Isn't this the same as having economists doing the work, just faster? You are still using past data to predict the future
Yes and no. In a sense, letting AI learn the salient traits of the available data just saves a human from needing to do it; but, you can do something with an algorithm that you can't reliably do with a human - You can model the existing system, then test billions of hypothetical situations to see how they respond.
Humans work amazingly well at pattern matching, even in the absence of understanding of "why". We can even get good at predicting what will likely happen if we change a few inputs to a system. But we don't do so well at figuring out what will happen if we tweak a large number of inputs simultaneously.
Think of this as nothing more than finally making batch hypothesis testing possible in an objective way, in a field where a persuasive argument matters more than facts and where a real experiment can take a few generations to fully show its outcome.
A year from now, I look forward to hearing Comcast whine about how "No legitimate user could seriously expect to pay $30 for 1.5 petabytes per month. Obviously, unlimited didn't mean unlimited - We intended it to give only another 300GB. We need to limit these greedy users out of fairness to our other customers."
Fuck 'em. I don't know who to consider dumber - Comcast, or any of their customers who fall for this again.
You need to quantify what you consider "good enough" in order to answer that.
First, in strict terms of bandwidth, no, today's best wireless just can't compete with today's best fiber. But how about tomorrow? No, tomorrow's best wireless still won't beat tomorrow's best fiber; but, with wireless, when 7G hits the scene everyone goes out and buys a new $50 modem and trucks don't need to physically roll to every end point on the network to upgrade their tubes.
Second, in more relaxed terms of bandwidth, when do we reach "enough" so that even revolutionary improvements don't really matter any more? Do I really need the ability to download a full 4k movie in under six seconds? I don't mean that as a "640k should be enough for anyone" argument, but at a point in time, yes, 640k did count as "enough" for most purposes, even though at that same point in time we had supercomputers with a whopping 16MB of main RAM.
Finally, and most importantly (I touched this in my first point), you asked specifically about "to the home". The biggest challenge in getting bits to the vast majority of homes has nothing to do with the throughput of the medium, but whether we can get it to the home in the first place. In the nearest city to me, I could get 1GB connections for a few hundred a month; living half an hour away, I don't even have the slowest of DSL available at any price. Whether or not fiber counts as "better" in that context doesn't mean a damned thing to me, because I won't ever see it.
When you ask about "good enough", keep in mind that the connection that meets all you needs, the connection that you can get, beats the much, much better one that you can't get.
Well, I have mine configured to give out leases from the range 192.168.100.100-199, which more than adequately serves my LAN for now. What sort of piece of crap modem do you have, that can only deal with a single client connection?
What fucking IP will things on the switch get?
I just answered that, but I'll repeat myself - Since switches work by transparently passing L2 traffic, they will get an address issued by the DHCP server on the modem, just like something directly plugged into the modem would. So something between 100 and 199.
What fucking IP will things on the router connected to the switch get?
They would get whatever range I configure the OnHub's own DHCP server to give out, exactly the same way it would work with any crappy $50 DLink/Linksys. Most likely I would pick 192.168.101.100-199, if you want an exact number.
How will traffic to and from these IPs be routed outside of your network?
DHCP leases include a gateway address. These can nest (almost) arbitrarily deep. How do you think your phone, connected to your WAP, in turn connected to your modem, manages to route traffic? Nothing magic here, dude.
This problem has a pretty straightforward answer - Just move, and take on a new identity as someone moderately wealthy (ie, no need to work, but not "solid-gold Veyron" level crazy-rich).
Incorrect. The Picostation has an omni antenna, but otherwise behaves just like all the rest of Ubiquiti's AirOS devices - It will act as any combination of {bridge / router / SOHO router} x {AP / Station / Client / Repeater}... And yes, a few of those combinations don't even make sense, but it will let you do it (never, ever disable the hard reset button on a Ubiquiti unless you know exactly what you want to do).
I absolutely love my Nanostations - Put one at one corner of an area you want covered, and bam, you will have five bars a quarter mile away in any direction (technically only a 60 degree beam, but it takes quite a distance from the antenna before that starts to matter).
I hate getting involved in this rapidly degenerating conversation, but...
You plug the switch into the LAN port and you plug your WAN connection into the WAN port. Hell the pictures from the article show two ports and two ethernet cables.
