and new malware usually passes through clamav for some short time before its db is updated. And I still have users on Windows, regrettably.
If I do rely on Spamhaus, I save bandwidth, processing time and of course human time dealing with false negatives, but I have to deal with the risk of some legitimate message bouncing off my servers because of a Spamhaus mistake. Which to my knowledge has never happened. Which would be trivial to correct anyway.
Actually, if the death ray is not a myth at all, then it is likely that it was not constructed the way Mythbusters and the MIT guys did. This is from the "Book of Histories" (Chiliades), by John Tzetzes (circa twelfth century AD):
When Marcellus withdrew them [his ships] a bow-shot, the old man [Archimedes] constructed a kind of hexagonal mirror, and at an interval proportionate to the size of the mirror he set similar small mirrors with four edges, moved by links and by a form of hinge, and made it the centre of the sun's beams--its noon-tide beam, whether in summer or in mid-winter. Afterwards, when the beams were reflected in the mirror, a fearful kindling of fire was raised in the ships, and at the distance of a bow-shot he turned them into ashes. In this way did the old man prevail over Marcellus with his weapons.
As much as I can picture such a device, I think what Archimedes could have done is set up an array of flat mirrors along with a mechanism to move them following the sun while keeping the light concentrated in a single point, like a solar furnace. Then he stood at the focus point with a concave mirror (maybe faceted, hence the account of it as "hexagonal") to bounce the light into parallel beams. A death ray, indeed. To hit a target, he only had to point at it with this single mirror.
I don't know how viable this device is, but it doesn't sound incredible to me. I guess there is some caveat I don't see right now; otherwise, I'd be very surprised that these reconstructions of the event didn't use this design.
Sorry, Manos: The Hand of Fate does top them all. Even mistied, it's painful. Un-touched, it'll make your eyeballs bleed.
Not that I disagree with you last remark, but I side with grandparent here: Red Zone Cuba is in fact worse than Manos, unbelieveable as that may be.
Manos may have been made by a fertilizer salesman (which shows, plenty), may have been shot out of focus in its entirety, may have the most ridiculous set of characters ever, but at least had some resemblance of a plot. Granted, a dumb plot, an annoying plot, but it's there. Plus, that lady was very good looking, IMO.
RZC, on the other hand, is technically as bad as Manos, and also completely incoherent. It is just an idiotic sequence of unrelated scenes, ranging from dull to unpleasant, up to frankly insulting.
Besides, RZC actually stars Coleman Francis, the man, the legend. With Anthony Cardoza to boot. That alone should be dreadful enough, but this is worse than the sum of the parts. For instance, RZC is much, much worse than Skydivers (my opinion again, of course---the IMDb for some weird reason ranks it better!).
I think this guitar playing thing is pretty cool, actually, even if it's music is indeed much worse than what a human player can do. The fact that it was built by some kids in a couple of months, and the fact that it actually works, are quite impressive by themselves.
First all they do is rehash the old crap and now we're replacing them with robots? Next they'll find a cheaper way to replace admins and you'll all be out of a job!
That's the idea, yes. And it probably will happen.
In the years I've been working with computers, I've seen many attempts to replace sysadmins, and even programmers, with software. To this day I know of no technology that can actually do that, but I'm not counting on this kind of job security. In fact, I'm almost sure that some day a computer will be able to do my job, probably better than I could ever do it. And my job demands quite a bit of creativity, mind you, I'm not a VB-coding monkey (no offense to any VB-coding monkeys around).
In fact, I think I'll see this happen---sometimes I think this will be the reason I'll retire. Which makes me a bit sad, because, hanging out my wife and daughter aside, there's nothing I like more than programming. But I've learned to accept it.
I was a bit surprised to see just how defensive this article got you musicians. I mean no disrespect to you, I know for a fact that a human artist can imbue in music subtleties far beyond my capability of perception (I'm a programmer, not a musician, even if I can play some things on a guitar). But I think you should get used to the idea of machines that can actually do precisely that, that can be as creative as humans. Which is the only reason synthetic music sounds "wrong" today---pure technique is something that a computer can match and surpass humans quite easily.
Creativity so far is a domain exclusive to humans, but I see no reason why this must be so, besides the fact that no computer has been able to emulate human thinking with sufficient accuracy. This, however, will probably change, and I think it will change rather soon.
I am pissed off that everytime something innovative comes out, someone has to be a token Slashdot user and say 'No Linux Version' or 'No *nix? This sucks' or whatever other equivalent you can pull out of Slashdot's archives.
I agree, to some extent. It always annoys me when people complain about free software, as if they had some right to it. It is a gift. Complaining about something that was given to you is pretty lame, indeed.
However, the lack of a GNU/Linux version is very frustrating, so I also understand the feeling. See, I do not have a Windows machine. I do have valid reasons (at least valid and important to me) for not having one, so I do not even intend to get one in the foreseeable future. So I will not be able to run this software, when otherwise I would be all over it.
This is my choice, of course, so I am not complaining. It is just very frustrating.
Even if there isn't a linux port, you shouldn't be complaining because someone has to take the time to code all that - are you willing to go put in the time and do that work?
