Because Python isn't Java. All of the Python apps I have used were fast and worked well. I can't say the same of Java apps.:(...and the really funny thing is this is mostly true even though a Java JIT compiler is typically quite a lot faster than the Python interpreter when benchmarked running equivalent code.
(Part of what this implies is that Java code really can be quite a lot faster than Python when it's written right using libraries that don't suck -- but the "libraries that don't suck" bit is important; a Python/GTK interface will be faster than Java/Swing any day, even though the Python interpreter is slower to process the program logic).
So basicly you're saying that there are no consequences to our actions?
Of course not -- I'm merely rejecting your premise that all actions with negative consequences are "evil", and consequently your presumption that testing for negative consequences is implicitly testing for evil.
BTW a great objective test for the presence of evil- Dead Human Bodies.
Counterexample: Old age isn't evil -- being an impersonal force it cannot be -- yet it results in dead bodies. Further, there are a great many things which are evil but which don't result in dead bodies. Theft, fraud... and so forth. Keep looking for that objective test.
No, I may be a fool. Therefore I must be careful. Neither are you unless you have acess to all relavent information and know when you have it all. Since some of the information I need to make a good decision will not exist until after the fact, at best I can only act in my percieved best interests.
Yes, of *course* you're being careful -- being careful when you need to is part of being competant, after all.
That said: You obviously have some kind of ruleset (perhaps one based in large part on the moral positions you're presently defending) which permits you to choose actions which you believe to be less disadvantageous than others -- if you had no such thing, your actions would be nothing but arbitrary. You believe that the actions you choose using this ruleset are, by and large, better than the actions you would choose without using this ruleset -- otherwise you wouldn't choose to use the ruleset, and certainly wouldn't be defending it here. (Remember, I didn't say "perfect", I said "competant").
No, I have and will be harmed by the unintended consequences of others actions. Unless you believe that my car getting trashed by a drunk while I was asleep in my bed is somehow my fault? The drunk had to mow down a fence and an apple tree to get to my car BTW.
*sigh*. I didn't ask if you would be subject to no damages, I asked if you would be subject to no additional damages -- that is to say, any damages which you are not already subject to. The drunk trashing your car, for instance, is obviously a form of damage to which you are already subject. By going off on this tangeant, you have completely avoided answering the question I asked.
You presume that no one is a fool. Incidently, fools do not believe they are fools. Lastly, many times a fool in his own cause can be wise in anothers cause. We should look out for each other.
I presume that a fool who does himself harm will be less of a fool next time around, as will his children and/or others who have the wisdom to learn from his mistake. I also presume that self-determination -- freedom, in short -- is a higher standard to uphold than wellbeing, with all the consequences this implies.
Are you a fool?
Heh. Well, I'm pretty damn foolish from time to time, but I wouldn't describe myself as a fool -- for that matter, I can't think of any people I would describe as unqualified fools. I can think of folks who are a bit short on common sense, or not particularly skilled in their profession, or helpless when faced with what should be a fairly simple automotive, plumbing or wiring problem -- and indeed, I fit into at least one of the above categories -- but even so, I'm disinclined to think that meeting any of these criteria makes an individual someone whom I'd without qualifications consider "a fool" except within the narrow scope to which that limitation applies.
If you define "a fool" as "anyone who makes poor decisions", then I no doubt am that, and frequently -- as, in that case, are we all. A definition so loose that it applies to anyone, however, is a definition so loose as to be meaningless; in this case, therefore, I prefer to use one which is substantially more restrictive.
First, consider the allegation. Copyrighted code is part of the Linux kernel. What if it's true? Shouldn't the copyright holder get its day in court?
Sure, the copyright holder should get their day in court. The copyright holder should also try to mitigate damages -- indeed, their ability to get the court's assistance will be severely limited if they don't. As of yet, they refuse to let us know exactly what we can do to reduce their damages (by refusing to identify the infringing code for removal). This is not consistant with behaviour with this allegation being true.
And consider the lawyer...David Boies. In case you've forgotten what what he's done...
No, I haven't forgotten about Boies losing every high-profile case he's been involved in -- and for good reason. Have you read SCO's legal briefs? They're downright amateurish next to IBM's filings.
Do you think the diligent researchers on Wall Street are confused by McBride's proclamations? No, they do research.
I'll believe that their research is effective when my own starts backing it up.
That's not addressing the problem, that's simply agreeing to ignore the problem. This is not a problem of legislation, this is more akin to natural law, like F = (m*a + V*dm/dt)!
You're presuming the conclusion: My definition only "agrees to ignore" harms done with consent from the so-called victim -- which, as you may recall, I argue to be no problem at all!
The "natural law" argument falls flat on its face: Nobody argues about the law of gravity, because such a thing can be objectively tested. I've yet to see an objective test for "evil", however -- perhaps a heat sensor combined with measurement of ambient sulfur levels?
As for the VERY BAD THINGS you so ominously warn of: If those VERY BAD THINGS happen with the consent of those to whom they occur, so be it! That said, if the argument which you so decried as "sophistry" is correct, and individuals will indeed more closely follow a moral code whose strictures they are in fact capable of complying with, then perhaps the number of VERY BAD THINGS which happen under my preferred code is fewer than those which happen under yours.
The responsibility of every human being is to realize what they do and act to mitigate it and to try to do some positive good.
You know, if I try at it for a bit, I'm sure I can come up with a broad, sweeping, mom-and-apple-pie sounding statement about things that other people ought to do, too. That's not to say that this statement actually creates some moral obligation on the part of others, or would make me a better person for having said it.
The only clear obligation which I see as essential to an effectively functioning society is "first, do no harm". To actively help others -- to do good -- is not a baseline that is expected of every individual (such that any individual who helps others is doing no more than the duty obliged them), but an outstanding act by which the individual who does good can be considered to have gone above and beyond the call of duty.
Once again, this mode of thought is intended to encourage good works; those who go above and beyond are allowed more room for self-praise -- and thus more encouragement to continue their good works -- than those who are merely trying to counterbalance the effects of those unavoidable evils they've committed. Likewise, trying to earn a few more points towards the fulfillment of an unending obligation is not an entirely encouraging situation. Compare the man working to pay off a $5 billion debt to the man working for pay he can keep; the former is far more likely to become unenthused regarding his efforts.
