I'm all about stiffer legislative penalties and more consumer control over the listing of their information. But I'm ALSO for the market improving its filtering, and I don't think it requires charging, and I don't think there's a good way to charge.
The key point that IS true is that spam will exist as long as stupid people buy stuff from spam in sufficient quantity. Short of improving education and waiting 30 years, the only solution is to keep the spam from getting to most users.
Here's what we really need: 1) Improved server-client spam communication. This is whatwe don't have: 1A. An open standard "spam points" header system - so that IF your receiving mail server has a "ranking" filter that gives a point score to emails it can pass an email to your mail client but tell the mail client "this is 75% spam" This lets you run advanced server-maintained filters but make user-specific decisions about how "strictly" to interpret them. Mail clients already by default ignore extra headers, so all I'm suggesting is that the server filters need to add it in a standard way for the clients to use if they so choose. For bonus points, it should have the main header and "this is 90% from a misDNSed mail server." etc. Mail clients should by default have a fairly strict checking, because the users who don't know how to set it are the same users who are likely to be phished.
1B. An open standard for the mail client telling the receiving mail server "my user thinks message 232432432 was spam" Obviously, users are wrong sometimes, but this would let users who find spam automatically report it to automatically improve their server-side filters. Many servers will ignore this feature, which is fine. But as long as all the clients try in the same way, at least it will be easy for a server to account for it.
2. SPF & friends - letting at least some servers prove who they are. This exists, although of course adoption could be better. If sender and receiver have SPF, people can't pretend to be you anymore.
3. Good, tracking weighted server side filters. These already exist. It should let through email that fails only a couple of tests, but should assign a point value based on many factors. Note that we don't need to force everyone to do this, just a the few biggest targets.
3A. They should take into account use of SPF, whether the maildomain has a valid DNS and some valid RDNS, whether the netblock is commonly used for spam, how long the domain has been active and normal content filtering of the message & content. Netcraft's phishing list, etc.
You can safely use things like the RBL this way, as long as you only assign a limited weight to them. In plain English, being on the RBL doesn't mean you're a spammer, but it does make it somewhat more likely. You only reject messages that have a lot of clues.
3B. It should _also_ take into account the current volume of identical or nearly identical messages. I suspect that a worldwide system for IMMEDIATELY sharing a hash of messages that occur in large volume would be helpful; I know some private companies already use a similar system.
3C. It should _also_ take into account the past history of the IP, rDNS domain, and netblock. This includes the past history of the stuff above and also the past history of user reports as mentioned in 1B.
3D. A valid tactic for certain kinds of messages is to slow down the processing of them. So if you get something you think is probably spam, you can delay a few minutes and see if its score gets better or worse. It will get worse, for instance, if you find you have a lot of identical messages, but that was the first one.
3E. Good servers should have a user-specifiable point cutoff.
Essentially anything will techically meet your needs, but somethings will make your life harder and some easier. Not to start a flame, but I think your description better suits Linux. If you want to try a BSD I'd try FreeBSD before OpenBSD.
The biggest difference is really in philosophy, because most really important things that any free OS has get shared.
Traditionally: OpenBSD is ridiculously secure, sometimes at the cost of speed or similarity of structure. It is ideal for important routers and servers. NetBSD is ridiculously compatible, but probably not better unless you're trying to run it on your toaster - which is probably supports. FreeBSD is not quite as secure as OpenBSD, but generally faster and more compatible. It is probably the right choice for many serious server applications that may value speed or ease over a usually small increase in security.
Linux has a lot more people working on desktop/workstation applications and a lot more popular mindshare. The most important effect is that there are a lot more tutorials on how to do every little thing using the standard linux tools. BSD can run linux binaries most of the time. But a lot of times something you want to use might not work - or it might work if you know how to make one tiny fix, but it won't be mentioned anywhere because the developer forgot BSD existed.
So from your description I would use Linux because it is vastly more popular as a desktop.
You're right that these are i386 limitations (although not all BIOS's are actually limited like this) But you're overstating how big of a limitation it is.
I think the way you said that is misleading, because it sounds like you're saying "OpenBSD must be installed in a primary DOS partition to be bootable"
That is definitely not true. OpenBSD does not necessarily have to touch a primary partition to be bootable.