I am certain that many people here would gladly "take the time to code all that", for free even, if they only could. But this is not an open source app, so that is not possible.
Come to think about it, this could make a great project. GPL'ed clone of Google Earth, anyone?
It is tiresome but it's still true. You might not notice if you have a newer machine, but for me on my 800mhz duron java apps are still unusably slow.
I think that sounds like a memory problem, not so much the CPU speed. Perhaps your machine does not have enough physical RAM to hold the huge virtual machine allocation and still cache all the shared libraries required by it. In that case, every startup of the virtual machine will be dreadfully slow, and swapping will only make things worse.
But Java itself is not slow, if you give it enough RAM. That's a misconception. A reasonable one, I will add, particularly if you used the early versions and then dropped it, like I did. But if you'll forgive a little rant, here's an anecdote.
A month ago or so I wrote a general C implementation of the a-star algorithm (AI stuff, commonly used to compute routes on a graph), and tested on some fairly demanding problems, both in memory and CPU cycles. I thought it was rather nice, e.g., solved most solvable combinations of the "eight-slides" puzzle in less than a second, and exhausted the search space for unsolvable combinations in 10 seconds or so.
Then, just for the hell of it, I wrote a quick Java implementation. Would you believe it was nearly 10 times faster? I couldn't, at first. But it really was.
I went back and actually profiled the C version, and found that my data structures were hideously slow. Long story short, I replaced some arrays with red-black trees, and in the end, yes, the C version ran faster than Java. But not much faster. Just some 10-20% faster. And it was much harder to get it right.
Now, this is not to be construed as an endorsement of Java. I don't like the stupid language, and Mono/.Net is much better IMO, as far as virtual machines go---at least you can choose the language. And I don't use programming tools where "Hello World" requires megabytes of memory to even start, just on general principle.
But my point is that, provided you have lots and lots of RAM, recent Java is fast. I couldn't believe exactly how fast until it hit me in the head.
To enable inline image display in w3m, you have to either compile it yourself, passing --enable-image to configure, or, if there is one for your distro, install the appropriate package, on debian you need w3m-img in addition to w3m.
Oh, no, sorry, I think I did not explain myself. I do have inline graphics, as you can see in the screenshot (btw, just apt-getted the standard Debian package, I did not build it myself). What I do not know is how w3m manages to display graphics at all, in a terminal window.
Fwiw, I did install a Netscape Directory Server on a HP-UX 11 machine, not that long ago. It was reasonably straightforward, except in that I had to install a number of OS patches and muck around with kernel parameters.
(Btw, what is it with these big proprietary apps that always want to change your kernel parameters? What on earth does Oracle need 2GB of shared memory for? And 64K file descriptors per process? That's beyond ridiculous. That sounds dangerously like extremely sloppy programming inside the product.)
But I digress. My point is that installing and configuring NDS is not hard, but nothing like "soo but soo easy" either (e.g., a far, far cry from "apt-get install slapd").
Enabling SSL is a PITA if you don't have the Netscape Certificate Server (which I didn't). I involves all manner of funky maneuvering with OpenSSL and some tools that you have to fetch from some obscure page at mozilla.org.
Management is more or less the same than with OpenLDAP, which is to say that it mostly depends on how good or bad are your LDAP client tools. In fairness, I hear the Netscape client is nice. I couldn't use it because the damn thing runs on Windows and I was not about to install that in my laptop just to see a stupid LDAP client.
Replication is probably better than OpenLDAP, though I haven't yet a chance to try it on either one.
As for big environments with many users and clients, until today I would have gone with OpenLDAP (or, if a PHB just had to see a lot of money spent in this, with Novell or Microsoft's directories). That's because nobody had source code to NDS and it was all but discontinued from the vendor. You don't want to find yourself in a position where you know there's a bug in the software, but you can't fix it and your vendor won't because they discontinued the product (and are pretty much out of business themselves, anyway).
Anyway. This is good news, certainly. Though I mostly hope there are parts and components that can be salvaged into slapd.
Just one person's anecdotal experience, but it proves that what he says can certainly happen.
Indeed it does. Rereading my post, now I see that it can be understood as it I thought that could not happen, ever. It can, it does, it has happened to me. It has happened also that Red Hat (before Fedora) or SuSE wouldn't install on some machine.
What I meant was that this is not something that happens often, and is hardly representative of my Linux-installing experience. And it usually has to do with some device poorly supported by Linux, in general. It has happened to me very few times with Debian (and, in fairness, very few times with other distros as well). I honestly can't say that one's better than other. I think Mandrake is good at this, but I've never tried it. I know Knoppix is good at this, save for one time that it didn't detect properly an audio card.
Anyway, I'd hope both of you had the time and the inclination to submit a bug report for consideration for the Debian Installer. I've been really happy with it lately, but if some hardware is not being detected correctly, and FC or other distro gets it right, it should not be too hard to fix the DI and we all would benefit.
Slashdot Mirror
Sigh. So you wanna talk GPL. Fine, you seem like a reasonable person, and I could use a little feedback. Let's do that.