Now, let's try looking at this from another perspective. A few questions (and yes, they're a bit leading):
Do you believe that you are competant to determine which actions are within your best interests?
Presuming that you do believe yourself competant, then, does it not follow that you -- not everyone, just you -- will not be subject to damages additional to those which presently befall you if the only harms done you by yourself or others are those harms which you have considered and decided are in your best interests to accept as risks?
If you yourself are competant, and you are capable of determining an appropriate course of action for yourself under the the above limitations, is it not the worst sort of egotism to determine that others are not so competant to determine an appropriate course for themselves as you are for your own self?
...because they were able to pick out combinations of optimizations that measuring each optimization individually wouldn't have discovered.
Consider the case in huffbench when -fmove-all-movables, -freduce-all-givs and -ftracer were discovered to result in great improvements when used with the other GA-selected optimizations -- but not when applied directly to -O3.
Sure, these situations may be specific to the test cases -- but that's not to say his software can't be used to evaluate real-world scenarios as well, such that companies doing a final release compile of their code let a GA churn for a few days to determine the best flags to use.
"without knowing consent" means that the same act can be evil or not depending on the assent of the target.
Absolutely. I think we agree that this in fact the case. If I consent to give you my car, and you take it, then what you've done is appropriate. If you take my car without my consent, what you've done is evil.
Giving you my car may be something that does me harm -- maybe I don't have another ready way to get to work in the morning -- but that's my own lookout; if I decide I want to give you my car (for whatever reason -- maybe you offered me something in return, maybe you needed it more than me and I like helping my friends), then you're doing nothing wrong by taking it.
I did not say the mitigating circumstances make doing an evil thing right. I am saying that regardless of circumstances the act is evil. War is evil; does that mean it's unnecessary, no; does that make it less evil, also no. We human beings can no more avoid doing evil than we can avoid breathing. But we can try.
If I had to come up with one reason I prefer my moral stance, you've made it abundantly clear: By my definition of evil, evil is something one can succesfully avoid. You may go and argue why evil things are unavoidable -- and engage in such actions while knowing them to be evil. Evil, by my definition, is something which can be avoided in full -- and when I don't avoid it, I have no excuse that it's "no more avoidable than breathing".
A less restrictive definition of "evil", thus, can result in an overall reduction in harmful behaviour in the same manner by which laws which the average man can realistically follow are better respected when unaccompanied by laws which he will not.
Okay, you assert that it's relative. I assert that it's not. Show me how "knowing consent" is any more relative than the definition of "destruction of a human being", and I might be impressed. Both have fuzzy areas -- places where you say it's destructive and I say it's not, places where I say consent is given and you say it's not -- but both have very clear, very firm areas where the condition is met or isn't.
[W]hat makes you think that imprisoning a human being is, in any way, shape, or form, anything but an evil act?
Because doing evil is wrong. If you believe that mitigating circumstances ("necessary, prudent" and so forth) can make doing an evil thing right, your moral position is one hell of a lot more relative than mine is.
My definition of evil is considerably older than your 20th century moral relativism.
What part of "harm without knowing consent" is relative? I'd call that a pretty damn black-and-white line... it's just a line that's different than the one you draw.
I'll grant that there are special cases -- but your definition has them too. Imprisoning or fining someone who's committed a crime is harm, and the only "consent" available is the implicit acceptance of possible consequences in commitment of the crime. That same case applies to your definition -- why isn't imprisoning someone judged guilty of a crime *evil*?
Porn is a distorted image of that, and people who build their lives around porn (strippers, prostitutes, pornographers, etc.) ultimatly are destroying themselves and the people around them.
[WARNING: I drift off-topic here. Consider that part of the risk you take in crafting a slippery-slope "porn leads to $FOO" argument -- it's a nice lead-in for me to ramble about $FOO].
People who build their lives around work (coders, designers, sysads, etc.) ultimately are destroying themselves and the people around them.
No, seriously -- I've seen waaay more peoples' relationships destroyed by workaholic tendencies than by porn and all the "alternative lifestyle" bits you associate with it
Granted, there are those whose behaviour is clearly self-destructive -- not merely when viewed through some set of moral guidelines but when taking an objective view of their wellbeing. Typically, though, self-destructive people are self-destructive people -- whether putting themselves into as many bad relationship as they can arrange, ignoring doctors' orders and letting a serious medical condition become life-threatening, or deciding to spend their days gaming rather than looking for work. Just because one can eat oneself' to the point of obesity doesn't mean that food is a bad thing, or even that everyone should be required to hold a 1500-calorie-a-day diet.
A lifestyle element which is destructive in one case may not be so in another. I know some potheads whose lives revolve around their drug -- but I know far more productive members of society who will occasionally partake of a passed-around joint. That alcohol can be used to excess, with exceedingly destructive effects, doesn't mean a one-drink-a-day maximum is always appropriate. Self-destructive people won't be stopped from self-destrution merely by taking away their mechanism, and doing damage to the range of choices available to those who are not self-destructive in the name of protecting those who are is an insult to the intelligence and free will of humanity as a whole.
And hitting on the "alternative lifestyle" bit again: I honestly don't believe that the traditional 1:1 husband/wife closed relationship is the only thing that works, because I've seen too many succesful (long-term, even!) relationships with different foundations at their core. Yup, I've seen unsuccesful alternative relationships as well -- but not grossly out of proportion, and largely involving people with self-destructive tendencies of the sort I mention above.
Okay, all that said, back to answering your points.
I think you have the answer implicit in your own statement. As soon as they test positive for an STD they are no longer working, biweekly tests remember?
