Note: when the Federal Reserve loans money, money is typically created - AFAIK nobody's bank balances goes down - what happens is devaluation of the currency.
Yes I know there's a legal difference between what the FR and NK do, but still...
What do you propose that solves the problem better than that algorithm? And what is "right" by your definition?
As for my proposal, it's more of an approach - since it seemed to me that a lot of people were not using the "time/age in router"[1] of a packet to help determine whether to discard it or not. To back that up, just look at all the RED papers, and all the talk about "bufferbloat" ( where bloat=size of buffers). Not using "time/age in router" to help solve the latency problem seems silly.
In my original post I did say:
If you really want to address latency what routers should do is keep track of how long packets have been in the router (in clock ticks, milliseconds or even microseconds) and use that with QoS stuff (and maybe some heuristics) to figure out which packets to send first, or to drop.
The simple/naive example was just for example... I'm well aware that you can't keep all packets. Some approaches are worse than others. If latency is a big issue then an approach that takes into account packet latency in the router is more likely to give better results than approaches that don't.
Uh. The real problem is latency, not buffer size. Right? That's what people are complaining about - packets are taking longer than they should.
You propose solving the latency problem by reducing buffer size.
I propose solving the latency problem by reducing the maximum time packets are allowed to spend in a router (AKA latency).
Which do you think actually deals with the real issue? If the problem is latency, why not deal with it rather than go off on loosely related tangents?
BTW I finally read the article and seems what they propose is actually related to my suggestion - use time in queue. I think their approach might work - it definitely should work better than stuff like RED (which is stupid IMO).
FWIW I came up with my suggestion more than a year ago ( http://tech.slashdot.org/comments.pl?sid=1939940&cid=34793154 ) it was rather obvious to me - fix latency by dealing with latency. Well duh! I later commented on a related article that was linked to by Slashdot e.g. http://queue.acm.org/detail.cfm?id=2071893 Note: the article author(s) were still on the wrong track back then (still focusing on size of buffers). That's why I posted my post. Maybe they took the idea and ran with it. Or they came up with it by themselves.
That's why I think most patents nowadays are BS. It's easy for anyone reasonably bright to come up with very many different good[1] ideas. The trouble is implementing them and then pushing them to the market takes way longer. I'm not currently in the business of writing router software.
Lastly if RED doesn't take into account packet size when it drops then it hurts lower bandwidth channels with small packets disproportionately more than the ones with big packets, and most latency sensitive applications use small packets. When communicating across say the pacific ocean, unnecessary packet loss can hurt a lot more (2 *50 milliseconds?).
I disagree. It's fine for buffers to be very big, practically "infinite" even. Buffer size does not have to be linked with latency at all. The reason it is currently is because most routers are doing it wrong;).
If you really want to address latency what routers should do is keep track of how long packets have been in the router (in clock ticks, milliseconds or even microseconds) and use that with QoS stuff (and maybe some heuristics) to figure out which packets to send first, or to drop.
For example, "bulk/throughput" packets (think email) might be kept around for 100+ milliseconds. In contrast, while latency sensitive packets get priority, they might be dropped if they cannot be sent within tens of milliseconds ( so both sides can earlier detect the comms channel can't cope and possibly deal with it sooner rather than wait till the latency gets intolerable).
With my proposed approach a latency insensitive burst of big packets through a high bandwidth channel does not necessarily have to cause packet loss at all - whether from latency sensitive or latency insensitive channels. There would be enough buffer space to hold all the big packets while letting the low latency stuff through. Whereas with the smaller buffer size approach you are more likely to have unnecessary packet loss - which reduces throughput.
Even if the router doesn't do any QoS stuff it's way more useful to people who care about latency to be able to configure a router so that the max latency it will ever cause is 10 milliseconds. Especially if the router has multiple network connections with different bandwidths (e.g. 2Mbps, 100Mbps).
Air war? China only has one aircraft carrier, and that's a rather recent development.
More likely to have a war of words and political posturing, with some cosy backroom wheeling and dealing between the "battling" leaders ("you let me say bad things about you to the US sheeple to win elections and I'll let GE transfer more nuclear tech to you", "Only if you let me say this to the Chinese peasants AND stop that trade sanctions bullshit", "It's not bullshit"... etc etc).
And big fucking deal. 0) The loans are payable in US dollars, not gold, not copper, not Yuan. 1) Go look up how much the USA owes China. 2) Then go look up how much the Federal Reserve gave out in loans in the last financial crisis (google for Federal Reserve trillions if you have problems). 3) Calculate how many times bigger 2) is compared to 1).