The limitation is really "SOMETHING has to pick what boots" usually (but not always) the i386 BIOS is pretty dumb about this, so something somewhere has to be on a primary partition.
One of the primary partitions on the first drive must be marked active, and that partition must contain a bootable OS OR boot loader that can find your OS - but that's trivial these days. - THE BOOT LOADER DOESN'T HAVE TO MATCH YOUR OS -
So you could have OpenBSD in a logical partition and have a linux boot loader in a primary partition that lets you select on boot which partition - primary or logical, on any drive - gets booted. You could also have this selection be automatic. You could have it boot OpenBSD if it's Thursday, if you wanted. Except for that last part, this is all very, very common freeware.
I had a dealership make me a new set of keys for about $6, without the car present, knowing only the VIN and the name of the original owner - who was neither me nor present. (It was a girlfriend at the time) They did have proof that a girl's voice at one point called them, but no identity verification of her.
Whoever got treated like a thief needs to act less suspicious, or call better locksmiths/dealerships.
I _was_ authorized to have that key made, but they had no way of knowing.
I believe that by "buy a new box" he meant if you lost "everything" you probably have to buy a new computer. Or at least a new harddrive. I wouldn't trust a harddrive that lost everything once already... unless maybe it was a virus or the like.
I NEVER said you could use a firearm SAFELY only from watching TV. I said you could cause significant effect - the likelyhood is that you can manage to cause damage to someone. And I specifically said you'd need to be at fairly close range.
More importantly, I thought I was very clearly negative about the wisdom of USING point and click interfaces without any knowledge and how it has unintended and negative consequences. Perhaps I was too subtle. I suppose I shouldn't expect people to follow humor on/. I would never recommend putting a firearm in the unsupervised hands of someone who's only seen them on TV.
To be clearer: I was specifically insulting the untrained use of firearms at the same time I was insulting the creation of any interface designed to give a lot of flexible power to user without training them.
I will happily admit that I was using nitpickable "popular" vocabulary, where "cock" is any possibly intimidating motion that prepares the weapon for firing, usually expressed in the movies as pulling back the slide. While I don't mind your narrower definition of "cock" and understand that it is the way a gun purist would use that word, you didn't present a good short replacement for my usage, and I think that in the context of _movies_ that's what cock means for any weapon with a slide.
This popular confusion probably stems from pulling the slide being synonymous with cocking for internal hammer weapons. AS YOU POINT OUT, the movies usually use internal hammer weapons. In the CONTEXT I was talking about, they are synonymous. Of course, I should also have known better than to expect context to be considered on/. either.
While I'll happily believe you may be more experienced than I am, that doesn't mean I'm either untrained or stupid. Your post is fairly insulting coming from someone making sweeping generalizations about me because you can't follow humor already metamoderated as such or the use of a word within the context that you later describe.
I have no combat experience, but I've certainly fired a few dozen different weapons and have been trained both privately and by the US Army. These weapons included gas operated and non gas operated semiautomatic pistols, bolt action, level action, semiautomatic and automatic rifles, breakbarrel and pump action shotguns and SA and SA/DA revolvers. Excepting the shotguns, each category was in a variety of calibers, and most of them I personally disassembled and reassembled to clean.
Regarding ammunition selection: Glazer rounds are too complicated for movies and TV - scoring the front of the round with a big knife is usually the end of it. But I've certainly seen shot used moderately often in movies. I'll completely agree that I've rarely seen characters consider accidentally hitting something through the wall.
I'd accept that the crossbow might've been the first ranged weapon designed with that idea in mind, but I sustain the technology wasn't there to accomplish the goal of untrained killing until the cartidge.
As someone else pointed out, the amount of effort to cock a crossbow isn't necessarily tremendous - but it's definitely there. As I pointed out, using an underling isn't a technological P&C.
And as you point out, it only takes about a week of training for a thickheaded soldier to use a crossbow. To use a firearm at close range only requires watching TV - people successfully kill all the time with only that much training.
(This is not to dismiss the very real skills someone with mastery of firearms has, but they aren't required to be very deadly - at least at short range.)
To address another nit somebody might pick, I'd agree that both crossbows and bows were easier and more effective than early firearms. That was not true for an untrained user by the time of common cartidges.