Assuming your modem has only one LAN port, you would do better to plug the modem into your switch, then you plug the OnHub's WAN side into the switch. This gives you full speed through the switch to your ISP for any wired devices you have, while not bogging down the processor on the OnHub dealing with non-wireless traffic. If your modem has two or more LAN ports, just connect one to your switch and one to your OnHub.
You would only want to use the LAN side of the OnHub in two, maybe three situations - You have only a single wired device in the whole house you need connected; you have a wired device that uses this hypothetical new Google spyOnYou protocol; and maybe you might put a second switch off the OnHub's LAN port if you had a strong need for an additional layer of segment isolation (and that assumes it truly isolates the LAN side, rather than merely acting as a two-port switch).
And before anyone points out that what I just proposed amounts to making your wired network a DMZ - Your modem already acts as a fully-functional SOHO router.
the point is not about detecting, the point is about being affected.
Even the worst "sufferers" of it can't successfully detect the presence of the very thing that supposedly leaves them in agony.
That is complete nonsense. With a bit of training, every one can do that.
the point is not about detecting, the point is about being affected.
"Well controlled and conducted double-blind studies have shown that symptoms were not correlated with EMF exposure."
Sorry, you are not listening, are you?
Not any more, nope. You've gone full circle with that shotgun o' logic; feel free to just keep recursively quoting yourself from the above two choices until you get bored.
Asking Facebook to follow German law while operating in Germany is somehow forcing "billions of Facebook users" to his ideology?
Yes, because Facebook doesn't exist only in Germany or only in the US.
If I, as a US citizen, want to deny the holocaust on Facebook, FB then has two choices - Remove the offending comment entirely, or at least block it from viewers in Germany. Either of those infringe on my right to express whatever brand of bigotry I may subscribe to despite living in an entirely different country that doesn't feel the need to outlaw critical thinking. I might not get arrested for it, but I would have had my voice silenced as a result of Germany's stupidity.
FWIW, I don't count as a holocaust denier. I arrived at that conclusion through rational consideration of the evidence, however, not because my government told me what to think - And in fact, the latter would make me less likely to believe it; any time the government really wants you to believe something, that raises the bar for the actual evidence a hell of a lot higher.
Here in the U.S., you cannot just say anything that you want without consequences. Hate speech, threats, and bullying are illegal here.
I agree with the rest of what you wrote, but one correction - Of those three, only credible threats actually break the law (with a few temporary state-by-state exceptions for cyberbullying).
Hate speech absolutely does not violate US law. Inciting to violence against them, sometimes (again, if credible); Ranting until you go horse about the evils of Muslims or gays or Canadians, no. You have every right to hate whatever groups you want and talk about it every chance you get - Hell, you can even do it while running for president!
Several states have passed anti-bullying laws, but not federally, and individual state supreme courts (e.g., New York) have already started overturning them as unconstitutional, and only a matter of time until the USSC does the same.
other than typical reactionary hate I don't see what the problem is.
You now have your init daemon providing an alternate attack pathway for gaining privileged access to the system, in a way that completely circumvents the well-established (and monitored by most IDSs) auditing capabilities of the platform.
Double blind. Unable to detect. What part of that don't you people get?
But hey, prove all the haters wrong! If you can do it so much better than everyone else, set up your own study and vindicate all these poor suffering folks condemned to a permanent vacation in a beautiful rural mountain village.
Interpolation is WORSE than nothing. you're discarding signal then adding noise in the hopes that it matches up with what should've been there kinda okay.
1, 2, 3, X, 5, 6. Guess the value of X... Congratulations, you just interpolated the right answer.
In the case of what the GP described, though, it works out even better than that, because the panel actually "knows" the right answer, so it hasn't "thrown away" information; it just lacks the luminance resolution to display it. It can, however, interpolate in the temporal domain way, way faster than the human eye can tell, to create a color we perceive as the correct value.
/ Go ahead, twitch gamers, tell us all about your ability to resolve sub-millisecond 1.5% color changes. XD
Could it be perhaps because nobody is subjected to double blind testing in order to determine whether or not they are disabled?
With most legitimate disabilities, a state licensed doctor can typically evaluate whether or not someone meets the criteria for a particular disability. How many legs does the patient have? Less than two? Okay, disabled.
And for the somewhat harder to prove disabilities like chronic pain, at least in the US the burden of proof rests on the individual to make their case, not the government to disprove it - Real sufferers wish they had a way to objectively prove their pain by something like a double-blind test.
