the experience of others indicates that baysean filters are no longer as effective as they need to be
That's interesting... I thought the proof would work the other way: if I can make it work then doesn't that prove they are effective? I don't know if you feel you are happy with your own spam filtering, but if you can install spamassassin I'd be happy to share my setup with you.
At the server level you're totally right: the battle is all but lost. But I did want to ask about your bayesian setup... I am using spamassassin as well with just the following config:
Each user feeds it ham/spam separately (I've admittedly only a handful of users). After the initial all-manual feeding (which was laborious) I simply feed it spam that gets through and all other mail as ham. I don't know, I'm just trying to help here. Maybe it only works for me, but it works pretty darn well. But yeah, like yours, my server has an amazing amount of spam coming in. If I had to see even 10% of it I'd pull my hair out.
Thanks for the reply. That's more or less what I thought. I didn't mean to offend anyone by implying spam isn't bad... it is bad. But it just seems like the industrious user can all but eliminate it these days. I think it is critical with bayesian classifiers that it is per-user. We have a site-wide implementation at work which sucks and I asked them to turn off (false positives and still too much spam slipping through) so I could manage it myself. Actually, I'm not even sure how a site-wide one could possibly work as intended. Some default configs also take shortcuts like having the filter self-teach, which is stupid, because that means it just learns to repeat mistakes.
My email address is already available online right now in several locations. You can find it in less than a minute, if you want. It has been available online since 1998. In fact I have multiple addresses listed on multiple public sites right now, all forwarding to my inbox. I've never put any effort into protecting it, I give to all sites that I sign up to. If you want to put extra special effort into attacking my inbox, well, I don't know if it would survive... go ahead if you want, but that's not really the point. The point is that my email has been and continues to be publicly available, so I'm not hiding from spam, yet my inbox is under control. So spam is not a problem for me.
Actually, I just checked and in the past 2 days I've received over 4300 spam messages and only 1 got through. The 700+ figure was based on the last time I bothered to check.
Baysean works different for different people. Maybe if it didn't work for you it's because the emails you want are not as different from spam as mine are? More likely is that you didn't train it properly or have the thresholds set right. Beyond that, I don't know what to tell you.
All reasonable points. And if the suggestions in the article helps things, excellent.
On the "time spent dealing" with it: I spent a couple hours back in 2003 figuring out how to install a baysean filter, then a few more hours (spread over several weeks) figuring out the best settings and the best way to "train" it. That was the initial investment, and it is higher than it should be for sure, though I bet there are more user friendly setups available. But now, I literally spend barely a minute a month on it... I just make sure to flag any spam that comes through (2-3 per day) and have everything else go to a "ham" folder when I delete them. A script runs and feeds those two sets to the filter automatically. No further tuning has been necessary.
I mention it all just in case there are people out there still suffering with spam: you can solve it, for all practical purposes.
Why do you think it "works" when your server has to scan and reject 700+ emails/day?
Because I don't have to:) My server has no trouble doing the work, and it doesn't cost me anything.
But I guess you're saying that if spam is out there, then it's still a problem. To me, if everyone who wants to stop it from reaching them can do so with minimum effort... it's just not a big issue any more. Let them spam. Let the morons who buy the stuff buy it. As long as there's a decent means of avoiding it for anyone who wants to. And I believe there is.
Are any of you people still living with spam? Do we really need another solution? I've found that a personally managed baysean filter is plenty good enough. I'm down from 700+ per day to 2-3 per day. I still dislike the fact that spam is out there, but I haven't actually had to deal with it in years. Has this not worked for other people? I mean, I do have to continue to feed the filter, but it's very little work. Nothing wrong with new ideas in the battle, but I thought that for anyone who cared it was already won.
All the tests in the world can only do one thing with string theory: show that we haven't found a way to disprove it yet. All scientific theories are open to being disproven, that is the beauty of science, that is why it is not a religion, as much as religious types would like it to be, and despite the fact that many so-called scientists actually use it as a religion. The best one can hope for is that observation continues to bear out the predictive abilities of the theory. And you can consider a well tested theory as being good enough for general use... but you'll never really know for sure. So, I say get used to it. Revel in the fact that we don't know, but can still make amazingly useful predictions about our world.