Yup. I find this adequate. Every life of business which generates good returns requires either substantial up-front investment or assumed risk; in this case it's the latter. Given that the licensing program requires the individual assuming such risks to be aware of said risks and skilled in minimizing their impact, I find I have few remaining moral objections with regard to assumption by the provider of any level of risk which may remain -- and risk to the customer is quite assuredly lower (by virtue of both the training and the regular tests) than the like risk with a member of the opposite sex picked up at a bar; hence, compared to some of the more popular alternatives, legalized and regulated prostitution is a very safe activity. (WRT risk to the providers: I'm putting on my caloused libertarian exterior here and arguing that someone who, being of legal age, contracts to undertake an activity which poses a risk regarding which said individual is fully informed, assumes said risk in such a way that they've only themselves to blame should the risked danger befall them -- same applies to factory workers risking industrial accidents, typists risking repetitive strain injuries and anyone else whose profession includes connotations of risk. It's cases where a
All I can say is that it can't be worse than what Sun's traditionally made.
I at one point had an opportunity to come in contact with a prototype embedded Sparc chip and reference board. The thing drew *70 watts* of power. For an embedded board.
My old university was recently (okay, it's been a few years now) donated a 64-unit sparc cluster by a national laboratory who was done with it. Last I'd heard they only had 2 units powered up -- they needed to have their (large, heavy-duty) server room refitted with more power and better A/C before they could turn on the rest.
Honestly -- there's no way AMD could do any worse.
Hmm -- you bring up a very important point: In my previous position, I wasn't thinking "internet" (though of course that's the topic of this story) -- none of the children I interact with on a regular basis are old enough to use chat or the like (few of them spent much time in front of a CRT at all, for that matter), and I just stopped thinking about it.
I might be inclined to agree with you that a level of caution makes sense online that is likely to be more counterproductive than useful when applied to real life.
[As a totally off-topic aside, I'm inclined to think that the not-much-time-in-front-of-a-CRT thing is a Good Thing at their present ages -- between 3 and 6 -- though I've done an SDL port of UCB's Logo implementation specifically to have something for my household's 5yo to learn with when she's ready].
Maybe they don't like to give away complete control over their patent licenses? Sounds like it makes perfect (business) sense to me.
It sounds like abuse of a section of law originally created for the protection of tangible inventions to me.
It makes perfect sense for a protocol's documentation to enjoy copyright protection. Patenting the protocol or format itself -- well, if someone died and made me the US legislature, there'd be a quick change of patent office policy on that count.
That 1% may or may not have been correct as original stated, but you've certainly misinterpreted it. It certainly does not say that if a specific child in this specific situation is not molested by his/her immediate family, that child has only 1% (or "likewise very small") of the "normal" risk for being molested. It'd be a reasonable guess that the risk is less than normal, given that the major risk factor in the general population has been eliminated. But that's all it is; a guess. That's not valid deductive reasoning; there are plenty of situations in which that sort of reasoning will get you a conclusion that is not true.
a - chance of general-population child being molested b - chance of child being molested by family c - chance of child being molested by strangers
a=b+c c=b*0.01
Now, let's say b is.2 (needless to say I hope this exceedingly pessimistic) and solve for a and c.
c = (.2)*0.01 = 0.002 a = (.2)+(.002) = 0.202
But wait, you say! I haven't proven that children not molested by family members aren't at a higher risk from strangers -- a dangerous assumption, until one thinks about it critically and realizes that even if all the risk of (c) were shifted onto the set not covered by (b),(c*.2) is still an absolutely tiny increase -- and any larger increase is not possible because it would increase the total percentage of molested children who were molested by strangers as opposed to family members.
So -- yes, it reasonably follows that any child who is not molested by family members has a very low chance of being molested at all -- low enough, in my view, that teaching children to fear anyone they don't know as a potential bogeyman does far more harm than good.
It's the people who really don't belong at a place that kids need to be scared of, the people who do belong to a place are usually good people there to help.
I'm not sure that automatically distrusting "out-of-place" people is really a good standard either. It's something the intentionally deceptive can and do fake, and it's prone to a high false-positive rate. Perhaps the best solution is to simply establish that only adults known to be trusted by Mommy and Daddy can make requests which would otherwise be rule-breaking (leaving the front of the school before Mommy shows up, for instance, is quite certainly rule-breaking) and, as you say, authority figures are good people to go to for help (but one should still think cautiously when asked to break rules on their say-so).
How effective would this be in real life? I don't know. The 5-year-old in my household has a remarkably good BS-detector for her age, better than that of some adults I know, and a willingness to question requests that don't make sense to her. Hopefully she'd be as quick at figuring out that some stranger (or, more importantly, some not-quite-a-stranger) is leading her on as she is when her family members do the same.
In any event, preparedness for strangers with ill intent is pretty much a non-issue; see the first half of my post. Saying "no" when her friends ask her to do something she knows is wrong... that's a much more real and pressing problem.
Your definition is incomplete. Certainly, humans will learn to seek rewards (rewards being things that cause the "goodness function" to evaluate high) -- but how is it determined what items have a high "goodness function" such that they function as a reward? The goodness function, then, isn't entirely determined *by* rewards; indeed, the goodness function determines what things *are* rewards.
Describing conditioning as the be all and end all of human behaviour is also incomplete -- ones' hunger for food, ones' need for affection, or even traits like curiosity are largely inbuilt -- though I'll certainly grant that conditioning later modifies the associated weightings.
One final bit -- I don't know if I need to spell this out, but might as well. A "goodness function" is called such not because it necessarily makes a human good, but earns its name as it's used to interpret how good one's curent state is versus how good ones' state would be if one just did $FOO. For selfish people this "goodness" function could evaluate high on a course of action that hurts others but results in personal gain, for instance), while folks with higher empathy weightings (however those are established -- I can think of a few plausible guesses but they're just that) more quickly decrease a scenario's goodness rating based on harm done to others.
In summary: Condition certainly affects later behaviour as it modifies the weights used in decisionmaking (including goodness-function evaluation). That said, humans are born with a pre-built goodness function -- if not our brains wouldn't know which stimuli to treat as rewards and to seek for. Coming up with a good equivalent to that inbuilt goodness function is a Hard Problem in artificial intelligence, and coming up with a general solution has a lot of potential to make AI problem-solving techniques we already posess much more useful in the context of building "strong" AI.
To me that's one requisite for intelligence, the ability to play a reasonable game with nothing other than raw rules and some practice.