To answer another uncle-post, with a dual action cartidge revolver, if you put the cartidges in the little holes and close it, all you need to do is pull the trigger.
And when you see somebody cock a gun in a movie, it's almost always stupid. Practicing proper gun safety generally means not having a round in the chamber. So you have to cock a semiautomatic or automatic weapon. Once. And you'd do this well before you entered combat, if you knew it was coming. If you cock your weapon after that, you're ejecting a good round from the chamber - wasting ammunition and making a fool of yourself. If a character walks up to another character and cocks his gun, either the director or the character is a fool.
While the firing of crossbows is certainly point and click, the loading and preparation is not. And my understanding is it's not reasonable to leave cocked crossbows lying around.
I usually take "point and click interface" to mean "interface that somebody probably smart put a lot of work into so that any unqualified moron can make do more or less what they want to great effect, although the user may not have considered or understand the ramifications of their actions"
That is why I usually consider cartidge firearms to be the original point and click interface. Sometimes I limit it to semiautomatic weapons and dual action revolvers, because you can click over and over with effect and without thought.
I respect your opinion, though, even though I think you're on a little bit of a slippery slope. Of course, the most effective point and click device is always a well-trained underling... but I was limited myself to technology.
I suppose it's possible I'm underinformed, but I believe the "BSD subsystem" of OSX is not compiled "into the kernel" and is entirely a compatibility layer on top of it.
I suspect this is exactly how to never violate the microkernel design and still have BSD compat.
I followed both your links, and they're wrong. Wrong enough to sound like vote-FUD to me.
IRV always gets the two most deserving candidates last and always picks the candidate most people this is somewhat positive. IRV also never requires any strategic voting - you just vote for everyone you want in the order you want on your ballot.
I think the author got bogged down in the details of how to count for IRV and got confused. Here's another way to think of it:
IRV will always pick the candidate the most people this is somewhat positive. This is because IRV will almost always pick the candidate that appears on the largest _number of ballots_. The order of votes only matters in elections where the top 2 candidates appeared on many of the same ballots as _each other_.
As long as you can make enough votes on the same ballot, this absolutely eliminates the spoiler effect - voters can always vote for both. MUCH more importantly to me, it eliminates the PERCIEVED spoiler effect where everbody doesn't vote for a third party candidate for fear of wasting their vote.
I'll give only one more short rebuttal detail: You have an example where you say the Republicans might loose an election to a Democrat because of a powerful Libertarian third party. And that certainly could happen. But if more people give higher votes to both the Democrat AND Libertarian parties, that Republican candidate surely deserves to lose.
Have you tried perhaps running some simulations to try to show what you mean?
You're quite correct, my apologies - I paid attention to the description more than the title. Thanks.
Also, to the post that said the biggest problems are voter apathy and short term votes: I think voter apathy without as much corporate manipulation wouldn't be nearly as meaningful. Similarly, I think most of the "short term voting" that the people do is because of what candidates are fed to them by the media.
More importantly, I was focussing on problems that have possible legislative solutions.
I'm not saying that all exit polls can perfectly predict the results or anything. I'm only saying that in 2004 the exit polls were dead on for most places that didn't have Diebold, and off where they DID have Diebold.
Some randomness would be totally reasonable, although I believe they've been very accurate recently. The pattern was not - it was a consistent deviation in Diebold battleground states.
Illinois, for instance, didn't have Diebold, has a largely democratic metro and largely republican rural area. It had a very high correlation between exit polls and paper vote results.
Lots of "opt-in" email isn't really opt-in. It means things like "somebody once didn't uncheck a borderline invisible box on a page when we told them they could win an ipod". And when we send them a message that says to unsubscribe it only unsubscribes them from THAT mailing list. Often it means "we spam people and call it opt-in to make you think it's not spam, but we're lying"
I have no idea about this particular company, of course.