Speaking of which: lets say that string theory survives this test. How far away are we from making useful predictions with string therory? That is, ones that are meaningfully more precise than quantum mechanics or general relativity? Last time I read about it they seemed nowhere near such a prediction because of the complexity of the mathematics. It seemed almost hopeless that they'd make predictions. Is that still the case? This test seems to say otherwise, but are the predictions notably different from what quantum mechanics predict?
the items being traded have value outside of the system itself
But do stocks have value "outside the system itself"? Where is "outside the system" of the stock market? I own stock, and I don't doubt that it is part of our functioning economy, but if I walk out of our economy, and try to pass off my paperwork to people who don't take part in our economy, they wouldn't be any more valuable than second-life items. And yes, regulation and legal recourse. Early on I think the idea of owning part of the company was important, but it is rarely the liquidity of a company that drives prices today. I think the stock market works simply because it is so well agreed upon and it spans such a large number of people that there practially is no "outside the system". It's not a real Ponzi scheme, per se, but somehing is weird because the price is dictated by the belief that someone in the future will pay more for the item than you will now, and if you follow that through it is circular, isn't it?
I hate to be a party pooper, but I doubt that something as revolutionary as the Chevy Volt will not be on the market in the next 5 years by any American automaker. I've been casually watching concept cars for a couple decades now, and very little of the dramatic tech makes it into production vehicles. It seems that they're just a way to keep people excited while they churn out the stuff that really sells: big cheap boxes.
On the other hand, I hope I'm wrong and if I am, I'd certainly be a potential customer for such a car, as I own a Prius and would love to see even more dramatic ways to reduce our oil dependency and polution.
Thanks for the well thought out response. I guess I have two points I'd like to respond to...
First, I agree that only I should decide what I'm interested in. I don't want spam to be illegal in the sense that someone else decides what is okay to mail, but rather that it should be illegal to mail me stuff if I've requested that you don't. Because after I request that you don't, it starts becoming harassment. Of course with spam there are so many people doing it that asking each one individually wouldn't do a thing to reduce it, so a centralized list needs to be created. For telemarketing this is done with a do-not-call list, and for snailmail it is done with the DMA opt-out list, both of which I've signed up for, and both of which work reasonably well, though not completely. I also call and politely request removal from any company that sends me catalogs because I did business with them at some point. Overall I've reduced junk intrusion into my life to a reasonable level. I don't think making spam illegal is a complete solution, but I think it could help, and even if it didnt help, it's "right" in the sense that the spammer is abusing the system and should be punished in some way. I think that should be the case even if it wasn't an effective deterrent or if there are other solutions. You harass others, you get smacked.
Second, I don't think it's fair to push the burden of proving the worth of your email should fall on my customers/fans. If they have a question, they should be able to email me with a subject of "question" and not be immediately demoted. Besides, with 700+ spam messages a day, even with perfect subjects I wouldn't be able to sort them in a timely manner without error. Thank the lord for custom trained baysean filtering:)
I guess I shouldn't have made that part of the comment. I am a fan of any viable green energy source, and solar PV and wind certainly qualify. I guess I was aiming that at the people who seem to think that green energy can only be supplemental or is expensive stuff forced into the market by the environmentalist agenda. Geothermal has the hardcore 24/7 power output of a nuclear plant without the radiation issues.
In any case, I'm glad to hear that solar is coming along nicely.
I believe in free speech when the technology used is "pull" instead of "push". Does that make sense? I believe you have the right to say whatever you want to whomever you want if they are willing to listen, but I also have the right to not have to listen. I should be able to say "stop bothering me" and have that respected.
And unfortunately spam is not just an issue of "don't open the message". Many messages don't have helpful subject lines, instead opting for "Re: your mail" or somesuch. And if you have any public interactions, you don't get all your messages from people you already know. I get over 700 spam messages per day, this gets in the way of me doing my responsibilities. I am lucky in that I am tech savvy enough to have set up sophisticated spam filtering (so only 5 or so a day get through), but the original poster is right: it is illegal to call someone's home after you've told them not to. Nearly all direct marketers will stop sending snail mail if you ask (if not, I'm sure there would be a law about that too). Spam is harassment, not free speech.
Well first of all, corporations are only accountable to their shareholders, not to anyone else.
