Date: Wed, 04 Sep 2002 06:23:49 -0700 From: Terry Lambert <tlambert2@mindspring.com> To: Dave Hayes <dave@jetcafe.org> Cc: chat@FreeBSD.ORG Subject: Re: Why did evolution fail? Message-ID: <3D760965.47339BA3@mindspring.com> References: <200209041140.g84BeK182877@hokkshideh2.jetcafe.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Dave Hayes wrote: > > The environment chooses the creatures which survive. > > It's not able to choose. The creatures are either able to adapt or > they are not. The creatures don't adapt or not adapt; they are born with the necessary survival characteristics, or they are not. If they are not, they die. If they are, they survive to propagate the genes which express as those characteristics. You seem to be leaning to the Lamarckian hypothesus, which has been disproven, that if you FedEx a bunch of humans to Mars, a portion of them will magically grow the lung tissue to be able to breathe 20 millibars of CO2 at 4 degrees Celcius, and that ability will then be inherited by their offspring. That is not how natural selection's fitness function operates. > Huh? I don't "solve problems" in this fashion. My life is not defined > as "one problem after another". Most of the problems I solve are > scientific in nature, but not even all of those are handlable by the > methodology you describe. All scientific problems are by definition solvable using the scientific method. If they aren't, then they are not scientific problems, they are some other class of problems. > > My personal preference it to analyze the problem, determine > > the class of problems it represents (if non-unique), and then > > solve for the set of problems in the space represented by the > > class, do it once, and never have to look back. > > Gah. What if the problem is dynamic? The method works anyway. > What if the problem mutates? Then you reanalyze it. > What if your classification was in error? Then you start over. > I bet I feel about this methodology what you feel about mine. ;) Unlikely... 8-). > All people are unique. Any classification you do on a unique > individual lowers your accuracy of estimation of said individual. Of course they are... any given indivudual ia unique... just like everyone else. 8-) 8-). > > You mean, like machine enforcement of the charters for technical > > mailing lists... > > Yes, that would be a contender. A machine restricting discourse has > a nauseous taste to it. As long as it only restricts it to the charter, I have no problem with it. If I want to go outside the charter, I take the discussion elsewhere. [ ... signed, timed, signature keys ... ] > Nope. All you are doing here is forcing the users to have a > "verifiable" identity. As most everything is, this is quite probably > hackable, subject to identity theft from careless users, etc. I'm also forcing that verifiable identity to obtain a limited time permission in order to post -- a lease -- which must be renewed to permit continued posting. This permits a feedback mechanism -- whatever mechanism the list membership consensually decides is appropriate -- to be used to enforce against continued abuse of the list. You are a SPAM'mer, and your identity loses posting rights. You are a troll, and your identity loses posting rights. Etc.. > > Also, all messages now non-repudiable (I view this as a > > disadvantage, but not in this context). > > Why not in this context? Because there is no power yielded by the individual to the group, apart from consensus control over posting rights. You are not going to be arreseted and sent to a reeducation camp, if you continually post about something both off topic and unpopular with the list membership. Unpopularity of a view is irrelevent, as long as the view is topical. If you change the feedback criteria, and the notification mechanism, then it can be an instrument of oppression. But the ability to use a hammer to kill someone doesn't make it any less of a tool than when you only used it to poind in nails. > >> > On the contrary. It is the nature of science to question assumptions. > >> > I see scientists question their own assumptions all the time; all that > >> > is required to trigger this is a contradictory observation. Scientists > >> > never hold forth facts, only hypothesis. > >> > >> Observational evidence contradicts this assertion. Really, I've rarely > >> seen this, and that fact is why I escaped academia years ago. (They tried > >> to hold me in but...) > > > > As I said before, you are hanging with the wrong peeps. > > Define "the right peeps". Whatever group it is, I don't belong, > period. I've walked the line between many classified groups ever since > I was born. People who call themselves scientists, but who don't walk the walk. > >> > [ ... profoundly bad example ... ] > >> Why? > > Because it analogizes an impedence mismatch with a convergent > > series. > > See? You aren't willing to think out of the box, or to critically > examine the concept. You dismiss it out of hand because of your > classifications. I dismiss it because it is a flawed analogy. Come up with a valid analogy, and I won't dismiss it. Your assumption about what happens when you sample something whose frequency is higher than the sample rate being similar to what happens when you set V > C in a Lorentz transformation is incorrect, because there is not equal symmetry