I'm afraid I'm not following you. You seem to be saying that we shouldn't blame Linus for choosing a license that later became incompatible with the CDDL, but we should blame him for choosing a license that later became incompatible with the GPLv3. So I guess the point is we should blame Linus only for not preemptively aligning himself with the FSF's philosophy, without reservations.
If that's what you're saying, and I'm honestly not sure, then I'm afraid we'll have to agree to disagree. I think it's become painfully obvious that the Linux "philosophy", to the extent there is one, and the FSF's, are two distinct points of view, not entirely compatible (see for instance Linux exception for proprietary modules).
Maybe it's that GPLv2 happens to be an adequate common ground what seems to have made people like you believe, mistakenly, that you shared the Linux point of view. Just like it made people like me believe, also mistakenly, that I shared the FSF's point of view. For my part, I now realize that this is not the case. I also believe my point of view is the better one, of course.
But never mind that. Linus' choices were and are consistent with my point of view, which I believe is shared by the Linux developers. GPLv3 is not.
You know what? You're absolutely right. Of course, I still think your preferred upgrade path (following the FSF's agenda) is completely unacceptable. But maybe I should have another look at that MPL.
Agreed. Can't see the irony, though. You mean that "early drafts" thing? I say that not because of the draft's rough quality, but because they made me aware of the direction the FSF intended to take, the things they consider important. As I said, I mistakenly believed that I shared their convictions. That was a rude awakening of sorts. And yes, it was entirely my mistake. Like, for instance, somehow I misunderstood or disregarded Stallman's writings. My bad, won't happen again.
Disagree. GPLv3 goes too far in its attempt to "protect the rights" of everyone. If someone wants to take my code and build a machine that doesn't boot modified software images, say a phone or some specialized portable device, as long as they release the source code I'm cool with that, but GPLv3 forbids it. If they want to write a client for SecondLife or some such, where the software bars the user from copying graphics or models or whatever, then they can't use my code either, because then they couldn't pursue patches disabling that restriction as "circunvention devices" under WIPO. Or so I think. Tell me if I'm wrong, because section 3 of the current draft is barely comprehensible.
(And yes, that last one would be a very stupid thing, but the point is that as long as the code
Yes, I know. And to think I just complained (elsewhere) about the current paranoid climate in open source :-S
Also, I'm complaining about something Sun is giving to some people, just because they seem to have deliberately chosen to exclude Linux. Complaining about people not giving you something you want is lame, I'm fully aware of that.
Still, I'm not trying to belittle or scorn Sun. I acknowledge Sun's contributions and I'm duly grateful for their generosity. I'm arguing because this dude with the hex handle said Linux is being "harmed" because the license its authors prefer does not conform to what Sun (or the FSF) would like, and I wanted to disagree on the grounds that Sun motives may not be in Linux best interests after all. I also have strong opinions concerning his advocacy of the GPLv3, but I really don't want to get into that.
Anyway, my point is, don't get too infatuated with Sun. Sun is a corporation, and one notoriously fickle at that. Schwartz primary responsibility is to Sun's shareholders, not the public in general and certainly not any open source community. This is not wrong or unfair or anything, it's just the way it is. They are not threatening patent litigation, thankfully, but that's now. What about next year? Five years from now? Patents last for 20 years, you know.
Consider that Caldera was once a Linux company. It doesn't get more Linux-friendly than that. If you'd tell me back then that the swines would backstab us all, I probably would have laughed at you, and now I'd be wondering how could I ever be so naïve. Being trustful and optimistic is a great way to go through life and still come out sane, but I think people shoult at least try to keep their feet on the ground.
Thank you for your answer. I think we are in the same page, as to what will happen if Sun releases ZFS under GPLv3. Minor nitpick: when I said Sun would be "forfeiting" the patents, I was just thinking about Sun losing that advantage over Linux, their main competitor. You know, what you referred to as the "main differentiator" (which btw I think it's a bit of an overstatement of the benefits of ZFS, but that's beside the point).
Right. But that's no different than the CDDL. Right now, to use ZFS you have to be able to glue a chunk of CDDL code into your project. Linux can't use it because of the GPL (any version). Shouldn't we, by your reasoning, say that that's Linus' fault for choosing the GPL instead of something else?
Anyway, my point is, it was Sun who chose to license it in a way that Linux couldn't touch it. And if they release it under GPLv3 (big if, IMO), that will also be Sun's choice. Now, you may take the charitative view and say that they had to do it that way, or that being incompatible with Linux is an unfortunate, unintended consequence. Me? Well... I guess I've just become too cynical in my old age.
Well, that's your perception. The short-sighted mistake, I mean. To me, it was a sensible choice. If you care about how people use your code, then you don't leave a backdoor for third parties to relicense your code as they see fit. Not even the FSF.
In fact, and this will sound trollish, and I do apologise for that, but after reading the GPLv3, particularly the early drafts, I have to qualify: especially not the FSF.
Oh and btw, I think BK was also a sensible choice at the time. And the offspring of that "fiasco", namely git, more than compensates for everything. IMO of course.