You're right. And Eichmann was only accountable to his superiors but ended up hanged anyways. Sorry for triggering Godwin's Law, but it is worth noting that there is widespread consensus that one is accountable for their actions even if accountability is not explicitly apparent in the system. I think that most of the SEC regulatory hoops and penalties are a not-wholly-unsuccessful effort to include ethics in the economic picture, so it's not true that they can't be accounted for. But it is a bit of a tragedy of the commons though since, as usual, the shortsighted people who misinterpret "corporations are only accountable to their shareholders" to mean "there's no place for ethics in the market" make an expensive pain in the ass for all of us.
Yes, they have a huge advantage in their ready access to both water and volcanic activity. It sounds like the report from the article indicates we've been underusing our own access to these resources.
If that's Cheney's plan, I may have to find a little respect for the man;)
I realize that they have some special circumstances (and I mention this at the end of my post), but the primary ingredients are volcanic activity and water, and they don't have the market cornered on those items:)
In other words, I realize it will take work to more widely deploy it, but I think it's an underexplored idea. And this MIT study seems to be saying just that. So I got excited:)
I'm all for wind and solar and tidal and whatever else works. My point is that these projects are supplemental at best because they don't have continuous guaranteed power output, and I want to cut through anyone who has the perception that green energy is only supplemental. Sorry if I offended you, but there are plenty of people who won't take green energy seriously. Geothermal has the power output profile of a primary energy source, and I want to promote that.
You are right: I am uneducated though, because I was wrong about geothermal usage in Iceland. It's actually only 54% of the total national consumption of primary energy, the rest being petroleum, hydro and coal, in that order. It actually only provides 17% of the electricity, the rest comes from hydro. Most of the geothermal is used for direct heating.
Do you realize that some sidewalks in Reykjavik are heated because they have so much cheap and clean geothermal energy? You have to admit that's pretty neat.
Yes, the hot water runoff is an equal problem in many energy systems, including geothermal and nuclear. The reason geothermal is better is that it doesn't also carry the radioactive disposal and meltdown risk issues. Both of which are still better than coal burning for environmental damage. I think nuclear is pretty good overall, but geothermal is a little better.
I visited Iceland a couple years ago, and I became sold on geothermal. I mean, Iceland is a small country, but they have fairly high power needs per capita because of the cold climate, and they run almost entirely off geothermal, as I understand it. This isn't some apologetic green technology that is decades or more from delivering affordable massive power, like solar, wind, etc. No, this is the real thing: a geothermal plant puts out power at nuclear reactor levels. And these things are clean.
My favorite part of the visit was swimming in the Blue Lagoon... a spa built alongside the runoff from a geothermal power plant. Seriously: you're in the middle of a lava rock field, and boiling hot waste water pours from the power plant into a huge outdoor pool. In the cold air you can nearly cook yourself as you swim closer to the power plant. But it's clean enough to swim in.
There are many criteria that need to be met to build a geothermal power station at a given location, but I think the research and development needed must be far less than for some other technologies, and the end result is completely proven, so the risks are minimal.
My ideal-yet-realistic world features geothermal and nuclear supplementing each other, with the preference towards geothermal.
Simply not true. A creative person may not be completely satisfied with every aspect of what they've created (which is often the impetus for the next creative project), but despite this stuff gets completed all the time without being cut short. As a creative person (developing, music, and film) I've completed many things without someone else telling me when it had to be done. There are a percentage of people like you describe, and yes, they do need someone to just give them a hard deadline. But there are many creative people who know when something has got as good as it's going to get and to mark it "complete" and move on. Don't judge everyone because of a few experiences you've had.
What empowers him? I think it has something to do with the long term effects being fatal to our way of life? If that's not reason enough to say "we will have to make changes" then I don't know what is. Getting angry at the messenger isn't going to help.
Slashdot Mirror
the experience of others indicates that baysean filters are no longer as effective as they need to be
That's interesting... I thought the proof would work the other way: if I can make it work then doesn't that prove they are effective? I don't know if you feel you are happy with your own spam filtering, but if you can install spamassassin I'd be happy to share my setup with you.
Cheers.