Quite a lot of AI techniques work this way -- genetic algorithms, neural networks (presuming some iterative training based on review of its actions, etc). The real trick is coming up with a "goodness function" which can be used for training -- once the machine has a function that it can seek to optimize, one can pretty much just let it go and figure out how to actually do the optimization itself given examples to watch and/or practice trying itself. Right now, though, that goodness function pretty much has to be written by hand and is very specific to what the AI in question is trying to do.
I'm fairly sure us humans (presuming, of course, that you're one) have a "goodness function" too -- though it's not a single number but a set of neurochemicals and other brain state -- and that we tend to learn to take actions which maximize this function (via receiving praise from others, satisfying basic needs, curiosities, etc). That said, coming up with a good general purpose one that could be used by a computer is no doubt far easier said than done.
All that said: I spent more time back at university in AI classes than most (and under some very good professors), but I by no means claim to be an expert on the subject; my knowledge is quite heavily weighted towards theoretical as opposed to practical content.
Last I knew, professional prostitutes (the legalized, licensed, Nevada kind) had a *much* lower STD rate than any other population segment -- given that in order to be licensed for their profession they require knowledge on best practices for avoiding contracting STDs and identifying those posessing them, are obliged submit to biweekly blood tests, etc. I'd expect that those involved in (professional) porn would have a similarly low rate, given the similar need for their employers to require regular testing pre-employment.
Finally, remember, we're discussing porn. Not prostitution, not sex clubs, not "alternative lifestyles", porn. I haven't heard of anyone getting an STD from a magazine yet.
So I think in this case you're best off assuming the danger is coming from strangers. Or maybe from other close acquaintances. Or from anyone but the immediate family, because they're the ones you're talking to.
Or maybe you're better off assuming that making your kids paranoid about anyone whom they don't know yet is a bigger threat to their overall wellbeing than that 1% (or likewise very small) remaining risk.
Are you the same kind of person who won't take your child on an airline because planes sometimes crash? There's a notion of "acceptable risk"; I consider talking to strangers (not blindly obeying strangers, but talking to strangers) well within that area.
4, typically -- it's the year before kindergarten, though there are occasionally side cases where one might end up starting preschool early and staying in it longer (typically because public schools often won't accept children a day under 5 into kindergarten, even if they're otherwise ready).
If you want to know why pornography is evil then just think about it from the point of view of the photographic objects. Imagine that was you, how degraded would you feel if you were the subject of those pictures. What kind of psycological defences would you have to put up just to get up in the morning? Kind of makes you understand why so many people in the porn industry are addicts doesn't it?
What makes you expect that said people would feel "degraded"? Your values, of course.
People with a different value set may have a different perspective.
I can't speak as to porn, but a few of my friends once worked as dancers in strip clubs (and another spent some time working as a nude model for the local university's art department). I can't say any of them ever saw their own work as degrading, despite the dissenting opinions of others.
One could even go as far as to argue that if it's the "degredation" meme that causes pornography to be harmful, then it's those (such as yourself) who perpetuate that meme who are responsible for the relevant harm.
Then your distribution will almost invariably come with the drivers bundled as a module that can be automatically loaded when the hardware is detected.
If you've got a *really* odd piece of hardware, then you might need to do a bit of trickery (download/compile a kernel module, or modify the code of an existing driver to recognize a new card using the same chipset), but that's rare -- and if you're buying hardware that's that unusual, you're typically an enthusiast, which means you should be competant to download/compile a 3rd-party kernel module yourself.
Please -- nobody's innovated in word processors or spreadsheets for years. What "innovative" new versions are, to 99% of users, is cost centers. Open source software shares development costs, and so reduces the costs of the businesses using it.
Being able to build a business that's doing something truly innovative while using zero-cost, open source software as building blocks means those engaged in true innovation are spending less for the privilege of being able to do business -- which means there's more money left to fund real innovation rather than throwing money away on licenses for software which reached the plateau of its usefulness years ago.
As a redhat stock holder, I'm extremely disappointed with the recent changes redhat has made, when they IPO'd and I got the stock as soon I could, not because I particularly liked redhat, I don't, I've nver used it as my primary OS and I prefer many other flavors of Linux over it; however I saw in redhat the best chance at a real competitor to microsoft; and now they are not.
How so?
They still sell and support a workstation product (Red Hat WS), and while they take the official position that it isn't ready for general (ie. home) desktop use, I haven't seen any statements to the effect that they don't still believe in Linux on the *corporate* desktop.
Just because they've made "Red Hat Linux" into "Fedora Linux" and stopped selling support for it such that their lowest-end supported desktop product is Red Hat WS, doesn't mean they're entirely out of the race.
True, but by understanding something one can understand where its weaknesses are, and one may take actions to minimize the likelihood of exploitation of these weaknesses (or maximize the likelihood of detection of said exploitation after-the-fact).
In the case of automated voting systems without a voter-verifiable paper trail, the ability to detect prior tampering is minimal; in a great many cases, detecting tampering after-the-fact is impossible. Likewise, because the system has several components which are difficult to oversee, there are a number of places where the results may be unduely influenced before the election itself occurs (ie. subtle undocumented features in the code).
It is because understanding is necessary to implement safeguards before-the-fact and audits after-the-fact that said understanding is such an important element -- not merely for understanding's own sake.
I say this, by the way, as the lead developer (a few years back) of an online voting system used for the elections of a public university in northern California and an associated $14M corporation. My system was delivered with detailed documentation describing its known vulnerabilities and describing circumstances where its use would be inappropriate.
Are you serious? About the dentistry thing? I'm not sure if I'm impressed or scared.
Once again, for "general maintenance" -- which is to say, brushing my teeth and such. As I intended to imply, recompiling a stock kernel is about on the same difficulty level as the aforementioned general-maintenance dental functions.:)
If you need a good source control system, try arch... it's got some really nifty features ala BitKeeper (changeset orientation, disconnected operation, distributed repository support, ultra-cheap/easy branches), but without the evil licensing or the repository-corrupting tendencies that BK had when I last used it (a few years ago).