RF sensitivity, amusingly enough, falls into a nice neat bin halfway between those two extremes. It has no externally measurable pathology, like chronic pain; but we do have a nice straightforward test to objectively disprove it as a legit disability - Even the worst "sufferers" of it can't successfully detect the presence of the very thing that supposedly leaves them in agony.
"No really, I swear, a shark bit my leg off! You just can't see it because [insert technobabble here]."
But given that where I feel this "whine" is my ear I don't think it is a stretch that it could be causing dizziness and nausea in others in fact is seems likely.
Do you feel confident that you could detect this whine under controlled experimental conditions, without any external information about when they turned the power on or off? And if not, what would that say about your actual ability to perceive that whine vs your beliefs about that whine?
That said, I don't disbelieve you about the whine. We can all hear it, because AC transformers and high voltage lines actually do make noise at the frequency of the AC - In the US, typically a 60Hz hum, but your choice of the word "whine" makes me think you most likely mean the 15kHz used in a cheap flyback transformer like you would find in an old TV.
it most definitely is not conclusive or concrete data.
If you claim $CAUSE gives you crippling pain, but can't tell whether or not $CAUSE exists without external confirmation, yes, that counts as both conclusive and concrete.
Try replacing $CAUSE with "a shark chewing on your leg". That "conclusive" enough for ya?:)
But spending any of his time or yours solving proprietary software licensing issues instead of making your own products work is a gigantic waste.
Great advice if you work in a pure-dev shop and the entire corporate food chain knows and likes Linux.
Career-ending advice, however, if you work in the other 99% of the IT industry and the CIO just wants the COTS ERP system to do its damned job.
I myself like and use Linux (at home), make no mistake. But suggesting someone "rewrite" the next version of their most-likely-3rd-party software to run on an open source platform just doesn't count as a practical, or often even possible, suggestion.
Is that actually the case? I thought a big purpose was to avoid voter intimidation by non-governmental vigilantes who oppose a particular candidate.
Absolutely! Your reason also holds true, but it comes in a distant second.
We tend to minimize the "Uncle Sam knows who you voted for" angle precisely because we don't live in a country where we routinely round up people who voted for the "wrong" candidate to torture or execute or "reeducate" them.
By contrast, consider (whatever your stance on the post-9/11 Iraq war) that Saddam Hussein routinely won reelection by an almost unanimous vote for precisely that reason.
None of the above really matter as long as any of them include the idea of "learning from your peers". If I pay a university to teach me something, they'd damned well better stick a relative expert on the subject matter in front of me for 40 hours over the next three months, whether in person, in realtime, or just "on demand".
Far, far too many online courses have roughly the same format as a Slashdot FP - Post the day's reading material, then require students to "discuss" it. Except, just like with Slashdot (browsing at 2+), the first few comments (almost always by the same few people) pretty much say it all, and everyone else tags along with "me too" - Albeit phrased much more verbosely to get credit for "participating".
Sorry, but I didn't pay to chat with people who know as little, or less, about the subject than I do. I don't have any interest in "learning" by helping my classmates catch up. I honestly do not give the least fuck about my "peers", and if I could afford to, I would have much preferred to only take classes with one-on-one instruction from a subject matter expert.
Bet it is cheaper, easier to accomplish, and better for everyone. Sure as hell will be cheaper then trying to change the climate on the whole planet.
Actually, it wouldn't.
We have the technology today to launch a massive fresnel lens to L1, at an estimated cost of only USD$20B over its lifetime.
For a manned mission to Mars - Not even talking about colonization here - NASA estimates it will cost over USD$100B and we won't have suitable technology available for a good 30 more years (though they could likely could speed that up by throwing more money at the problem).
No such requirement exists, however, to simply visit someone's Twitter page. I see this (extremely valuable) tool as likely rewritten into a straightforward page-scraper by the end of the day. Block that, Twitter!
Hell, I might rewrite it as such if I have a slow afternoon.
We don't need no stinkin' TOS to load a public website!
Slashdot Mirror
Isn't this the same as having economists doing the work, just faster? You are still using past data to predict the future
Yes and no. In a sense, letting AI learn the salient traits of the available data just saves a human from needing to do it; but, you can do something with an algorithm that you can't reliably do with a human - You can model the existing system, then test billions of hypothetical situations to see how they respond.
Humans work amazingly well at pattern matching, even in the absence of understanding of "why". We can even get good at predicting what will likely happen if we change a few inputs to a system. But we don't do so well at figuring out what will happen if we tweak a large number of inputs simultaneously.