At the server level you're totally right: the battle is all but lost. But I did want to ask about your bayesian setup... I am using spamassassin as well with just the following config:
required_hits 5
rewrite_subject 0
bayes_auto_learn 0
score BAYES_00 -2.0
score BAYES_01 -2.0
score BAYES_10 -1.0
score BAYES_20 -1.0
score BAYES_30 -1.0
score BAYES_40 -1.0
score BAYES_44 -1.0
score BAYES_50 5.0
score BAYES_56 5.0
score BAYES_60 5.0
score BAYES_70 5.0
score BAYES_80 6.0
score BAYES_90 6.0
score BAYES_95 7.0
score BAYES_99 7.0
score ALL_TRUSTED 0
Each user feeds it ham/spam separately (I've admittedly only a handful of users). After the initial all-manual feeding (which was laborious) I simply feed it spam that gets through and all other mail as ham. I don't know, I'm just trying to help here. Maybe it only works for me, but it works pretty darn well. But yeah, like yours, my server has an amazing amount of spam coming in. If I had to see even 10% of it I'd pull my hair out.
Cheers.
Thanks for the reply. That's more or less what I thought. I didn't mean to offend anyone by implying spam isn't bad... it is bad. But it just seems like the industrious user can all but eliminate it these days. I think it is critical with bayesian classifiers that it is per-user. We have a site-wide implementation at work which sucks and I asked them to turn off (false positives and still too much spam slipping through) so I could manage it myself. Actually, I'm not even sure how a site-wide one could possibly work as intended. Some default configs also take shortcuts like having the filter self-teach, which is stupid, because that means it just learns to repeat mistakes.
Anyways, as usual, the devil is in the details.
Cheers.
Whoops -- I confused my work email with my personal email there for a sec and mixed up the info. The correct stats are:
4300 spam at work in about 45 days, 2 spam got through.
900 spam at home in about 2 days, 1 spam got through.
It varies, of course, so sorry if my previous numbers were a) wrong or b) outdated. Still, I think my original point stands.
Cheers.
My email address is already available online right now in several locations. You can find it in less than a minute, if you want. It has been available online since 1998. In fact I have multiple addresses listed on multiple public sites right now, all forwarding to my inbox. I've never put any effort into protecting it, I give to all sites that I sign up to. If you want to put extra special effort into attacking my inbox, well, I don't know if it would survive... go ahead if you want, but that's not really the point. The point is that my email has been and continues to be publicly available, so I'm not hiding from spam, yet my inbox is under control. So spam is not a problem for me.
Actually, I just checked and in the past 2 days I've received over 4300 spam messages and only 1 got through. The 700+ figure was based on the last time I bothered to check.
Baysean works different for different people. Maybe if it didn't work for you it's because the emails you want are not as different from spam as mine are? More likely is that you didn't train it properly or have the thresholds set right. Beyond that, I don't know what to tell you.
Cheers.
All reasonable points. And if the suggestions in the article helps things, excellent.
On the "time spent dealing" with it: I spent a couple hours back in 2003 figuring out how to install a baysean filter, then a few more hours (spread over several weeks) figuring out the best settings and the best way to "train" it. That was the initial investment, and it is higher than it should be for sure, though I bet there are more user friendly setups available. But now, I literally spend barely a minute a month on it... I just make sure to flag any spam that comes through (2-3 per day) and have everything else go to a "ham" folder when I delete them. A script runs and feeds those two sets to the filter automatically. No further tuning has been necessary.
I mention it all just in case there are people out there still suffering with spam: you can solve it, for all practical purposes.
Cheers.
Why do you think it "works" when your server has to scan and reject 700+ emails/day?
:) My server has no trouble doing the work, and it doesn't cost me anything.
Because I don't have to
But I guess you're saying that if spam is out there, then it's still a problem. To me, if everyone who wants to stop it from reaching them can do so with minimum effort... it's just not a big issue any more. Let them spam. Let the morons who buy the stuff buy it. As long as there's a decent means of avoiding it for anyone who wants to. And I believe there is.
Cheers.
Are any of you people still living with spam? Do we really need another solution? I've found that a personally managed baysean filter is plenty good enough. I'm down from 700+ per day to 2-3 per day. I still dislike the fact that spam is out there, but I haven't actually had to deal with it in years. Has this not worked for other people? I mean, I do have to continue to feed the filter, but it's very little work. Nothing wrong with new ideas in the battle, but I thought that for anyone who cared it was already won.
Cheers.
All the tests in the world can only do one thing with string theory: show that we haven't found a way to disprove it yet. All scientific theories are open to being disproven, that is the beauty of science, that is why it is not a religion, as much as religious types would like it to be, and despite the fact that many so-called scientists actually use it as a religion. The best one can hope for is that observation continues to bear out the predictive abilities of the theory. And you can consider a well tested theory as being good enough for general use... but you'll never really know for sure. So, I say get used to it. Revel in the fact that we don't know, but can still make amazingly useful predictions about our world.