I'll admit to only having read the abstract, but I think they're missing the point. They compared the tendency to vote Republican with having touchscreens. I'd compare the tendency to have the vote say Republican and the exit poll say Democrat with having particular brands of touch screens. The paper-ballot states had dead-on exit polls; the hotly contested states with Diebold had very large pro-Republican variations. To me, that's the killer info. [Note: I don't think this needed to be a large or powerful conspiracy, I'm more than willing to believe it was a "Lone Gunman" ]
More on the parent's topic, here (US) we usually call that "IRV" - Instant Runoff Voting - and we're using it in some local elections. http://www.fairvote.org/index.php?page=19
And in response to one of the sibling posts, I strongly believe it does make a difference. Not in how much somebody can "game the system", but on how much the two parties matter - it gives a mostly fair shake to a third party candidate. Politicians here vote along party lines with reckless disregard to what they think about issues - like in the recent Bolton stuff. Because the parties have all the control.
I'd rate the partisan stranglehold as the top problem in US politics today.
I'd rate the elimination of most journalistic integrity from the popular media second.
I'd rate the ability of corporations to outvote citizens third. This is partly weak campaign finance laws and partly citizen apathy.
I believe that if we fixed these three problems most of the details would start to fix themselves.
I believe there are some flaws in the experimental details.
More importantly, this is a clearcut case of finding only what you were already looking for and not looking for other options. Experimental results in psychology often don't represent the underlying truth the experimenter presents in the summary.
Personally, I think the moral of this experiment is much more likely to be
"many people are A$$holes if they think they're guaranteed to get away with it. This is much more true if they don't know the victim."
Put that way, I'm surprised it's only 65%. And it completely explains
The experiment depends heavily on the flawed assumption that "complaining about something" = "tendency to not do it" A presumption that is to me obviously not true - think about the stereotypical yelling marriage... There have been a variety of studies demonstrating that people's verbal behavior are often - even usually - far removed from their actions.
At best the experiment demonstrated that: 1) Most people are well trained to _give the appearance_ of not wanting to violate societal norms. 2) Most people will violate the norms _as_ they complain about them for _some reason_ especially if they can believably be told that consequences will be absent. Those reasons might be: a)authority figures. b) sadism if consequences are percieved to be absent. c) curiosity when presented with a situation where consequences are absent.
It is completely unclear from the experiment what the impact of these various factors is.
As I understand it, Apple released all of their changes - enough that you can completely build Safari from that source. What they didn't do was spend extra time making it easy to merge back into KHTML.
From what I know of Safari and Konq from a user point of view they ought to be porting Safari to other platforms more than they ought to be trying to pick and choose each patch.
This Dish thing - it depends on details of how linked they are. Details I certainly don't know at this time.
If there was an infrastructure in for parts for mice in single quantities that was self-sufficient (considering the very low quantities) such parts would each individually cost almost as much - or more - as a whole new mouse.
Generally in a free economy if this WASN'T true someone would sell compatible parts, offer a service of replacing them or buyback your broken mouse for reconditioning.
If performance is your goal, I believe there could be significant benefits to using the same drive. If the integrity and availability of your data are most important to you it should be stored on redundant drives that are ideally from different manufacturers.
This latter part is what the linux High Availability Howto says, also.
While I appreciate the "esc" key - and would mod you up for mentioning it - it doesn't solve the problem. Because if you click "next" it starts all over again. So it's at least an extra keystroke on every single page load.
The problem isn't that you can STOP it from reloading, it's that I want to SUPPRESS it from having automatic reload. I want a setting where it loads a page, and when it finishes it automatically stops the page.
If you click the stop icon, then "next" it doesn't hold the "stop" from the previous screen. Which is incredibly annoying. You need to click "slow" several times to make it useable.
Or, better yet, I need a firefox extension that disables meta-refresh but lets you activate it manually for certain pages. Webshots is terrible about this.
The previous cousin post had it right - a terrible interface designed to force unclickable ad impressions.
Although if they'd come out with a second button on a powerbook it'd be a lot better.
Apple is NEAR the top in laptop construction. Not as tough as a real Toughbook and not as nice as a Thinkpad, but better than everything else. And usually with better battery life because the architecture is more conservative.
However being able to use OSX is more important than any of those details. Anyone who hasn't seriously tried using OSX for a long period should definitely try it.
Also, of course, IBM doesn't make the Thinkpads anymore...
Slashdot Mirror
I'm all about stiffer legislative penalties and more consumer control over the listing of their information. But I'm ALSO for the market improving its filtering, and I don't think it requires charging, and I don't think there's a good way to charge.