Think of this as nothing more than finally making batch hypothesis testing possible in an objective way, in a field where a persuasive argument matters more than facts and where a real experiment can take a few generations to fully show its outcome.
A year from now, I look forward to hearing Comcast whine about how "No legitimate user could seriously expect to pay $30 for 1.5 petabytes per month. Obviously, unlimited didn't mean unlimited - We intended it to give only another 300GB. We need to limit these greedy users out of fairness to our other customers."
Fuck 'em. I don't know who to consider dumber - Comcast, or any of their customers who fall for this again.
You need to quantify what you consider "good enough" in order to answer that.
First, in strict terms of bandwidth, no, today's best wireless just can't compete with today's best fiber. But how about tomorrow? No, tomorrow's best wireless still won't beat tomorrow's best fiber; but, with wireless, when 7G hits the scene everyone goes out and buys a new $50 modem and trucks don't need to physically roll to every end point on the network to upgrade their tubes.
Second, in more relaxed terms of bandwidth, when do we reach "enough" so that even revolutionary improvements don't really matter any more? Do I really need the ability to download a full 4k movie in under six seconds? I don't mean that as a "640k should be enough for anyone" argument, but at a point in time, yes, 640k did count as "enough" for most purposes, even though at that same point in time we had supercomputers with a whopping 16MB of main RAM.
Finally, and most importantly (I touched this in my first point), you asked specifically about "to the home". The biggest challenge in getting bits to the vast majority of homes has nothing to do with the throughput of the medium, but whether we can get it to the home in the first place. In the nearest city to me, I could get 1GB connections for a few hundred a month; living half an hour away, I don't even have the slowest of DSL available at any price. Whether or not fiber counts as "better" in that context doesn't mean a damned thing to me, because I won't ever see it.
When you ask about "good enough", keep in mind that the connection that meets all you needs, the connection that you can get, beats the much, much better one that you can't get.
Uh-huh. Post your topology (including model numbers and salient points of your configs) if you have anything even remotely different.
No rush, I won't hold my breath.
What fucking IP will your modem give you?
Well, I have mine configured to give out leases from the range 192.168.100.100-199, which more than adequately serves my LAN for now. What sort of piece of crap modem do you have, that can only deal with a single client connection?
What fucking IP will things on the switch get?
I just answered that, but I'll repeat myself - Since switches work by transparently passing L2 traffic, they will get an address issued by the DHCP server on the modem, just like something directly plugged into the modem would. So something between 100 and 199.
What fucking IP will things on the router connected to the switch get?
They would get whatever range I configure the OnHub's own DHCP server to give out, exactly the same way it would work with any crappy $50 DLink/Linksys. Most likely I would pick 192.168.101.100-199, if you want an exact number.
How will traffic to and from these IPs be routed outside of your network?
DHCP leases include a gateway address. These can nest (almost) arbitrarily deep. How do you think your phone, connected to your WAP, in turn connected to your modem, manages to route traffic? Nothing magic here, dude.
This problem has a pretty straightforward answer - Just move, and take on a new identity as someone moderately wealthy (ie, no need to work, but not "solid-gold Veyron" level crazy-rich).
Incorrect. The Picostation has an omni antenna, but otherwise behaves just like all the rest of Ubiquiti's AirOS devices - It will act as any combination of {bridge / router / SOHO router} x {AP / Station / Client / Repeater}... And yes, a few of those combinations don't even make sense, but it will let you do it (never, ever disable the hard reset button on a Ubiquiti unless you know exactly what you want to do).
I absolutely love my Nanostations - Put one at one corner of an area you want covered, and bam, you will have five bars a quarter mile away in any direction (technically only a 60 degree beam, but it takes quite a distance from the antenna before that starts to matter).
I hate getting involved in this rapidly degenerating conversation, but...
You plug the switch into the LAN port and you plug your WAN connection into the WAN port. Hell the pictures from the article show two ports and two ethernet cables.
Assuming your modem has only one LAN port, you would do better to plug the modem into your switch, then you plug the OnHub's WAN side into the switch. This gives you full speed through the switch to your ISP for any wired devices you have, while not bogging down the processor on the OnHub dealing with non-wireless traffic. If your modem has two or more LAN ports, just connect one to your switch and one to your OnHub.