Speaking of which: lets say that string theory survives this test. How far away are we from making useful predictions with string therory? That is, ones that are meaningfully more precise than quantum mechanics or general relativity? Last time I read about it they seemed nowhere near such a prediction because of the complexity of the mathematics. It seemed almost hopeless that they'd make predictions. Is that still the case? This test seems to say otherwise, but are the predictions notably different from what quantum mechanics predict?
Cheers.
A friend of mine does this.
Get a new friend.
the items being traded have value outside of the system itself
But do stocks have value "outside the system itself"? Where is "outside the system" of the stock market? I own stock, and I don't doubt that it is part of our functioning economy, but if I walk out of our economy, and try to pass off my paperwork to people who don't take part in our economy, they wouldn't be any more valuable than second-life items. And yes, regulation and legal recourse. Early on I think the idea of owning part of the company was important, but it is rarely the liquidity of a company that drives prices today. I think the stock market works simply because it is so well agreed upon and it spans such a large number of people that there practially is no "outside the system". It's not a real Ponzi scheme, per se, but somehing is weird because the price is dictated by the belief that someone in the future will pay more for the item than you will now, and if you follow that through it is circular, isn't it?
Cheers.
I hate to be a party pooper, but I doubt that something as revolutionary as the Chevy Volt will not be on the market in the next 5 years by any American automaker. I've been casually watching concept cars for a couple decades now, and very little of the dramatic tech makes it into production vehicles. It seems that they're just a way to keep people excited while they churn out the stuff that really sells: big cheap boxes.
On the other hand, I hope I'm wrong and if I am, I'd certainly be a potential customer for such a car, as I own a Prius and would love to see even more dramatic ways to reduce our oil dependency and polution.
Cheers.
Thanks for the well thought out response. I guess I have two points I'd like to respond to...
:)
First, I agree that only I should decide what I'm interested in. I don't want spam to be illegal in the sense that someone else decides what is okay to mail, but rather that it should be illegal to mail me stuff if I've requested that you don't. Because after I request that you don't, it starts becoming harassment. Of course with spam there are so many people doing it that asking each one individually wouldn't do a thing to reduce it, so a centralized list needs to be created. For telemarketing this is done with a do-not-call list, and for snailmail it is done with the DMA opt-out list, both of which I've signed up for, and both of which work reasonably well, though not completely. I also call and politely request removal from any company that sends me catalogs because I did business with them at some point. Overall I've reduced junk intrusion into my life to a reasonable level. I don't think making spam illegal is a complete solution, but I think it could help, and even if it didnt help, it's "right" in the sense that the spammer is abusing the system and should be punished in some way. I think that should be the case even if it wasn't an effective deterrent or if there are other solutions. You harass others, you get smacked.
Second, I don't think it's fair to push the burden of proving the worth of your email should fall on my customers/fans. If they have a question, they should be able to email me with a subject of "question" and not be immediately demoted. Besides, with 700+ spam messages a day, even with perfect subjects I wouldn't be able to sort them in a timely manner without error. Thank the lord for custom trained baysean filtering
Cheers.
I guess I shouldn't have made that part of the comment. I am a fan of any viable green energy source, and solar PV and wind certainly qualify. I guess I was aiming that at the people who seem to think that green energy can only be supplemental or is expensive stuff forced into the market by the environmentalist agenda. Geothermal has the hardcore 24/7 power output of a nuclear plant without the radiation issues.
In any case, I'm glad to hear that solar is coming along nicely.
I believe in free speech when the technology used is "pull" instead of "push". Does that make sense? I believe you have the right to say whatever you want to whomever you want if they are willing to listen, but I also have the right to not have to listen. I should be able to say "stop bothering me" and have that respected.
And unfortunately spam is not just an issue of "don't open the message". Many messages don't have helpful subject lines, instead opting for "Re: your mail" or somesuch. And if you have any public interactions, you don't get all your messages from people you already know. I get over 700 spam messages per day, this gets in the way of me doing my responsibilities. I am lucky in that I am tech savvy enough to have set up sophisticated spam filtering (so only 5 or so a day get through), but the original poster is right: it is illegal to call someone's home after you've told them not to. Nearly all direct marketers will stop sending snail mail if you ask (if not, I'm sure there would be a law about that too). Spam is harassment, not free speech.