The key point that IS true is that spam will exist as long as stupid people buy stuff from spam in sufficient quantity. Short of improving education and waiting 30 years, the only solution is to keep the spam from getting to most users.
Here's what we really need:
1) Improved server-client spam communication. This is whatwe don't have:
1A. An open standard "spam points" header system - so that IF your receiving mail server has a "ranking" filter that gives a point score to emails it can pass an email to your mail client but tell the mail client "this is 75% spam" This lets you run advanced server-maintained filters but make user-specific decisions about how "strictly" to interpret them. Mail clients already by default ignore extra headers, so all I'm suggesting is that the server filters need to add it in a standard way for the clients to use if they so choose. For bonus points, it should have the main header and "this is 90% from a misDNSed mail server." etc. Mail clients should by default have a fairly strict checking, because the users who don't know how to set it are the same users who are likely to be phished.
1B. An open standard for the mail client telling the receiving mail server "my user thinks message 232432432 was spam" Obviously, users are wrong sometimes, but this would let users who find spam automatically report it to automatically improve their server-side filters. Many servers will ignore this feature, which is fine. But as long as all the clients try in the same way, at least it will be easy for a server to account for it.
2. SPF & friends - letting at least some servers prove who they are. This exists, although of course adoption could be better. If sender and receiver have SPF, people can't pretend to be you anymore.
3. Good, tracking weighted server side filters. These already exist. It should let through email that fails only a couple of tests, but should assign a point value based on many factors. Note that we don't need to force everyone to do this, just a the few biggest targets.
3A. They should take into account use of SPF, whether the maildomain has a valid DNS and some valid RDNS, whether the netblock is commonly used for spam, how long the domain has been active and normal content filtering of the message & content. Netcraft's phishing list, etc.
You can safely use things like the RBL this way, as long as you only assign a limited weight to them. In plain English, being on the RBL doesn't mean you're a spammer, but it does make it somewhat more likely. You only reject messages that have a lot of clues.
3B. It should _also_ take into account the current volume of identical or nearly identical messages. I suspect that a worldwide system for IMMEDIATELY sharing a hash of messages that occur in large volume would be helpful; I know some private companies already use a similar system.
3C. It should _also_ take into account the past history of the IP, rDNS domain, and netblock. This includes the past history of the stuff above and also the past history of user reports as mentioned in 1B.
3D. A valid tactic for certain kinds of messages is to slow down the processing of them. So if you get something you think is probably spam, you can delay a few minutes and see if its score gets better or worse. It will get worse, for instance, if you find you have a lot of identical messages, but that was the first one.
3E. Good servers should have a user-specifiable point cutoff.
Essentially anything will techically meet your needs, but somethings will make your life harder and some easier. Not to start a flame, but I think your description better suits Linux. If you want to try a BSD I'd try FreeBSD before OpenBSD.
The biggest difference is really in philosophy, because most really important things that any free OS has get shared.
Traditionally:
OpenBSD is ridiculously secure, sometimes at the cost of speed or similarity of structure. It is ideal for important routers and servers.
NetBSD is ridiculously compatible, but probably not better unless you're trying to run it on your toaster - which is probably supports.
FreeBSD is not quite as secure as OpenBSD, but generally faster and more compatible. It is probably the right choice for many serious server applications that may value speed or ease over a usually small increase in security.
Linux has a lot more people working on desktop/workstation applications and a lot more popular mindshare. The most important effect is that there are a lot more tutorials on how to do every little thing using the standard linux tools. BSD can run linux binaries most of the time. But a lot of times something you want to use might not work - or it might work if you know how to make one tiny fix, but it won't be mentioned anywhere because the developer forgot BSD existed.
So from your description I would use Linux because it is vastly more popular as a desktop.
You're right that these are i386 limitations (although not all BIOS's are actually limited like this) But you're overstating how big of a limitation it is.
I think the way you said that is misleading, because it sounds like you're saying "OpenBSD must be installed in a primary DOS partition to be bootable"
That is definitely not true. OpenBSD does not necessarily have to touch a primary partition to be bootable.
The limitation is really "SOMETHING has to pick what boots" usually (but not always) the i386 BIOS is pretty dumb about this, so something somewhere has to be on a primary partition.