You would only want to use the LAN side of the OnHub in two, maybe three situations - You have only a single wired device in the whole house you need connected; you have a wired device that uses this hypothetical new Google spyOnYou protocol; and maybe you might put a second switch off the OnHub's LAN port if you had a strong need for an additional layer of segment isolation (and that assumes it truly isolates the LAN side, rather than merely acting as a two-port switch).
And before anyone points out that what I just proposed amounts to making your wired network a DMZ - Your modem already acts as a fully-functional SOHO router.
the point is not about detecting, the point is about being affected.
"Well controlled and conducted double-blind studies have shown that symptoms were not correlated with EMF exposure."
Sorry, you are not listening, are you?
Not any more, nope. You've gone full circle with that shotgun o' logic; feel free to just keep recursively quoting yourself from the above two choices until you get bored.
Let me know when you successfully detect the WiFi in a double-blind study.
Asking Facebook to follow German law while operating in Germany is somehow forcing "billions of Facebook users" to his ideology?
Yes, because Facebook doesn't exist only in Germany or only in the US.
If I, as a US citizen, want to deny the holocaust on Facebook, FB then has two choices - Remove the offending comment entirely, or at least block it from viewers in Germany. Either of those infringe on my right to express whatever brand of bigotry I may subscribe to despite living in an entirely different country that doesn't feel the need to outlaw critical thinking. I might not get arrested for it, but I would have had my voice silenced as a result of Germany's stupidity.
FWIW, I don't count as a holocaust denier. I arrived at that conclusion through rational consideration of the evidence, however, not because my government told me what to think - And in fact, the latter would make me less likely to believe it; any time the government really wants you to believe something, that raises the bar for the actual evidence a hell of a lot higher.
Here in the U.S., you cannot just say anything that you want without consequences. Hate speech, threats, and bullying are illegal here.
I agree with the rest of what you wrote, but one correction - Of those three, only credible threats actually break the law (with a few temporary state-by-state exceptions for cyberbullying).
Hate speech absolutely does not violate US law. Inciting to violence against them, sometimes (again, if credible); Ranting until you go horse about the evils of Muslims or gays or Canadians, no. You have every right to hate whatever groups you want and talk about it every chance you get - Hell, you can even do it while running for president!
Several states have passed anti-bullying laws, but not federally, and individual state supreme courts (e.g., New York) have already started overturning them as unconstitutional, and only a matter of time until the USSC does the same.
So a country that decided to throw its weight around to force its ideologies on a few million Jews (by killing them)...
...has decided to throw its weight around to force its ideologies on a few billion Facebook users (albeit without death resulting, for now)?
Nice!
other than typical reactionary hate I don't see what the problem is.
You now have your init daemon providing an alternate attack pathway for gaining privileged access to the system, in a way that completely circumvents the well-established (and monitored by most IDSs) auditing capabilities of the platform.
I'd call that a problem, but YMMV.
Never mind following my link, did you even read the one sentence summary I quoted in my original post???
Here, if the World Health Organization doesn't count as a good enough source for you, how about a nice high quality Wikipedia link:
several double-blind experiments have been published, each of which has suggested that people who report electromagnetic hypersensitivity are unable to detect the presence of electromagnetic fields and are as likely to report ill health following a sham exposure as they are following exposure to genuine electromagnetic fields
Double blind. Unable to detect. What part of that don't you people get?
But hey, prove all the haters wrong! If you can do it so much better than everyone else, set up your own study and vindicate all these poor suffering folks condemned to a permanent vacation in a beautiful rural mountain village.
Interpolation is WORSE than nothing. you're discarding signal then adding noise in the hopes that it matches up with what should've been there kinda okay.
1, 2, 3, X, 5, 6. Guess the value of X... Congratulations, you just interpolated the right answer.
In the case of what the GP described, though, it works out even better than that, because the panel actually "knows" the right answer, so it hasn't "thrown away" information; it just lacks the luminance resolution to display it. It can, however, interpolate in the temporal domain way, way faster than the human eye can tell, to create a color we perceive as the correct value.
/ Go ahead, twitch gamers, tell us all about your ability to resolve sub-millisecond 1.5% color changes. XD
Could it be perhaps because nobody is subjected to double blind testing in order to determine whether or not they are disabled?
With most legitimate disabilities, a state licensed doctor can typically evaluate whether or not someone meets the criteria for a particular disability. How many legs does the patient have? Less than two? Okay, disabled.
And for the somewhat harder to prove disabilities like chronic pain, at least in the US the burden of proof rests on the individual to make their case, not the government to disprove it - Real sufferers wish they had a way to objectively prove their pain by something like a double-blind test.