China has a pile of US dollars it would prefer to spend on useful stuff before it gets worth even less. Doing silly stuff like you mentioned would cost them a fair bit.
I'm not an electrician but if you're moderately paranoid: 1) Press the breaker test switch to try to turn stuff off, and to test the breaker. 2) Test the wires with a multimeter to make sure 3) short the wires, so if some wise guy tries to turn stuff back on you don't get zapped so badly (the breaker should trip as tested in 1) ). 4) Work on stuff. 5) unshort the wires 6) reset the breaker. This only applies for lower voltages.
For the HV stuff you better be more careful - even if stuff is off the residual charge can be enough to kill you.
Yeah some time back I had a crazy bad idea for a law.
Basically a motorcyclist that was provably not wearing a helmet and ended up in a fatal accident is automatically regarded as having consented to donating his/her organs.
I suppose to make it less bad we could add the clause, "unless said person has already explicitly and officially specified nonconsent to organ donation". But who'd want to live in a society with such ruthless laws?
In particular, writing: It's designed that right-handers are dragging the writing implement behind their hand in a smooth gliding motion
Left handers may prefer Arabic or Hebrew (right to left)... Or old style Chinese...(top to down, right to left).
That's specifically the reason why my uncle (for example) was forced to switch by my grandparents tying his left hand behind his back.
Trivia: Rafael Nadal a top tennis player is right-handed but was encouraged by his uncle to play left-handed (an advantage in such sports)... Seems to be doing OK.
We may be better at learning to do stuff with our dominant hand, but it's definitely possible for most to learn how to do complex stuff with the "other hand". After all when you're doing one thing with your dominant hand, your other hand ends up having to do what's left... I learned to drive stick-shift in UK style cars - I doubt I'd be good at driving stick-shift with my right hand. And a car had the clutch pedals swapped with the accelerator, I'm sure my right foot would be bad at clutch-control till I got enough practice.
Conventional fighting game controllers have the joystick on the left and the buttons on the right. At the beginner/intermediate levels the complex and quick motions have to be done mainly with the joystick - so the left-handers may find it easier than the right handers. But for some games at the advanced levels of play it gets complex on the buttons too (e.g. streetfighter: plinking, double tapping, "pianoing"). That said there's a guy (Mike Begum) who plays Streetfighter with his _mouth_ at a rather high level...
A government subsiding on low income would be limited in the damage it could do, further helping its reputation.
The good it could do would also be limited. Even Hong Kong and Singapore tax rates aren't that low (I doubt those oil nations can sustain their low tax rates when they run out of oil). Is the USA really rich enough to do run an effective Government on 2% tax? What sort of Government or country do you expect to get? http://en.wikipedia.org/wiki/List_of_countries_by_tax_rates
Are you one of those libertarians who foolishly keep obsessing over quantity instead of quality?
As someone here already said:
Libertarians seem to believe that the US has become a corporate dictatorship so as a solution they want to remove the middle man and go straight to a corporate dictatorship.
Because a corporate dictatorship is what you'd get if your "2%" Government is too feeble to regulate the corporations. A weak Gov might say to the Corporations "Stop that or else", and the Corporations could ask "who is going to make us? You and WHAT army?".
Apple and gang won't hold elections every few years to let the pesky US voters decide who gets to be the bosses.
I believe right now you need to be rich enough so that it becomes worth the hassle.
But once you say corporations don't get taxed at all, nearly everyone that currently pays significant income tax will join/start a corporation (individually or with partners), and stop paying any tax.
In hindsight you can say it was nothing. But it's all based on the information they had at that time. When a potentially infected passenger is in flight, you'll just have to quarantine everyone on the flight till you know things are OK.
What do you propose they do instead? a) Let everyone including her go? b) Check her while letting the rest of the passengers go?
What if it turns out she is infected with something nasty and contagious? You now have even more people to quarantine - since the other passengers would have travelled "everywhere" and come into contact with even more people.
They violated their own protocols.
Can you provide actual evidence that they violated their own protocols?
No reason? Based on the information they had, they had good reason to suspect the passenger might have a serious contagious disease from Uganda.
And that passenger was on a plane that will land in the USA. They hadn't seen the patient to be sure that she didn't have monkeypox OR something worse yet.
So do you actually propose letting the passenger and all of them head off to their various destinations without examining any of them? Yeah maybe the fire trucks were unnecessary (not sure why they were there) but the rest is reasonable assuming shit actually happened- cops to handle any uncooperative passengers, ambulances to take the cooperative infected away.