One of the primary partitions on the first drive must be marked active, and that partition must contain a bootable OS OR boot loader that can find your OS - but that's trivial these days. - THE BOOT LOADER DOESN'T HAVE TO MATCH YOUR OS -
So you could have OpenBSD in a logical partition and have a linux boot loader in a primary partition that lets you select on boot which partition - primary or logical, on any drive - gets booted. You could also have this selection be automatic. You could have it boot OpenBSD if it's Thursday, if you wanted. Except for that last part, this is all very, very common freeware.
http://pclt.cis.yale.edu/pclt/BOOT/PARTITIO.HTM
http://www.theregister.co.uk/2002/05/16/gummi_bear s_defeat_fingerprint_sensors/
I had a dealership make me a new set of keys for about $6, without the car present, knowing only the VIN and the name of the original owner - who was neither me nor present. (It was a girlfriend at the time) They did have proof that a girl's voice at one point called them, but no identity verification of her.
Whoever got treated like a thief needs to act less suspicious, or call better locksmiths/dealerships.
I _was_ authorized to have that key made, but they had no way of knowing.
I believe that by "buy a new box" he meant if you lost "everything" you probably have to buy a new computer. Or at least a new harddrive. I wouldn't trust a harddrive that lost everything once already... unless maybe it was a virus or the like.
I NEVER said you could use a firearm SAFELY only from watching TV. I said you could cause significant effect - the likelyhood is that you can manage to cause damage to someone. And I specifically said you'd need to be at fairly close range.
/. I would never recommend putting a firearm in the unsupervised hands of someone who's only seen them on TV.
/. either.
More importantly, I thought I was very clearly negative about the wisdom of USING point and click interfaces without any knowledge and how it has unintended and negative consequences. Perhaps I was too subtle. I suppose I shouldn't expect people to follow humor on
To be clearer: I was specifically insulting the untrained use of firearms at the same time I was insulting the creation of any interface designed to give a lot of flexible power to user without training them.
I will happily admit that I was using nitpickable "popular" vocabulary, where "cock" is any possibly intimidating motion that prepares the weapon for firing, usually expressed in the movies as pulling back the slide. While I don't mind your narrower definition of "cock" and understand that it is the way a gun purist would use that word, you didn't present a good short replacement for my usage, and I think that in the context of _movies_ that's what cock means for any weapon with a slide.
This popular confusion probably stems from pulling the slide being synonymous with cocking for internal hammer weapons. AS YOU POINT OUT, the movies usually use internal hammer weapons. In the CONTEXT I was talking about, they are synonymous. Of course, I should also have known better than to expect context to be considered on
While I'll happily believe you may be more experienced than I am, that doesn't mean I'm either untrained or stupid. Your post is fairly insulting coming from someone making sweeping generalizations about me because you can't follow humor already metamoderated as such or the use of a word within the context that you later describe.
I have no combat experience, but I've certainly fired a few dozen different weapons and have been trained both privately and by the US Army. These weapons included gas operated and non gas operated semiautomatic pistols, bolt action, level action, semiautomatic and automatic rifles, breakbarrel and pump action shotguns and SA and SA/DA revolvers. Excepting the shotguns, each category was in a variety of calibers, and most of them I personally disassembled and reassembled to clean.
Regarding ammunition selection: Glazer rounds are too complicated for movies and TV - scoring the front of the round with a big knife is usually the end of it. But I've certainly seen shot used moderately often in movies. I'll completely agree that I've rarely seen characters consider accidentally hitting something through the wall.
I'd accept that the crossbow might've been the first ranged weapon designed with that idea in mind, but I sustain the technology wasn't there to accomplish the goal of untrained killing until the cartidge.
As someone else pointed out, the amount of effort to cock a crossbow isn't necessarily tremendous - but it's definitely there. As I pointed out, using an underling isn't a technological P&C.
And as you point out, it only takes about a week of training for a thickheaded soldier to use a crossbow. To use a firearm at close range only requires watching TV - people successfully kill all the time with only that much training.
(This is not to dismiss the very real skills someone with mastery of firearms has, but they aren't required to be very deadly - at least at short range.)
To address another nit somebody might pick, I'd agree that both crossbows and bows were easier and more effective than early firearms. That was not true for an untrained user by the time of common cartidges.