RF sensitivity, amusingly enough, falls into a nice neat bin halfway between those two extremes. It has no externally measurable pathology, like chronic pain; but we do have a nice straightforward test to objectively disprove it as a legit disability - Even the worst "sufferers" of it can't successfully detect the presence of the very thing that supposedly leaves them in agony.
"No really, I swear, a shark bit my leg off! You just can't see it because [insert technobabble here]."
But given that where I feel this "whine" is my ear I don't think it is a stretch that it could be causing dizziness and nausea in others in fact is seems likely.
:)
Do you feel confident that you could detect this whine under controlled experimental conditions, without any external information about when they turned the power on or off? And if not, what would that say about your actual ability to perceive that whine vs your beliefs about that whine?
That said, I don't disbelieve you about the whine. We can all hear it, because AC transformers and high voltage lines actually do make noise at the frequency of the AC - In the US, typically a 60Hz hum, but your choice of the word "whine" makes me think you most likely mean the 15kHz used in a cheap flyback transformer like you would find in an old TV.
it most definitely is not conclusive or concrete data.
If you claim $CAUSE gives you crippling pain, but can't tell whether or not $CAUSE exists without external confirmation, yes, that counts as both conclusive and concrete.
Try replacing $CAUSE with "a shark chewing on your leg". That "conclusive" enough for ya?
"Well controlled and conducted double-blind studies have shown that symptoms were not correlated with EMF exposure."
Done in one.
But spending any of his time or yours solving proprietary software licensing issues instead of making your own products work is a gigantic waste.
Great advice if you work in a pure-dev shop and the entire corporate food chain knows and likes Linux.
Career-ending advice, however, if you work in the other 99% of the IT industry and the CIO just wants the COTS ERP system to do its damned job.
I myself like and use Linux (at home), make no mistake. But suggesting someone "rewrite" the next version of their most-likely-3rd-party software to run on an open source platform just doesn't count as a practical, or often even possible, suggestion.
Is that actually the case? I thought a big purpose was to avoid voter intimidation by non-governmental vigilantes who oppose a particular candidate.
Absolutely! Your reason also holds true, but it comes in a distant second.
We tend to minimize the "Uncle Sam knows who you voted for" angle precisely because we don't live in a country where we routinely round up people who voted for the "wrong" candidate to torture or execute or "reeducate" them.
By contrast, consider (whatever your stance on the post-9/11 Iraq war) that Saddam Hussein routinely won reelection by an almost unanimous vote for precisely that reason.
None of the above really matter as long as any of them include the idea of "learning from your peers". If I pay a university to teach me something, they'd damned well better stick a relative expert on the subject matter in front of me for 40 hours over the next three months, whether in person, in realtime, or just "on demand".
Far, far too many online courses have roughly the same format as a Slashdot FP - Post the day's reading material, then require students to "discuss" it. Except, just like with Slashdot (browsing at 2+), the first few comments (almost always by the same few people) pretty much say it all, and everyone else tags along with "me too" - Albeit phrased much more verbosely to get credit for "participating".
Sorry, but I didn't pay to chat with people who know as little, or less, about the subject than I do. I don't have any interest in "learning" by helping my classmates catch up. I honestly do not give the least fuck about my "peers", and if I could afford to, I would have much preferred to only take classes with one-on-one instruction from a subject matter expert.
Enabling this option however will, quite rightly, prohibit the subscriber from accessing other free WiFi spots
I guess I don't quite get the whole concept of "free" as used, then.
So the general public can use it, but a paying customer who doesn't want to subsidize Virgin's electric bill can't?
Virgin has an interesting concept of "fair".
Bet it is cheaper, easier to accomplish, and better for everyone. Sure as hell will be cheaper then trying to change the climate on the whole planet.
Actually, it wouldn't.
We have the technology today to launch a massive fresnel lens to L1, at an estimated cost of only USD$20B over its lifetime.
For a manned mission to Mars - Not even talking about colonization here - NASA estimates it will cost over USD$100B and we won't have suitable technology available for a good 30 more years (though they could likely could speed that up by throwing more money at the problem).
No such requirement exists, however, to simply visit someone's Twitter page. I see this (extremely valuable) tool as likely rewritten into a straightforward page-scraper by the end of the day. Block that, Twitter!
Hell, I might rewrite it as such if I have a slow afternoon.
We don't need no stinkin' TOS to load a public website!