My guess is that the QVCS docs won't port directly to Debian without some... err... creative interpretation; YMMV. That said, the Debian packages are *mostly* pretty good, excluding those times when they're not; Debian includes packages for qmail, vmailmgr, courier and squirrelmail in unstable (and possibly other branches, I haven't checked).
Slashdot Mirror
Because Python isn't Java. All of the Python apps I have used were fast and worked well. I can't say the same of Java apps. :( ...and the really funny thing is this is mostly true even though a Java JIT compiler is typically quite a lot faster than the Python interpreter when benchmarked running equivalent code.
(Part of what this implies is that Java code really can be quite a lot faster than Python when it's written right using libraries that don't suck -- but the "libraries that don't suck" bit is important; a Python/GTK interface will be faster than Java/Swing any day, even though the Python interpreter is slower to process the program logic).
Of course not -- I'm merely rejecting your premise that all actions with negative consequences are "evil", and consequently your presumption that testing for negative consequences is implicitly testing for evil.
BTW a great objective test for the presence of evil- Dead Human Bodies.
Counterexample: Old age isn't evil -- being an impersonal force it cannot be -- yet it results in dead bodies. Further, there are a great many things which are evil but which don't result in dead bodies. Theft, fraud... and so forth. Keep looking for that objective test.
No, I may be a fool. Therefore I must be careful. Neither are you unless you have acess to all relavent information and know when you have it all. Since some of the information I need to make a good decision will not exist until after the fact, at best I can only act in my percieved best interests.
Yes, of *course* you're being careful -- being careful when you need to is part of being competant, after all.
That said: You obviously have some kind of ruleset (perhaps one based in large part on the moral positions you're presently defending) which permits you to choose actions which you believe to be less disadvantageous than others -- if you had no such thing, your actions would be nothing but arbitrary. You believe that the actions you choose using this ruleset are, by and large, better than the actions you would choose without using this ruleset -- otherwise you wouldn't choose to use the ruleset, and certainly wouldn't be defending it here. (Remember, I didn't say "perfect", I said "competant").
No, I have and will be harmed by the unintended consequences of others actions. Unless you believe that my car getting trashed by a drunk while I was asleep in my bed is somehow my fault? The drunk had to mow down a fence and an apple tree to get to my car BTW.
*sigh*. I didn't ask if you would be subject to no damages, I asked if you would be subject to no additional damages -- that is to say, any damages which you are not already subject to. The drunk trashing your car, for instance, is obviously a form of damage to which you are already subject. By going off on this tangeant, you have completely avoided answering the question I asked.
You presume that no one is a fool. Incidently, fools do not believe they are fools. Lastly, many times a fool in his own cause can be wise in anothers cause. We should look out for each other.
I presume that a fool who does himself harm will be less of a fool next time around, as will his children and/or others who have the wisdom to learn from his mistake. I also presume that self-determination -- freedom, in short -- is a higher standard to uphold than wellbeing, with all the consequences this implies.
Are you a fool?
Heh. Well, I'm pretty damn foolish from time to time, but I wouldn't describe myself as a fool -- for that matter, I can't think of any people I would describe as unqualified fools. I can think of folks who are a bit short on common sense, or not particularly skilled in their profession, or helpless when faced with what should be a fairly simple automotive, plumbing or wiring problem -- and indeed, I fit into at least one of the above categories -- but even so, I'm disinclined to think that meeting any of these criteria makes an individual someone whom I'd without qualifications consider "a fool" except within the narrow scope to which that limitation applies.
If you define "a fool" as "anyone who makes poor decisions", then I no doubt am that, and frequently -- as, in that case, are we all. A definition so loose that it applies to anyone, however, is a definition so loose as to be meaningless; in this case, therefore, I prefer to use one which is substantially more restrictive.
First, consider the allegation. Copyrighted code is part of the Linux kernel. What if it's true? Shouldn't the copyright holder get its day in court?
Sure, the copyright holder should get their day in court. The copyright holder should also try to mitigate damages -- indeed, their ability to get the court's assistance will be severely limited if they don't. As of yet, they refuse to let us know exactly what we can do to reduce their damages (by refusing to identify the infringing code for removal). This is not consistant with behaviour with this allegation being true.
And consider the lawyer...David Boies. In case you've forgotten what what he's done...
No, I haven't forgotten about Boies losing every high-profile case he's been involved in -- and for good reason. Have you read SCO's legal briefs? They're downright amateurish next to IBM's filings.
Do you think the diligent researchers on Wall Street are confused by McBride's proclamations? No, they do research.
I'll believe that their research is effective when my own starts backing it up.
You're presuming the conclusion: My definition only "agrees to ignore" harms done with consent from the so-called victim -- which, as you may recall, I argue to be no problem at all!
The "natural law" argument falls flat on its face: Nobody argues about the law of gravity, because such a thing can be objectively tested. I've yet to see an objective test for "evil", however -- perhaps a heat sensor combined with measurement of ambient sulfur levels?
As for the VERY BAD THINGS you so ominously warn of: If those VERY BAD THINGS happen with the consent of those to whom they occur, so be it! That said, if the argument which you so decried as "sophistry" is correct, and individuals will indeed more closely follow a moral code whose strictures they are in fact capable of complying with, then perhaps the number of VERY BAD THINGS which happen under my preferred code is fewer than those which happen under yours.
The responsibility of every human being is to realize what they do and act to mitigate it and to try to do some positive good.
You know, if I try at it for a bit, I'm sure I can come up with a broad, sweeping, mom-and-apple-pie sounding statement about things that other people ought to do, too. That's not to say that this statement actually creates some moral obligation on the part of others, or would make me a better person for having said it.
The only clear obligation which I see as essential to an effectively functioning society is "first, do no harm". To actively help others -- to do good -- is not a baseline that is expected of every individual (such that any individual who helps others is doing no more than the duty obliged them), but an outstanding act by which the individual who does good can be considered to have gone above and beyond the call of duty.
Once again, this mode of thought is intended to encourage good works; those who go above and beyond are allowed more room for self-praise -- and thus more encouragement to continue their good works -- than those who are merely trying to counterbalance the effects of those unavoidable evils they've committed. Likewise, trying to earn a few more points towards the fulfillment of an unending obligation is not an entirely encouraging situation. Compare the man working to pay off a $5 billion debt to the man working for pay he can keep; the former is far more likely to become unenthused regarding his efforts.