You'll be glad to know that I do not work in the medical field. However I also hope you do not work in the medical field either. You seem even more incompetent than I am when it comes to medical matters.
Secondary transmission is human-to-human, resulting from close contact with infected respiratory tract excretions, with the skin lesions of an infected person or with recently contaminated objects. Transmission via droplet respiratory particles has also been documented. Transmission can also occur by inoculation or via the placenta (congenital monkeypox). There is no evidence to date that person-to-person transmission alone can sustain monkeypox in the human population.
Despite the last sentence, based on the rest of the paragraph I doubt you'd still want anyone who might have monkeypox to roam freely. 1) It's still a nasty disease to get 2) Some other animal might get it from the human, and monkeypox might be sustainable in that species.
Huh, overreaction? Do you really want monkey pox and other nasty diseases endemic in the Africa to not be rare in your country?
It's not overreaction at all. It's appropriate reaction. When the next super pandemic breaks out, the main solution is likely to be quarantine. The hospitals won't cope, and when the doctors and nurses start getting sick themselves it all falls apart and you'd be better off not going to the hospital for anything. Stay at home, and wait for the disease to kill all it can kill and/or evolve to be less lethal (that can happen in a short time - it can't spread if it kills too fast[1]).
[1] Quarantine in theory can actually work very well in breeding diseases to be less nasty and lethal.
Most of that "elections" game theory reasoning is only true if you are playing the game by assume there is only one election, and that's it. Then yes you vote for the 2nd place finisher. But if there's going to be more than one election it's a stupid idea, unless you genuinely would be more happy with the 2nd place finisher than the 3r place or other.
Because if people voted for the candidate they actually wanted rather than their second least hated candidate, then even if the 1st candidate still wins, the results could show everyone how many voters really wanted the other candidates. That sends a signal.
Then the voters in the next election might figure that one of the alternative candidates actually might have a chance, isn't so bad, and so vote for that candidate.
Alternatively the 2 Parties might realize that more and mroe people really want something different and change accordingly to try to ensure the other candidates don't win. While this does mean the 2 Parties might still win, the voters could still get the changes they wanted.
If you bunch are trying all that "game theory" crap when voting and are NOT succeeding, perhaps it's time to realize that the Two Parties are better at playing the game than you are. So stop doing that stupid shit which is not working.
Lastly: the first past the post system also works the same way for the "3rd candidate" if that candidate happens to get enough votes. Then suddenly the incumbent gets thrown out. That doesn't happen because either a) the voters stupidly think they are so smart at gaming the system or b) the voters actually prefer the Two Parties. Which may be true despite what people here might prefer to believe - there are certainly very many Obama fans (same goes for the R fans). Often it's as if it's their religion or similar. So it'll take a lot to convince them to vote differently.
Slashdot Mirror
If counterfeiting tens of millions a year = war then I think the US Federal Reserve has declared war on the USA. ;). http://www.google.com/search?q=federal+reserve+trillions
Note: when the Federal Reserve loans money, money is typically created - AFAIK nobody's bank balances goes down - what happens is devaluation of the currency.
Yes I know there's a legal difference between what the FR and NK do, but still...
Is this not a fancy latency-based algorithm? http://queue.acm.org/detail.cfm?id=2209336
What do you propose that solves the problem better than that algorithm? And what is "right" by your definition?
As for my proposal, it's more of an approach - since it seemed to me that a lot of people were not using the "time/age in router"[1] of a packet to help determine whether to discard it or not. To back that up, just look at all the RED papers, and all the talk about "bufferbloat" ( where bloat=size of buffers). Not using "time/age in router" to help solve the latency problem seems silly.
In my original post I did say:
If you really want to address latency what routers should do is keep track of how long packets have been in the router (in clock ticks, milliseconds or even microseconds) and use that with QoS stuff (and maybe some heuristics) to figure out which packets to send first, or to drop.
The simple/naive example was just for example... I'm well aware that you can't keep all packets. Some approaches are worse than others. If latency is a big issue then an approach that takes into account packet latency in the router is more likely to give better results than approaches that don't.
[1] There's TTL but that's total age of packet.
Uh. The real problem is latency, not buffer size. Right? That's what people are complaining about - packets are taking longer than they should.
You propose solving the latency problem by reducing buffer size.
I propose solving the latency problem by reducing the maximum time packets are allowed to spend in a router (AKA latency).
Which do you think actually deals with the real issue? If the problem is latency, why not deal with it rather than go off on loosely related tangents?