I'm really sorry for being difficult, but no, I would not agree. C'mon, I keep reading GPLv3 advocates saying as much: GPLv2 is not going anywhere, you are as free to use it as you are to choose GPLv3. Now, regardless of how things got the way they are, or whose "fault" was it, Linux is GPLv2, and it doesn't seem likely that that will change anytime soon. Sun can choose GPLv2 if they want. In fact, I think there's a good chance they'll do precisely that, just so they can use the drivers.
I just read notamisfit's reply to my post. He's right, BSD is pretty much free to use it under the CDDL, as is that guy doing the FUSE port. My mistake.
But you're right too, if Sun does that, Linux will not benefit. In fact, unless Sun allows GPLv2, Linux is pretty much out of luck, and the patents ensure that it will stay out of luck. Another unintended consequence, perhaps.
It would be highly desirable to have it on Linux, yes. If Sun were to license ZFS under GPLv3, then that would be a good reason for considering moving Linux to GPLv3. In fact, this may be the very first actual justification for such a huge undertaking. But that doesn't mean Linux must get it that way. Switching to a compatible license may not be possible at all, and even if it were, whether it would be desirable is questionable. There are always alternatives.
Yeah, about two years behing would be my WAG. I'm thinking along the lines that rewriting the whole thing would be easier than changing the license of the kernel, and that it took Sun some 6 years to implement ZFS but cloning something is much easier than writing it the first time. But see below.
And Linux, via FUSE. It's slow, but getting better.
Whoa dude, hold on a minute. Three points here. First, the kernel devs (not only Linus) chose GPLv2 because they liked it. They don't seem to like GPLv3, so removing the "later version" language was evidently a prescient and wise move.
Second, whether GPLv3 is a better license for Linux is highly debatable. I happen to dislike it for somewhat amorphous ethical and aesthetic reasons... but doesn't matter, that's irrelevant. There are practical concerns: embedded Linux is huge, for instance, and GPLv3 may be poison to that. ZFS alone is probably not worth the risk of losing that "market share" (God, I hate that term).
Third, and most important, you seem to be granting as fact that Sun will license ZFS under GPL. I think that's unlikely. Linus makes some reasonable points in that article, and here's another: Sun itself has got some 50 patents around ZFS. Licensing under GPLv3 would amount to forfeiting these for all practical purposes. In fact, licensing under GPLv2 will probably do that as well.
Now, if Sun actually steps ahead and does it, I will gladly take back all this, apologise, and run to the kitchen to get some Tabasco for my nice plate of crow. And I will enjoy chewing on it, because it would be great to have ZFS. But, like Linus, I won't hold my breath waiting for that to happen.
And while we're talking patents, I don't think you should feel that safe using ZFS on BSD. The fact that there's an implementation available does not release you from Sun's patents, which are still hanging over your head, just the same as if you choose FUSE ZFS. As long as those patents are there, it may be in everyone's best interests (meaning BSD and Linux people) to start working in a free, non-encumbered, superior alternative. I mean, merely cloning ZFS may be not enough.
(*sigh* what a waste of resources... fuckin' evil software patents)
Oh, and Apple may have a patent license from Sun, yes. But somehow I don't think you're advocating switching to a proprietary OS just to get ZFS.
Funny, I've been considering switching from Emacs, too, but for different reasons. In my case, I'm worried by the GPLv3 and all that jazz. I guess I'm having second thoughts about the FSF---all that "forcing evil tivoizers" to "preserve people's freedoms" is starting to rub me the wrong way.
Or maybe I'm just getting paranoid and peculiar in my old age. I hope that's it.
But anyway, politics aside, I've been thinking, hypothetically, how hard would it be to go without GNU. Not only switching to BSD or OSX or something, but actual life without GNU at all. Worst case scenario. Oh and, as you say, I've not switched yet, I don't think things are that bad, and anyway I probably couldn't switch anytime soon, even if I really wanted to. So please don't kill me either.
Right now, there are two pieces of GNU software I absolutely can't live without: Emacs and GCC. To a lesser extent, Bash, GDB and GNU Make. Everything else is negotiable, I think.
Currently GCC can't be replaced, of course, no need to waste time on that one. But Emacs should be replaceable. My problem is nxml. Just thinking of giving up nxml gives me the creeps.
So I guess this rant is to ask, how do people edit XML in vim, or something else? Is there an alternative at all? Would I need a dedicated XML editor instead, besides the one I'd use to write code and plain text?
Oh for the love of root. Nobody is about to be sued by Microsoft. That's just a bluff, a good old fashioned piece of FUD, nothing more.
But hey, even if someone were, what makes you so sure Linus could do something about it? You're probably thinking about him switching Linux to GPLv3, but I have to say, even if that were feasible, I'm not convinced at all that it would be a change for the better. In fact, if the new GPL is designed to do this sort of thing, trying to bind and control people who don't even distribute, then it's starting to look quite evil to me.
Now I'm not a lawyer, and I certainly wouldn't dream of outsmarting Moglen at his game. So I won't even guess if this is actually proper and legal and not an abuse of copyright. That's how it feels, though, and I want no part of that. I'm glad Linus has rejected it.