To answer another uncle-post, with a dual action cartidge revolver, if you put the cartidges in the little holes and close it, all you need to do is pull the trigger.
And when you see somebody cock a gun in a movie, it's almost always stupid. Practicing proper gun safety generally means not having a round in the chamber. So you have to cock a semiautomatic or automatic weapon. Once. And you'd do this well before you entered combat, if you knew it was coming. If you cock your weapon after that, you're ejecting a good round from the chamber - wasting ammunition and making a fool of yourself. If a character walks up to another character and cocks his gun, either the director or the character is a fool.
While the firing of crossbows is certainly point and click, the loading and preparation is not. And my understanding is it's not reasonable to leave cocked crossbows lying around.
I usually take "point and click interface" to mean "interface that somebody probably smart put a lot of work into so that any unqualified moron can make do more or less what they want to great effect, although the user may not have considered or understand the ramifications of their actions"
That is why I usually consider cartidge firearms to be the original point and click interface. Sometimes I limit it to semiautomatic weapons and dual action revolvers, because you can click over and over with effect and without thought.
I respect your opinion, though, even though I think you're on a little bit of a slippery slope. Of course, the most effective point and click device is always a well-trained underling... but I was limited myself to technology.
Proving once again that I am a nerd.
Firearms: the original point and click interface.
Apparently that's all I have to say, but amazingly it's on topic in this story.
I stand corrected; my apologies.
I suppose it's possible I'm underinformed, but I believe the "BSD subsystem" of OSX is not compiled "into the kernel" and is entirely a compatibility layer on top of it.
I suspect this is exactly how to never violate the microkernel design and still have BSD compat.
I followed both your links, and they're wrong. Wrong enough to sound like vote-FUD to me.
IRV always gets the two most deserving candidates last and always picks the candidate most people this is somewhat positive. IRV also never requires any strategic voting - you just vote for everyone you want in the order you want on your ballot.
I think the author got bogged down in the details of how to count for IRV and got confused. Here's another way to think of it:
IRV will always pick the candidate the most people this is somewhat positive. This is because IRV will almost always pick the candidate that appears on the largest _number of ballots_. The order of votes only matters in elections where the top 2 candidates appeared on many of the same ballots as _each other_.
As long as you can make enough votes on the same ballot, this absolutely eliminates the spoiler effect - voters can always vote for both. MUCH more importantly to me, it eliminates the PERCIEVED spoiler effect where everbody doesn't vote for a third party candidate for fear of wasting their vote.
I'll give only one more short rebuttal detail: You have an example where you say the Republicans might loose an election to a Democrat because of a powerful Libertarian third party. And that certainly could happen. But if more people give higher votes to both the Democrat AND Libertarian parties, that Republican candidate surely deserves to lose.
Have you tried perhaps running some simulations to try to show what you mean?
You're quite correct, my apologies - I paid attention to the description more than the title. Thanks.
Also, to the post that said the biggest problems are voter apathy and short term votes: I think voter apathy without as much corporate manipulation wouldn't be nearly as meaningful. Similarly, I think most of the "short term voting" that the people do is because of what candidates are fed to them by the media.
More importantly, I was focussing on problems that have possible legislative solutions.
I'm not saying that all exit polls can perfectly predict the results or anything. I'm only saying that in 2004 the exit polls were dead on for most places that didn't have Diebold, and off where they DID have Diebold.
Some randomness would be totally reasonable, although I believe they've been very accurate recently. The pattern was not - it was a consistent deviation in Diebold battleground states.
Illinois, for instance, didn't have Diebold, has a largely democratic metro and largely republican rural area. It had a very high correlation between exit polls and paper vote results.
And for the record, I'm not a Democrat.
Lots of "opt-in" email isn't really opt-in. It means things like "somebody once didn't uncheck a borderline invisible box on a page when we told them they could win an ipod". And when we send them a message that says to unsubscribe it only unsubscribes them from THAT mailing list. Often it means "we spam people and call it opt-in to make you think it's not spam, but we're lying"
I have no idea about this particular company, of course.