Cheers.
Well first of all, corporations are only accountable to their shareholders, not to anyone else.
You're right. And Eichmann was only accountable to his superiors but ended up hanged anyways. Sorry for triggering Godwin's Law, but it is worth noting that there is widespread consensus that one is accountable for their actions even if accountability is not explicitly apparent in the system. I think that most of the SEC regulatory hoops and penalties are a not-wholly-unsuccessful effort to include ethics in the economic picture, so it's not true that they can't be accounted for. But it is a bit of a tragedy of the commons though since, as usual, the shortsighted people who misinterpret "corporations are only accountable to their shareholders" to mean "there's no place for ethics in the market" make an expensive pain in the ass for all of us.
Cheers.
Are you referring to the meat stew of tourists in the lagoon? Or in the restaurant? ;)
I did eat there, and yes, it was excellent. Hope to visit again someday, and spend more time exploring the glaciers... I was only there for one night.
Yes, they have a huge advantage in their ready access to both water and volcanic activity. It sounds like the report from the article indicates we've been underusing our own access to these resources.
;)
If that's Cheney's plan, I may have to find a little respect for the man
Cheers.
I realize that they have some special circumstances (and I mention this at the end of my post), but the primary ingredients are volcanic activity and water, and they don't have the market cornered on those items :)
:)
In other words, I realize it will take work to more widely deploy it, but I think it's an underexplored idea. And this MIT study seems to be saying just that. So I got excited
Yeah, the last line in my post mentions that. But I question "like nobody else". They aren't the only place that has volcanos and water.
I'm all for wind and solar and tidal and whatever else works. My point is that these projects are supplemental at best because they don't have continuous guaranteed power output, and I want to cut through anyone who has the perception that green energy is only supplemental. Sorry if I offended you, but there are plenty of people who won't take green energy seriously. Geothermal has the power output profile of a primary energy source, and I want to promote that.
You are right: I am uneducated though, because I was wrong about geothermal usage in Iceland. It's actually only 54% of the total national consumption of primary energy, the rest being petroleum, hydro and coal, in that order. It actually only provides 17% of the electricity, the rest comes from hydro. Most of the geothermal is used for direct heating.
Do you realize that some sidewalks in Reykjavik are heated because they have so much cheap and clean geothermal energy? You have to admit that's pretty neat.
Cheers.
Yes, the hot water runoff is an equal problem in many energy systems, including geothermal and nuclear. The reason geothermal is better is that it doesn't also carry the radioactive disposal and meltdown risk issues. Both of which are still better than coal burning for environmental damage. I think nuclear is pretty good overall, but geothermal is a little better.
Cheers.
I visited Iceland a couple years ago, and I became sold on geothermal. I mean, Iceland is a small country, but they have fairly high power needs per capita because of the cold climate, and they run almost entirely off geothermal, as I understand it. This isn't some apologetic green technology that is decades or more from delivering affordable massive power, like solar, wind, etc. No, this is the real thing: a geothermal plant puts out power at nuclear reactor levels. And these things are clean.
My favorite part of the visit was swimming in the Blue Lagoon... a spa built alongside the runoff from a geothermal power plant. Seriously: you're in the middle of a lava rock field, and boiling hot waste water pours from the power plant into a huge outdoor pool. In the cold air you can nearly cook yourself as you swim closer to the power plant. But it's clean enough to swim in.
There are many criteria that need to be met to build a geothermal power station at a given location, but I think the research and development needed must be far less than for some other technologies, and the end result is completely proven, so the risks are minimal.
My ideal-yet-realistic world features geothermal and nuclear supplementing each other, with the preference towards geothermal.
Cheers.
Simply not true. A creative person may not be completely satisfied with every aspect of what they've created (which is often the impetus for the next creative project), but despite this stuff gets completed all the time without being cut short. As a creative person (developing, music, and film) I've completed many things without someone else telling me when it had to be done. There are a percentage of people like you describe, and yes, they do need someone to just give them a hard deadline. But there are many creative people who know when something has got as good as it's going to get and to mark it "complete" and move on. Don't judge everyone because of a few experiences you've had.
What empowers him? I think it has something to do with the long term effects being fatal to our way of life? If that's not reason enough to say "we will have to make changes" then I don't know what is. Getting angry at the messenger isn't going to help.