Btw, that BK "fiasco" was a lot noise, indeed, but look at the bright side: nobody got sued, nobody even acted unfairly (not even Larry McVoy, who was entirely within his right to be an asshole). And in fact the whole episode gave us git, a wonderful piece of software I use everyday now. Not much of a tragedy if you look at it this way, no?
Dude... I'm hoping there's some sarcasm there that I missed. You can't be serious. C'mon, you can't even take Iraq!
No, really. Taking on China is out of the question. That would be Dr. Strangelove-grade insanity.
Besides, I'm pretty sure nobody who matters wants confrontation. The Chinese could screw your economy quite thoroughly, yes, but I don't think they'd want to lose their biggest customer---not to mention seeing the 400 billion they hold in US treasury securities become toilet paper. And American corporations (who seem to run your country anyway) would hate to lose dirt-cheap, profit-maximizing Chinese manufacturing.
So I don't see any sort of confrontation any time soon (thankfully). What I see is American and Chinese billionaries getting even richer, while the general standard of living in China raises marginally, and in America goes to the dogs.
But what do I know, really.
Oh, and as for the pollution thing, that's nasty, indeed, shame on the Chinese. But they're not the world's most polluting country---that would be the assholes who emit 24% of the world's greenhouse gases, and yet refuse to sign up to Kyoto.
Yeah, well, it's kinda sad if a constructive user with some coding skills and no patience for bureaucrats must resort to something like forking, in order to get a useful feature in the GNOME code base. Come to think of it, wasn't something like that attitude that killed XFree86?
But hey, in this case, with Linus being a sacred cow and all, his changes will most likely be taken in, and I'll see that small improvement in a future version. Maybe I'll even find it useful, since I also use the focus-follow-mouse mode, and tend not to maximize windows. Still, a bit worrying if you have to be a sacred cow to get someone to listen at all, at GNOME Towers.
That said, I don't think it's that bad with GNOME, really. As I said, I think most of the screaming is being done neither by Linus nor most of the GNOME coders; just a few people hypersensible to harsh truths and direct language.
Geez. Did you read the thread in that mailing list? I did. Linus is right. Not only right, what he actually did is much more than what I probably would have done in his place (I probably would have sent just the "fuck you", sans patch).
Now I use GNOME, which to me is much prettier than KDE, has a nice shell where I spend most of the time, and lets me run Emacs. So I do appreciate the work the GNOME guys do and all. And never even noticed that right-clicking on the border of windows brings up a useless menu, and that there's no way to change this behavior. So Linus noticed, and it seems he talked to some maintainer and the guys in that list and eventually got fed up. Why, I don't exactly know. In fact, it does surprise me a bit, since I know Linus is a no-nonsense kind of guy, and kinda smart too, and the people who coded this beautiful user interface just can't be morons.
So I think that this brouhaha was caused by, sorry, people like you, getting all worked up in the name of "the norms of human interaction in a public forum" or (more likely in that case) the "norms for talking sweetly to the commitee so a change they didn't thought of is even considered". I'm sorry, but you sound exasperating. *Fuck* the norms of human interaction yadda-yadda, the change is an improvement, it does not affect usability at all (as I said, I didn't even noticed before), and if someone told Linus that they were not going to do it because GNOME is better off without it, then that someone was being a bureaucratic prick and deserved a sound fuck-you.
You do realize how extremely unlikely is that an organism fossilizes after it dies, right? Most of them just decompose and dissolve back into the environment. It is a freak occurrence for a fossil to form: the organism has to be covered with sediment quickly after it dies. Think about it. Then ask yourself if it is really surprising that there are gaps in the fossil record.
Nevertheless, the continuous evolution of life has been going on for so long, and produced so many variations, that transitional fossils are actually quite common. Google for them.
I am not familiar with neither of the "sawed bone" or "glued feathers" anecdotes. Any references to those? In any case, "Lucy" is not the only Australopithecus fossil that has been found. And the dinosaur model with feathers glued onto it... well, I am sorry, but I just can not see the relevance of such a thing existing, if it even does.
As for the genetic code, I do not know who those scientists you are talking about are. But anyway, what is surprising to me is that mutations are so rare. I mean, I have seen bytes changed in tapes and disks, for no apparent reason (exposure to some magnetic field, no doubt). Intuitively I would think the DNA mechanism would be even flakier, what with all the million ways it could go wrong, from simple physical trauma to cosmic rays and so. It obviously is not. Intuition does have its limits.
Evolution is most likely a fact. Regardless of the available evidence, the model works: I have seen it and I have used it in practice. I have made LISP lambda expressions that evolve from gibberish to actual useful functions, using only random mutations and fitness feedback. In fact, I'm working on a pet project which has, at its heart, a genetic algorithm that relies on evolution. And it most certainly works.
Oh, and carbon-14 is a short-range dating technique, as you are plainly aware of. There are longer range radiometric dating techniques available, of course, from Uranium-Thorium to Potassium-Argon and Uranium-Lead. It is not my duty to "show you how" anything is done, but if you are genuinely curious, just search for them, just as for the transitional fossils. Detailed information is abundant.