Now, let's try looking at this from another perspective. A few questions (and yes, they're a bit leading):
...because they were able to pick out combinations of optimizations that measuring each optimization individually wouldn't have discovered.
Consider the case in huffbench when
-fmove-all-movables, -freduce-all-givs and -ftracer were discovered to result in great improvements when used with the other GA-selected optimizations -- but not when applied directly to -O3.
Sure, these situations may be specific to the test cases -- but that's not to say his software can't be used to evaluate real-world scenarios as well, such that companies doing a final release compile of their code let a GA churn for a few days to determine the best flags to use.
That's a Good Thing, no?
"without knowing consent" means that the same act can be evil or not depending on the assent of the target.
Absolutely. I think we agree that this in fact the case. If I consent to give you my car, and you take it, then what you've done is appropriate. If you take my car without my consent, what you've done is evil.
Giving you my car may be something that does me harm -- maybe I don't have another ready way to get to work in the morning -- but that's my own lookout; if I decide I want to give you my car (for whatever reason -- maybe you offered me something in return, maybe you needed it more than me and I like helping my friends), then you're doing nothing wrong by taking it.
I did not say the mitigating circumstances make doing an evil thing right. I am saying that regardless of circumstances the act is evil. War is evil; does that mean it's unnecessary, no; does that make it less evil, also no. We human beings can no more avoid doing evil than we can avoid breathing. But we can try.
If I had to come up with one reason I prefer my moral stance, you've made it abundantly clear: By my definition of evil, evil is something one can succesfully avoid. You may go and argue why evil things are unavoidable -- and engage in such actions while knowing them to be evil. Evil, by my definition, is something which can be avoided in full -- and when I don't avoid it, I have no excuse that it's "no more avoidable than breathing".
A less restrictive definition of "evil", thus, can result in an overall reduction in harmful behaviour in the same manner by which laws which the average man can realistically follow are better respected when unaccompanied by laws which he will not.
"without knowing consent."
Okay, you assert that it's relative. I assert that it's not. Show me how "knowing consent" is any more relative than the definition of "destruction of a human being", and I might be impressed. Both have fuzzy areas -- places where you say it's destructive and I say it's not, places where I say consent is given and you say it's not -- but both have very clear, very firm areas where the condition is met or isn't.
[W]hat makes you think that imprisoning a human being is, in any way, shape, or form, anything but an evil act?
Because doing evil is wrong. If you believe that mitigating circumstances ("necessary, prudent" and so forth) can make doing an evil thing right, your moral position is one hell of a lot more relative than mine is.
What part of "harm without knowing consent" is relative? I'd call that a pretty damn black-and-white line... it's just a line that's different than the one you draw.
I'll grant that there are special cases -- but your definition has them too. Imprisoning or fining someone who's committed a crime is harm, and the only "consent" available is the implicit acceptance of possible consequences in commitment of the crime. That same case applies to your definition -- why isn't imprisoning someone judged guilty of a crime *evil*?
Porn is a distorted image of that, and people who build their lives around porn (strippers, prostitutes, pornographers, etc.) ultimatly are destroying themselves and the people around them.
[WARNING: I drift off-topic here. Consider that part of the risk you take in crafting a slippery-slope "porn leads to $FOO" argument -- it's a nice lead-in for me to ramble about $FOO].
People who build their lives around work (coders, designers, sysads, etc.) ultimately are destroying themselves and the people around them.
No, seriously -- I've seen waaay more peoples' relationships destroyed by workaholic tendencies than by porn and all the "alternative lifestyle" bits you associate with it
Granted, there are those whose behaviour is clearly self-destructive -- not merely when viewed through some set of moral guidelines but when taking an objective view of their wellbeing. Typically, though, self-destructive people are self-destructive people -- whether putting themselves into as many bad relationship as they can arrange, ignoring doctors' orders and letting a serious medical condition become life-threatening, or deciding to spend their days gaming rather than looking for work. Just because one can eat oneself' to the point of obesity doesn't mean that food is a bad thing, or even that everyone should be required to hold a 1500-calorie-a-day diet.
A lifestyle element which is destructive in one case may not be so in another. I know some potheads whose lives revolve around their drug -- but I know far more productive members of society who will occasionally partake of a passed-around joint. That alcohol can be used to excess, with exceedingly destructive effects, doesn't mean a one-drink-a-day maximum is always appropriate. Self-destructive people won't be stopped from self-destrution merely by taking away their mechanism, and doing damage to the range of choices available to those who are not self-destructive in the name of protecting those who are is an insult to the intelligence and free will of humanity as a whole.
And hitting on the "alternative lifestyle" bit again: I honestly don't believe that the traditional 1:1 husband/wife closed relationship is the only thing that works, because I've seen too many succesful (long-term, even!) relationships with different foundations at their core. Yup, I've seen unsuccesful alternative relationships as well -- but not grossly out of proportion, and largely involving people with self-destructive tendencies of the sort I mention above.
Okay, all that said, back to answering your points.
I think you have the answer implicit in your own statement. As soon as they test positive for an STD they are no longer working, biweekly tests remember?
Yup. I find this adequate. Every life of business which generates good returns requires either substantial up-front investment or assumed risk; in this case it's the latter. Given that the licensing program requires the individual assuming such risks to be aware of said risks and skilled in minimizing their impact, I find I have few remaining moral objections with regard to assumption by the provider of any level of risk which may remain -- and risk to the customer is quite assuredly lower (by virtue of both the training and the regular tests) than the like risk with a member of the opposite sex picked up at a bar; hence, compared to some of the more popular alternatives, legalized and regulated prostitution is a very safe activity. (WRT risk to the providers: I'm putting on my caloused libertarian exterior here and arguing that someone who, being of legal age, contracts to undertake an activity which poses a risk regarding which said individual is fully informed, assumes said risk in such a way that they've only themselves to blame should the risked danger befall them -- same applies to factory workers risking industrial accidents, typists risking repetitive strain injuries and anyone else whose profession includes connotations of risk. It's cases where a
All I can say is that it can't be worse than what Sun's traditionally made.