around the centerpoint. > > Never test for an error condition you don't know how to handle. > > -- Steinbach > > How do you know you can handle it before you get it? What does your program do, when it can't read the file, but your process has sufficient priviledge to change the access controls on the file to permit it to be read by your program? > >> Sometimes, a model that doesn't "academically work" can still > >> "practically work". > > "Finger quotes"?!? > > Eh? The use of ``"practically work"'' instead of ``practically work'' says that you were attempting to imply a non-traditional meaning. > >> > Why can't it be orthogonalized? You are effectively arguing > >> > against the Taniyama-Shimura conjecture... which has been > >> > proven. > >> > >> The who? Good grief. Is this an authority? ;) > > > > "All elliptic curves have modular forms" > > So? How does this imply that you can orthogonalize -all- aspects of > life? I never said you could. I responded to your statement: | You can't orthogonalize this. You can't just apply a transform and have | the troll component vanish, you still affect the other communication. If you can identify the trolls, you can in fact, find a modular space in which there is a manifold dividing the space, with all the trolls on one side of the manifold, and everyone else on the other. Then you can apply a simple binary "trollness" test. It has nothing whatsoever to do with "orthogonalizing -all- aspects of life". Nice try, though. > >> I can't agree with that at all. The world of humans doesn't always > >> obey any strict mathematical definition, and as such is not a > >> candidate for scientific manners of investigation. > > > > Oh, this is so wrong. > > We have arrived at another fundamental disagreement then. Nevertheless, I will continue to use such manners of investigation, so long as they continue to yield highly accurate predictive models. 8-). > > Individual humans are not completely predictable (yet), but > > statistically, groups of humans are very, very predicatable. > > Statistical arguments are generally inconclusive. They are hard > to accept unless you can guarantee a bunch of hard to guarantee > things about the evidence. I disagree. Perhaps what you feel is hard and what I feel is hard are two different things. > That being said, there are certain human foibles that you can see in > most people like clockwork. These could resemble statistical > observations, but they are best understood by watching individual > examples and then applying that experiential knowledge to further > observation. I didn't say that it was not possible to model individual behaviour entirely, only that, as yet, accurate predictive models can not be built from such observations, except in the most trivial cases. Otherwise, we would have software which can pass the turing test, simply by running such a model. > >> I did that. Free.* was taken over by Tim Skirvin... > > > > That was a hierarchy within the context of the genereal usenet. > > I'm talking about non-interoperation. > > The entire point wasn't to make my own sandbox and see who would play > in it. This was a very common straw man. It's irrelevant to the drive > I had at the time to express common sense and teach people (by the > action of not moderating) to -freaking- press the "delete" or "next > messsage" key when you don't like what someone posted. What is so > damn -hard- about that? Why can't people just do this? Moving the > finger takes very little caloric energy, less energy than continuing > to read and get worked up. By not making it "your own sandbox", you failed to put a border between your society and Tim's. The result was predictable. > Just look. -You- want to spend a lot of time and energy devising > secure identified email or coming up with who knows what just so that > the laziness of humanity can prevail over common sense. Hardly. I want common sense to prevail. But the trolls refuse to exhibit it. > This is all just more evidence that Earth is really a comedic stage > for the amusement of whatever cosmic being(s) are out there > watching. ;) I resemble that remark... ;^). > > Any existing system that fulfills a similar societal role is a > > control. I think you are confusing the society itself, which is > > an independent entity, with the communications media within > > which its internal systems operate. The two are not identical. > > Maybe so, but they sure lose a lot of distinction in the process. > Also, however correct you are, the people -in- the society > don't seem to agree with this. They tend to percieve them as one. That's why I keep suggesting that the "laws of physics" need to be built into the the pathways, rather than externally imposed. You keep arguing that internal imposition won't work. Fine. Take that as a working hypothesis, and impose the rules externally instead. > >> Perhaps support for your "paid troll" theory can be had by noting > >> that paid trolls don't really care about the response as long as > >> they can shut down the list. (I'm trying to think like you here, > >> correct me if I am wrong but I think this is your theory.) > > > > Yes, this is my theory. > > They got to ya then. ;) It would appear you are at least somewhat > worried about the list being shut down by trolls. If that's true, > they've managed to win the first round. Hardly. Their