I can't believe you are serious about this. The notion of Microsoft pulling out of Europe is just too bizarre, even as a thought experiment. I mean, thought experiments are not as fun when they are that disconnected from reality.
I don't think Microsoft can just "cut off" Europe in the way that's been suggested. Not unless they are prepared to refund every company and individual who would be left unable to use the operating system they paid for. I don't know how many copies of Windows could be "shut down by autoupdate", but I guess the amount Microsoft would have to refund would be quite large. Good luck convincing your shareholders that diverting a big chunk of the company's cash to refund people (or to pay for litigation) is preferrable to having to document some protocols.
A little less outlandish, though not even close to realistic, would be for Microsoft to refrain from selling any more products in Europe, and convincing the court to let them skip complying with the previous order on that grounds. But I don't think that would be a very smart choice by Microsoft either. It would cause them to lose a huge market, and all that, but there's something even more important...
Microsoft has had great success imbuing in the minds of people this weird feeling of dependence of Windows. That prediction of yours is a prime example of this. Even in the face of steep prices, defective products, vulnerabilities and all, people just think they need Microsoft to use their computers---it's almost like an addiction.
Now this really puzzles Mac and Linux users, because they know for a fact that there are perfectly serviceable replacements for most of Microsoft products. Not only serviceable, but in many cases technically superior, cheaper, or both. But people at large shun those and go along with Microsoft, out of fud, herd mentality, or whatever---I don't claim to know how that happens. But I do know Microsoft would not even consider forcing people out of their addiction to its products. Force Europe to switch to Linux or whatever, and that many people will find out that there really is no need to pay Microsoft for anything. I submit that would be much worse for Microsoft than losing the European market. In fact, I think that would be the end of Microsoft's dominance worldwide.
That's just disgusting.
Anyway, back here in reality, I think Microsoft will "cave in" and document the stupid protocols. Really, they have no choice. It will be harder for them to keep their monopoly, having fewer ways to lock-in people and all, but at least it won't cost them their entire business. Not immediately, at least.
I don't know, but if I had to guess, I'd say probably not much. I mean, if that particular datum were newsworthy, it'd be all over the news by now.
But... why is that particular question relevant? I mean, if the Democrats were doing the same, does it somehow justify the Republicans? Would it make it right, or at least not as wrong?
Anyway. I don't really care for, or know much about, your political parties. I'm just another foreigner watching the process from home, concerned about the possibility of your neocons keeping their power. That'd be just sad.
I hope it's like you say, because in the media it came more like this: ...
Funny. However, I think this is probably just delay in the media picking up the Spamhaus story. For instance, the last three stories about this on Slashdot concern with reactions and commentary, by different users and reporters, on a single Spamhaus press release. Both Steve Linford and Richard Cox have been asked to comment, and to the best of my knowledge, they've just repeated what can already be found in Spamhaus' answer to the proposed order:
What I'm saying is, I don't really think it is Spamhaus whining over and over about impending doom. They said their piece. Now it's the media slowly picking it up, with every new article coming through as if Spamhaus made yet another dramatic warning about the fate of a world without them.
But then again, what do I know? I just can't understand how some people can take a stand against Spamhaus. IMO they are the best asset the Internet has right now for keeping spam at bay; I know my job would be considerably harder without them, and my users would waste a lot more time digging through the junk. So Spamhaus gave the finger to some judge, and that was not the smartest thing to do. Fine. They probably should apologize to that judge or something. It doesn't change the fact that the judge should not have any authority over them. And I'd also say we should be helping them, not adding to their worries right now.
I think you should wait before changing anything. I don't think spamhaus.org.uk, or any other name besides spamhaus.org, will ever resolve the Spamhaus RBLs.
From Spamhaus' response to the proposed order (proposed, people, by the spammer's counsel, no judge has ordered ICANN anything), it seems they'll intend to contest this. They mention they don't think that ICANN suspending them can actually happen, for reasons I in fact agree with (go read them at their site). They also mention that "one U.S. government agency has begun working on a response."
However, if worse comes to worst, they probably won't switch to any other domain name. They state: "... if Spamhaus gets around the court order by switching domain to maintain the blocking, the judge would very likely then rule us in criminal contempt. We don't want a criminal record for the sake of fighting spam. We normally help fit the spammers with criminal records, not the other way round."
Which I read as, if this order is enforced, and ICANN caves in and all that, there will be no more Spamhaus, period.
Which would really piss me off. The whole episode already already seems like a bad dream to me. To see Spamhaus destroyed by some spammer scum would be just depressing. One thing's for certain, though: it'll be a cold day in hell before any site I manage will exchange traffic with this spammer.
Worse than the problem? So, what you're saying is that if I disable the Spamhaus check in my servers, me and my users will be better off?
Let's see. As things are, currently, my MTA rejects any IP address flagged by either the SBL ("direct UBE sources, verified spam services and ROKSO spammers") or the XBL ("illegal 3rd party exploits, including proxies, worms and trojan exploits"). This happens right before the first SMTP command is accepted, which saves me bandwidth and CPU time I'd otherwise use accepting the headers and scanning them, and, if the message passes that, accepting the body and running clamav and bogofilter on it.