I at one point had an opportunity to come in contact with a prototype embedded Sparc chip and reference board. The thing drew *70 watts* of power. For an embedded board.
My old university was recently (okay, it's been a few years now) donated a 64-unit sparc cluster by a national laboratory who was done with it. Last I'd heard they only had 2 units powered up -- they needed to have their (large, heavy-duty) server room refitted with more power and better A/C before they could turn on the rest.
Honestly -- there's no way AMD could do any worse.
Hmm -- you bring up a very important point: In my previous position, I wasn't thinking "internet" (though of course that's the topic of this story) -- none of the children I interact with on a regular basis are old enough to use chat or the like (few of them spent much time in front of a CRT at all, for that matter), and I just stopped thinking about it.
I might be inclined to agree with you that a level of caution makes sense online that is likely to be more counterproductive than useful when applied to real life.
[As a totally off-topic aside, I'm inclined to think that the not-much-time-in-front-of-a-CRT thing is a Good Thing at their present ages -- between 3 and 6 -- though I've done an SDL port of UCB's Logo implementation specifically to have something for my household's 5yo to learn with when she's ready].
Maybe they don't like to give away complete control over their patent licenses? Sounds like it makes perfect (business) sense to me.
It sounds like abuse of a section of law originally created for the protection of tangible inventions to me.
It makes perfect sense for a protocol's documentation to enjoy copyright protection. Patenting the protocol or format itself -- well, if someone died and made me the US legislature, there'd be a quick change of patent office policy on that count.
That 1% may or may not have been correct as original stated, but you've certainly misinterpreted it. It certainly does not say that if a specific child in this specific situation is not molested by his/her immediate family, that child has only 1% (or "likewise very small") of the "normal" risk for being molested. It'd be a reasonable guess that the risk is less than normal, given that the major risk factor in the general population has been eliminated. But that's all it is; a guess. That's not valid deductive reasoning; there are plenty of situations in which that sort of reasoning will get you a conclusion that is not true.
.2 (needless to say I hope this exceedingly pessimistic) and solve for a and c.
a - chance of general-population child being molested
b - chance of child being molested by family
c - chance of child being molested by strangers
a=b+c
c=b*0.01
Now, let's say b is
c = (.2)*0.01 = 0.002
a = (.2)+(.002) = 0.202
But wait, you say! I haven't proven that children not molested by family members aren't at a higher risk from strangers -- a dangerous assumption, until one thinks about it critically and realizes that even if all the risk of (c) were shifted onto the set not covered by (b),(c*.2) is still an absolutely tiny increase -- and any larger increase is not possible because it would increase the total percentage of molested children who were molested by strangers as opposed to family members.
So -- yes, it reasonably follows that any child who is not molested by family members has a very low chance of being molested at all -- low enough, in my view, that teaching children to fear anyone they don't know as a potential bogeyman does far more harm than good.
It's the people who really don't belong at a place that kids need to be scared of, the people who do belong to a place are usually good people there to help.
I'm not sure that automatically distrusting "out-of-place" people is really a good standard either. It's something the intentionally deceptive can and do fake, and it's prone to a high false-positive rate. Perhaps the best solution is to simply establish that only adults known to be trusted by Mommy and Daddy can make requests which would otherwise be rule-breaking (leaving the front of the school before Mommy shows up, for instance, is quite certainly rule-breaking) and, as you say, authority figures are good people to go to for help (but one should still think cautiously when asked to break rules on their say-so).
How effective would this be in real life? I don't know. The 5-year-old in my household has a remarkably good BS-detector for her age, better than that of some adults I know, and a willingness to question requests that don't make sense to her. Hopefully she'd be as quick at figuring out that some stranger (or, more importantly, some not-quite-a-stranger) is leading her on as she is when her family members do the same.
In any event, preparedness for strangers with ill intent is pretty much a non-issue; see the first half of my post. Saying "no" when her friends ask her to do something she knows is wrong... that's a much more real and pressing problem.
Your definition is incomplete. Certainly, humans will learn to seek rewards (rewards being things that cause the "goodness function" to evaluate high) -- but how is it determined what items have a high "goodness function" such that they function as a reward? The goodness function, then, isn't entirely determined *by* rewards; indeed, the goodness function determines what things *are* rewards.
Describing conditioning as the be all and end all of human behaviour is also incomplete -- ones' hunger for food, ones' need for affection, or even traits like curiosity are largely inbuilt -- though I'll certainly grant that conditioning later modifies the associated weightings.
One final bit -- I don't know if I need to spell this out, but might as well. A "goodness function" is called such not because it necessarily makes a human good, but earns its name as it's used to interpret how good one's curent state is versus how good ones' state would be if one just did $FOO. For selfish people this "goodness" function could evaluate high on a course of action that hurts others but results in personal gain, for instance), while folks with higher empathy weightings (however those are established -- I can think of a few plausible guesses but they're just that) more quickly decrease a scenario's goodness rating based on harm done to others.
In summary: Condition certainly affects later behaviour as it modifies the weights used in decisionmaking (including goodness-function evaluation). That said, humans are born with a pre-built goodness function -- if not our brains wouldn't know which stimuli to treat as rewards and to seek for. Coming up with a good equivalent to that inbuilt goodness function is a Hard Problem in artificial intelligence, and coming up with a general solution has a lot of potential to make AI problem-solving techniques we already posess much more useful in the context of building "strong" AI.
To me that's one requisite for intelligence, the ability to play a reasonable game with nothing other than raw rules and some practice.
Quite a lot of AI techniques work this way -- genetic algorithms, neural networks (presuming some iterative training based on review of its actions, etc). The real trick is coming up with a "goodness function" which can be used for training -- once the machine has a function that it can seek to optimize, one can pretty much just let it go and figure out how to actually do the optimization itself given examples to watch and/or practice trying itself. Right now, though, that goodness function pretty much has to be written by hand and is very specific to what the AI in question is trying to do.