goal and their actual ability to achieve it are very different things. > >> > It depends on what you mean by "moderation". > >> > >> Classic moderation is where a small subset of "society" gets all > >> the messages destined for a forum. They then determine whether to post > >> those or not. > > > > This doesn't work. Not because of the reasons you keep claiming, > > but because it will not scale. > > So a superposition of reasons then. I claim classic moderation chills > speech. I don't think anyone can prove otherwise, rationally or > experientially. It doesn't chill speech within the consensus of the meaning of the charter, if implemented correctly. Any you will have a hard time proving a negative in that case, too. 8-). > > Not a dodge. My Uncle-by-marriage's sister is the person who > > dispenses Charles Manson's medication. Some people yanked out > > out their interface cables before the programming was complete. > > Some people didn't trust the code and that's why they yanked. Our > "society" is not perfect, and I daresay far from it. People like > this guy are a reaction to it, which has been increasing in past > years. It's OK. We'll lock them up and prevent their genes from propagating. > >> Not if I have the Dallas Cowboy Cheerleaders at gunpoint. ;) > > > > I can still ignore what you have to say, and report on the > > whack-job with the famous hostages... > > And your boss can fire you and assign another reporter, yes. Not really. I will be giving the boss what he wants: viewers; how many people have actually *read* "The Unibomber Manifesto" (or "The GNU Manifesto")? A circus doesn't have to have a plot. > >> Not at gunpoint, but I do have over 35 active FreeBSD systems > >> to care for...I think there's an imperative there don't you? > > > > So what information pertinent to that situation are you getting > > from the "FreeBSD is Dead" trolls? > > Starting with the obvious, Someone feels threatened by FreeBSD. I'll grant that. We got that the first time they posted. They've posted more than once. What *new* information was present in each subsequent posting, which was not present in previous postings? Remember that the mechanism I have proposed would not have stopped the initial posting. > >> Maybe this response is "Hey, friend..."? (Ok, so that's -my- utopia, > >> not yours.) > > > > Hey, if it worked... but it wouldn't. > > It's worked for me in the past. I wouldn't call it reliable, but then > again...to do this one you have to be impeccably appropriate. So it worked with Tim, did it? > >> > The noosphere is not bounded to a finite competitive resource > >> > domain, as you keep implying with your "move to an island" > >> > analogy. > >> > >> I'll grant you finitely uncountable, but really the limit is > >> in how long you have to peruse it. > > > > Not long. You have filters, right? > > Yes. I still have trouble keeping up with it all. Yet you expect people not dedicated to your ideal to keep up, even when you, a dedicated person, can not? > >> The difference between you and I is, I can operate independent of > >> my axioms. Sometimes without thought even. If I'm lucky, complete > >> mental shutdown. > > > > You may as well be a puppet, if you give them that much control > > over you. > > Them? Nope. This is my control over me. Deprogramming my mind and > letting who I really am surface. That's exactly what a pupet in your position would say. 8-) 8-). > >> As the limit of time approaches infinity, you can't. ;) > > > > Functionally decompose the problem space, and distribute the > > processing. You're asking the same thing of personal filtering, > > only you are asking it of a multiplicity equivalent to the fan > > out for a given mailing list. > > I'm not asking anything. I'm implying that unless it's done this way, > it's not honorable. People should determine what they want to read. > The converse is just as multiplicative; you have to sit there and make > presumptions about what N people want to read. As N grows large, you > are bound to make decisions that a portion of N would disagree with. > This is what stagnates a list, since you have to LCD the presumptions > to get "the most people" happy. Hardly. Topicality is not arbitrary, even if choices about the content of the charter are. > > Mailing lists are push model. They are not Usenet. Stop pretending > > they are. > > The distinction is irrelevant in this case. Functionally, they are the > same thing, just on different scales. Wrong. The distiction is critical. It defined the tipping point. > > You could have just ignore my response. So by your argument, > > you should take responsibility for initiating this entire > > diatribe. > > I merely posted a thought. You attacked that thought. That started > the diatribe. Stop weasling, you must have known I wouldn't just > back off. ;) I merely posted a thought about your thought; there was no attack. > >> So if someone is chopping the hedges on your side of the fence, > >> and you blow him away with a 12-gauge shotgun, it's ok because > >> it was effective? > > > > If they cart you off to prision, and in two years someone buys > > the house next door, and chops the hedges on your side of the > > fence, then it wasn't effective. > > Interesting dodge. How is that a dodge? -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3D760965.47339BA3>