So yes, I "blindly" reject clients listed by Spamhaus. I'm perfectly aware that it is possible I could be blocking some innocent client somewhere. What I'd expect in such a case would be some friend or contact or user would complain that her mail bounced, at which point I'd probably whitelist her and be done with it. Of course, if such problems were common, I'd probably stop using Spamhaus, as their criteria for blocking would clearly be too different than mine and my users'. But I've never had a complaint, so I guess false positives are rare, or my users are really shy (yeah, right).
Anyway. If I don't rely on Spamhaus, or any other RBL because, as you say, they are "worse than the problem", then I have to accept all mail and scan it. I think I usually get over 95% accuracy with bogofilter, but keeping it that well trained is time consuming. And of course, I can't block mail with it, just tag it (a bayesian filter trained with the spam *I* receive is not necessarily accurate for my users), so everyone has to setup rules to actually use the filter rating. If I know my users, they won't do that, they'll rely on the "[SPAM]" added to the subject to filter manually. And 5% of a lot of mail a lot anyway. As for malware, it gets worse: messages are usually very large, and new malware usually passes through clamav for some short time before its db is updated. And I still have users on Windows, regrettably.
If I do rely on Spamhaus, I save bandwidth, processing time and of course human time dealing with false negatives, but I have to deal with the risk of some legitimate message bouncing off my servers because of a Spamhaus mistake. Which to my knowledge has never happened. Which would be trivial to correct anyway.
Hm. Though call...
But I think I'll go with Spamhaus, thank you.
Actually, if the death ray is not a myth at all, then it is likely that it was not constructed the way Mythbusters and the MIT guys did. This is from the "Book of Histories" (Chiliades), by John Tzetzes (circa twelfth century AD):
As much as I can picture such a device, I think what Archimedes could have done is set up an array of flat mirrors along with a mechanism to move them following the sun while keeping the light concentrated in a single point, like a solar furnace. Then he stood at the focus point with a concave mirror (maybe faceted, hence the account of it as "hexagonal") to bounce the light into parallel beams. A death ray, indeed. To hit a target, he only had to point at it with this single mirror.
I don't know how viable this device is, but it doesn't sound incredible to me. I guess there is some caveat I don't see right now; otherwise, I'd be very surprised that these reconstructions of the event didn't use this design.
Not that I disagree with you last remark, but I side with grandparent here: Red Zone Cuba is in fact worse than Manos, unbelieveable as that may be.
Manos may have been made by a fertilizer salesman (which shows, plenty), may have been shot out of focus in its entirety, may have the most ridiculous set of characters ever, but at least had some resemblance of a plot. Granted, a dumb plot, an annoying plot, but it's there. Plus, that lady was very good looking, IMO.
RZC, on the other hand, is technically as bad as Manos, and also completely incoherent. It is just an idiotic sequence of unrelated scenes, ranging from dull to unpleasant, up to frankly insulting.
Besides, RZC actually stars Coleman Francis, the man, the legend. With Anthony Cardoza to boot. That alone should be dreadful enough, but this is worse than the sum of the parts. For instance, RZC is much, much worse than Skydivers (my opinion again, of course---the IMDb for some weird reason ranks it better!).
Dude, the "numbered days" remark was a joke.
I think this guitar playing thing is pretty cool, actually, even if it's music is indeed much worse than what a human player can do. The fact that it was built by some kids in a couple of months, and the fact that it actually works, are quite impressive by themselves.
That's the idea, yes. And it probably will happen.
In the years I've been working with computers, I've seen many attempts to replace sysadmins, and even programmers, with software. To this day I know of no technology that can actually do that, but I'm not counting on this kind of job security. In fact, I'm almost sure that some day a computer will be able to do my job, probably better than I could ever do it. And my job demands quite a bit of creativity, mind you, I'm not a VB-coding monkey (no offense to any VB-coding monkeys around).
In fact, I think I'll see this happen---sometimes I think this will be the reason I'll retire. Which makes me a bit sad, because, hanging out my wife and daughter aside, there's nothing I like more than programming. But I've learned to accept it.
I was a bit surprised to see just how defensive this article got you musicians. I mean no disrespect to you, I know for a fact that a human artist can imbue in music subtleties far beyond my capability of perception (I'm a programmer, not a musician, even if I can play some things on a guitar). But I think you should get used to the idea of machines that can actually do precisely that, that can be as creative as humans. Which is the only reason synthetic music sounds "wrong" today---pure technique is something that a computer can match and surpass humans quite easily.
Creativity so far is a domain exclusive to humans, but I see no reason why this must be so, besides the fact that no computer has been able to emulate human thinking with sufficient accuracy. This, however, will probably change, and I think it will change rather soon.
I agree, to some extent. It always annoys me when people complain about free software, as if they had some right to it. It is a gift. Complaining about something that was given to you is pretty lame, indeed.
However, the lack of a GNU/Linux version is very frustrating, so I also understand the feeling. See, I do not have a Windows machine. I do have valid reasons (at least valid and important to me) for not having one, so I do not even intend to get one in the foreseeable future. So I will not be able to run this software, when otherwise I would be all over it.