I'll admit to only having read the abstract, but I think they're missing the point. They compared the tendency to vote Republican with having touchscreens. I'd compare the tendency to have the vote say Republican and the exit poll say Democrat with having particular brands of touch screens. The paper-ballot states had dead-on exit polls; the hotly contested states with Diebold had very large pro-Republican variations. To me, that's the killer info. [Note: I don't think this needed to be a large or powerful conspiracy, I'm more than willing to believe it was a "Lone Gunman" ]
More on the parent's topic, here (US) we usually call that "IRV" - Instant Runoff Voting - and we're using it in some local elections.
http://www.fairvote.org/index.php?page=19
And in response to one of the sibling posts, I strongly believe it does make a difference. Not in how much somebody can "game the system", but on how much the two parties matter - it gives a mostly fair shake to a third party candidate. Politicians here vote along party lines with reckless disregard to what they think about issues - like in the recent Bolton stuff. Because the parties have all the control.
I'd rate the partisan stranglehold as the top problem in US politics today.
I'd rate the elimination of most journalistic integrity from the popular media second.
I'd rate the ability of corporations to outvote citizens third. This is partly weak campaign finance laws and partly citizen apathy.
I believe that if we fixed these three problems most of the details would start to fix themselves.
I think there a basic difference in US laws...
The US trademark and patent laws, as a rule, make no sense anymore.
I believe there are some flaws in the experimental details.
More importantly, this is a clearcut case of finding only what you were already looking for and not looking for other options. Experimental results in psychology often don't represent the underlying truth the experimenter presents in the summary.
Personally, I think the moral of this experiment is much more likely to be
"many people are A$$holes if they think they're guaranteed to get away with it. This is much more true if they don't know the victim."
Put that way, I'm surprised it's only 65%. And it completely explains
The experiment depends heavily on the flawed assumption that "complaining about something" = "tendency to not do it" A presumption that is to me obviously not true - think about the stereotypical yelling marriage... There have been a variety of studies demonstrating that people's verbal behavior are often - even usually - far removed from their actions.
At best the experiment demonstrated that:
1) Most people are well trained to _give the appearance_ of not wanting to violate societal norms.
2) Most people will violate the norms _as_ they complain about them for _some reason_ especially if they can believably be told that consequences will be absent. Those reasons might be:
a)authority figures.
b) sadism if consequences are percieved to be absent.
c) curiosity when presented with a situation where consequences are absent.
It is completely unclear from the experiment what the impact of these various factors is.
As I understand it, Apple released all of their changes - enough that you can completely build Safari from that source. What they didn't do was spend extra time making it easy to merge back into KHTML.
From what I know of Safari and Konq from a user point of view they ought to be porting Safari to other platforms more than they ought to be trying to pick and choose each patch.
This Dish thing - it depends on details of how linked they are. Details I certainly don't know at this time.
If there was an infrastructure in for parts for mice in single quantities that was self-sufficient (considering the very low quantities) such parts would each individually cost almost as much - or more - as a whole new mouse.
Generally in a free economy if this WASN'T true someone would sell compatible parts, offer a service of replacing them or buyback your broken mouse for reconditioning.
If performance is your goal, I believe there could be significant benefits to using the same drive. If the integrity and availability of your data are most important to you it should be stored on redundant drives that are ideally from different manufacturers.
This latter part is what the linux High Availability Howto says, also.
While I appreciate the "esc" key - and would mod you up for mentioning it - it doesn't solve the problem. Because if you click "next" it starts all over again. So it's at least an extra keystroke on every single page load.
The problem isn't that you can STOP it from reloading, it's that I want to SUPPRESS it from having automatic reload. I want a setting where it loads a page, and when it finishes it automatically stops the page.
If you click the stop icon, then "next" it doesn't hold the "stop" from the previous screen. Which is incredibly annoying. You need to click "slow" several times to make it useable.
Or, better yet, I need a firefox extension that disables meta-refresh but lets you activate it manually for certain pages. Webshots is terrible about this.
The previous cousin post had it right - a terrible interface designed to force unclickable ad impressions.
Although if they'd come out with a second button on a powerbook it'd be a lot better.
Apple is NEAR the top in laptop construction. Not as tough as a real Toughbook and not as nice as a Thinkpad, but better than everything else. And usually with better battery life because the architecture is more conservative.
However being able to use OSX is more important than any of those details. Anyone who hasn't seriously tried using OSX for a long period should definitely try it.
Also, of course, IBM doesn't make the Thinkpads anymore...