I'm fairly sure us humans (presuming, of course, that you're one) have a "goodness function" too -- though it's not a single number but a set of neurochemicals and other brain state -- and that we tend to learn to take actions which maximize this function (via receiving praise from others, satisfying basic needs, curiosities, etc). That said, coming up with a good general purpose one that could be used by a computer is no doubt far easier said than done.
All that said: I spent more time back at university in AI classes than most (and under some very good professors), but I by no means claim to be an expert on the subject; my knowledge is quite heavily weighted towards theoretical as opposed to practical content.
Hmm?
Last I knew, professional prostitutes (the legalized, licensed, Nevada kind) had a *much* lower STD rate than any other population segment -- given that in order to be licensed for their profession they require knowledge on best practices for avoiding contracting STDs and identifying those posessing them, are obliged submit to biweekly blood tests, etc. I'd expect that those involved in (professional) porn would have a similarly low rate, given the similar need for their employers to require regular testing pre-employment.
Finally, remember, we're discussing porn. Not prostitution, not sex clubs, not "alternative lifestyles", porn. I haven't heard of anyone getting an STD from a magazine yet.
So I think in this case you're best off assuming the danger is coming from strangers. Or maybe from other close acquaintances. Or from anyone but the immediate family, because they're the ones you're talking to.
Or maybe you're better off assuming that making your kids paranoid about anyone whom they don't know yet is a bigger threat to their overall wellbeing than that 1% (or likewise very small) remaining risk.
Are you the same kind of person who won't take your child on an airline because planes sometimes crash? There's a notion of "acceptable risk"; I consider talking to strangers (not blindly obeying strangers, but talking to strangers) well within that area.
4, typically -- it's the year before kindergarten, though there are occasionally side cases where one might end up starting preschool early and staying in it longer (typically because public schools often won't accept children a day under 5 into kindergarten, even if they're otherwise ready).
If you want to know why pornography is evil then just think about it from the point of view of the photographic objects. Imagine that was you, how degraded would you feel if you were the subject of those pictures. What kind of psycological defences would you have to put up just to get up in the morning? Kind of makes you understand why so many people in the porn industry are addicts doesn't it?
What makes you expect that said people would feel "degraded"? Your values, of course.
People with a different value set may have a different perspective.
I can't speak as to porn, but a few of my friends once worked as dancers in strip clubs (and another spent some time working as a nude model for the local university's art department). I can't say any of them ever saw their own work as degrading, despite the dissenting opinions of others.
One could even go as far as to argue that if it's the "degredation" meme that causes pornography to be harmful, then it's those (such as yourself) who perpetuate that meme who are responsible for the relevant harm.
Then your distribution will almost invariably come with the drivers bundled as a module that can be automatically loaded when the hardware is detected.
If you've got a *really* odd piece of hardware, then you might need to do a bit of trickery (download/compile a kernel module, or modify the code of an existing driver to recognize a new card using the same chipset), but that's rare -- and if you're buying hardware that's that unusual, you're typically an enthusiast, which means you should be competant to download/compile a 3rd-party kernel module yourself.
Seen Fedora's printer setup? Click on the print queue icon, click on "add printer", select the printer from the dialog box, and it DTRT.
Granted, this is a *little* more complex than "plug in and go"... but not much so at all.
Please -- nobody's innovated in word processors or spreadsheets for years. What "innovative" new versions are, to 99% of users, is cost centers. Open source software shares development costs, and so reduces the costs of the businesses using it.
Being able to build a business that's doing something truly innovative while using zero-cost, open source software as building blocks means those engaged in true innovation are spending less for the privilege of being able to do business -- which means there's more money left to fund real innovation rather than throwing money away on licenses for software which reached the plateau of its usefulness years ago.
Yes, I know, IHBT; I'll be sure to HAND.
As a redhat stock holder, I'm extremely disappointed with the recent changes redhat has made, when they IPO'd and I got the stock as soon I could, not because I particularly liked redhat, I don't, I've nver used it as my primary OS and I prefer many other flavors of Linux over it; however I saw in redhat the best chance at a real competitor to microsoft; and now they are not.
How so?
They still sell and support a workstation product (Red Hat WS), and while they take the official position that it isn't ready for general (ie. home) desktop use, I haven't seen any statements to the effect that they don't still believe in Linux on the *corporate* desktop.
Just because they've made "Red Hat Linux" into "Fedora Linux" and stopped selling support for it such that their lowest-end supported desktop product is Red Hat WS, doesn't mean they're entirely out of the race.
True, but by understanding something one can understand where its weaknesses are, and one may take actions to minimize the likelihood of exploitation of these weaknesses (or maximize the likelihood of detection of said exploitation after-the-fact).
In the case of automated voting systems without a voter-verifiable paper trail, the ability to detect prior tampering is minimal; in a great many cases, detecting tampering after-the-fact is impossible. Likewise, because the system has several components which are difficult to oversee, there are a number of places where the results may be unduely influenced before the election itself occurs (ie. subtle undocumented features in the code).
It is because understanding is necessary to implement safeguards before-the-fact and audits after-the-fact that said understanding is such an important element -- not merely for understanding's own sake.
I say this, by the way, as the lead developer (a few years back) of an online voting system used for the elections of a public university in northern California and an associated $14M corporation. My system was delivered with detailed documentation describing its known vulnerabilities and describing circumstances where its use would be inappropriate.
Are you serious? About the dentistry thing? I'm not sure if I'm impressed or scared.
:)
Once again, for "general maintenance" -- which is to say, brushing my teeth and such. As I intended to imply, recompiling a stock kernel is about on the same difficulty level as the aforementioned general-maintenance dental functions.
If you need a good source control system, try arch... it's got some really nifty features ala BitKeeper (changeset orientation, disconnected operation, distributed repository support, ultra-cheap/easy branches), but without the evil licensing or the repository-corrupting tendencies that BK had when I last used it (a few years ago).
My guess is that the QVCS docs won't port directly to Debian without some... err... creative interpretation; YMMV. That said, the Debian packages are *mostly* pretty good, excluding those times when they're not; Debian includes packages for qmail, vmailmgr, courier and squirrelmail in unstable (and possibly other branches, I haven't checked).