This is my choice, of course, so I am not complaining. It is just very frustrating.
I am certain that many people here would gladly "take the time to code all that", for free even, if they only could. But this is not an open source app, so that is not possible.
Come to think about it, this could make a great project. GPL'ed clone of Google Earth, anyone?
I think that sounds like a memory problem, not so much the CPU speed. Perhaps your machine does not have enough physical RAM to hold the huge virtual machine allocation and still cache all the shared libraries required by it. In that case, every startup of the virtual machine will be dreadfully slow, and swapping will only make things worse.
But Java itself is not slow, if you give it enough RAM. That's a misconception. A reasonable one, I will add, particularly if you used the early versions and then dropped it, like I did. But if you'll forgive a little rant, here's an anecdote.
A month ago or so I wrote a general C implementation of the a-star algorithm (AI stuff, commonly used to compute routes on a graph), and tested on some fairly demanding problems, both in memory and CPU cycles. I thought it was rather nice, e.g., solved most solvable combinations of the "eight-slides" puzzle in less than a second, and exhausted the search space for unsolvable combinations in 10 seconds or so.
Then, just for the hell of it, I wrote a quick Java implementation. Would you believe it was nearly 10 times faster? I couldn't, at first. But it really was.
I went back and actually profiled the C version, and found that my data structures were hideously slow. Long story short, I replaced some arrays with red-black trees, and in the end, yes, the C version ran faster than Java. But not much faster. Just some 10-20% faster. And it was much harder to get it right.
Now, this is not to be construed as an endorsement of Java. I don't like the stupid language, and Mono/.Net is much better IMO, as far as virtual machines go---at least you can choose the language. And I don't use programming tools where "Hello World" requires megabytes of memory to even start, just on general principle.
But my point is that, provided you have lots and lots of RAM, recent Java is fast. I couldn't believe exactly how fast until it hit me in the head.
Oh, no, sorry, I think I did not explain myself. I do have inline graphics, as you can see in the screenshot (btw, just apt-getted the standard Debian package, I did not build it myself). What I do not know is how w3m manages to display graphics at all, in a terminal window.
W3m is nice. Here is a screenshot, running in the Gnome terminal:
http://et.com.mx/~crincon/w3m.pngNo, I do not know how they do the graphics.
And here is links, which was also mentioned:
http://et.com.mx/~crincon/links2.pngLinks is my browser of choice when I'm on a system without X, or just in a hurry. It is surprising just how useable (not to mention fast) it is.
Fwiw, I did install a Netscape Directory Server on a HP-UX 11 machine, not that long ago. It was reasonably straightforward, except in that I had to install a number of OS patches and muck around with kernel parameters.
(Btw, what is it with these big proprietary apps that always want to change your kernel parameters? What on earth does Oracle need 2GB of shared memory for? And 64K file descriptors per process? That's beyond ridiculous. That sounds dangerously like extremely sloppy programming inside the product.)
But I digress. My point is that installing and configuring NDS is not hard, but nothing like "soo but soo easy" either (e.g., a far, far cry from "apt-get install slapd").
Enabling SSL is a PITA if you don't have the Netscape Certificate Server (which I didn't). I involves all manner of funky maneuvering with OpenSSL and some tools that you have to fetch from some obscure page at mozilla.org.
Management is more or less the same than with OpenLDAP, which is to say that it mostly depends on how good or bad are your LDAP client tools. In fairness, I hear the Netscape client is nice. I couldn't use it because the damn thing runs on Windows and I was not about to install that in my laptop just to see a stupid LDAP client.
Replication is probably better than OpenLDAP, though I haven't yet a chance to try it on either one.
As for big environments with many users and clients, until today I would have gone with OpenLDAP (or, if a PHB just had to see a lot of money spent in this, with Novell or Microsoft's directories). That's because nobody had source code to NDS and it was all but discontinued from the vendor. You don't want to find yourself in a position where you know there's a bug in the software, but you can't fix it and your vendor won't because they discontinued the product (and are pretty much out of business themselves, anyway).
Anyway. This is good news, certainly. Though I mostly hope there are parts and components that can be salvaged into slapd.
Indeed it does. Rereading my post, now I see that it can be understood as it I thought that could not happen, ever. It can, it does, it has happened to me. It has happened also that Red Hat (before Fedora) or SuSE wouldn't install on some machine.
What I meant was that this is not something that happens often, and is hardly representative of my Linux-installing experience. And it usually has to do with some device poorly supported by Linux, in general. It has happened to me very few times with Debian (and, in fairness, very few times with other distros as well). I honestly can't say that one's better than other. I think Mandrake is good at this, but I've never tried it. I know Knoppix is good at this, save for one time that it didn't detect properly an audio card.
Anyway, I'd hope both of you had the time and the inclination to submit a bug report for consideration for the Debian Installer. I've been really happy with it lately, but if some hardware is not being detected correctly, and FC or other distro gets it right, it should not be too hard to fix the DI and we all would benefit.
Best regards.
I second that. Thanks a lot, guys.