BTW I finally read the article and seems what they propose is actually related to my suggestion - use time in queue. I think their approach might work - it definitely should work better than stuff like RED (which is stupid IMO).
FWIW I came up with my suggestion more than a year ago ( http://tech.slashdot.org/comments.pl?sid=1939940&cid=34793154 ) it was rather obvious to me - fix latency by dealing with latency. Well duh! I later commented on a related article that was linked to by Slashdot e.g. http://queue.acm.org/detail.cfm?id=2071893
Note: the article author(s) were still on the wrong track back then (still focusing on size of buffers). That's why I posted my post. Maybe they took the idea and ran with it. Or they came up with it by themselves.
That's why I think most patents nowadays are BS. It's easy for anyone reasonably bright to come up with very many different good[1] ideas. The trouble is implementing them and then pushing them to the market takes way longer. I'm not currently in the business of writing router software.
[1] Insanely great is a different matter ;).
RED seems to be a primitive hack job to me.
My proposal is this: http://tech.slashdot.org/comments.pl?sid=2837433&cid=39941029
Lastly if RED doesn't take into account packet size when it drops then it hurts lower bandwidth channels with small packets disproportionately more than the ones with big packets, and most latency sensitive applications use small packets. When communicating across say the pacific ocean, unnecessary packet loss can hurt a lot more (2 *50 milliseconds?).
I disagree. It's fine for buffers to be very big, practically "infinite" even. Buffer size does not have to be linked with latency at all. The reason it is currently is because most routers are doing it wrong ;).
If you really want to address latency what routers should do is keep track of how long packets have been in the router (in clock ticks, milliseconds or even microseconds) and use that with QoS stuff (and maybe some heuristics) to figure out which packets to send first, or to drop.
For example, "bulk/throughput" packets (think email) might be kept around for 100+ milliseconds. In contrast, while latency sensitive packets get priority, they might be dropped if they cannot be sent within tens of milliseconds ( so both sides can earlier detect the comms channel can't cope and possibly deal with it sooner rather than wait till the latency gets intolerable).
With my proposed approach a latency insensitive burst of big packets through a high bandwidth channel does not necessarily have to cause packet loss at all - whether from latency sensitive or latency insensitive channels. There would be enough buffer space to hold all the big packets while letting the low latency stuff through. Whereas with the smaller buffer size approach you are more likely to have unnecessary packet loss - which reduces throughput.
Even if the router doesn't do any QoS stuff it's way more useful to people who care about latency to be able to configure a router so that the max latency it will ever cause is 10 milliseconds. Especially if the router has multiple network connections with different bandwidths (e.g. 2Mbps, 100Mbps).
Air war? China only has one aircraft carrier, and that's a rather recent development.
... etc etc).
More likely to have a war of words and political posturing, with some cosy backroom wheeling and dealing between the "battling" leaders ("you let me say bad things about you to the US sheeple to win elections and I'll let GE transfer more nuclear tech to you", "Only if you let me say this to the Chinese peasants AND stop that trade sanctions bullshit", "It's not bullshit"
And big fucking deal.
0) The loans are payable in US dollars, not gold, not copper, not Yuan.
1) Go look up how much the USA owes China.
2) Then go look up how much the Federal Reserve gave out in loans in the last financial crisis (google for Federal Reserve trillions if you have problems).
3) Calculate how many times bigger 2) is compared to 1).
China has a pile of US dollars it would prefer to spend on useful stuff before it gets worth even less. Doing silly stuff like you mentioned would cost them a fair bit.
it's certainly an overwhelming survival advantage.
That's not proven yet. We haven't been around for that long and there's a high chance that we could kill ourselves first.
If we can sustainably reproduce off this planet then we do actually have a better chance, otherwise I doubt we'd do better even than the dinos.
So far the odds are bacteria would out-survive us. And there aren't many scientists who would claim that bacteria are that intelligent.
It could also happen at the meme/belief system level.
The substrate or hosts do not even need to be limited to a single species or even conventional organisms.
Whatever happened to "sticks and stones will break my bones, but words will never hurt me."
What happened? That remark has never been true soon after words were invented.
And just wait till voice controlled stuff becomes even more popular and prevalent.
I'm not an electrician but if you're moderately paranoid:
1) Press the breaker test switch to try to turn stuff off, and to test the breaker.
2) Test the wires with a multimeter to make sure
3) short the wires, so if some wise guy tries to turn stuff back on you don't get zapped so badly (the breaker should trip as tested in 1) ).
4) Work on stuff.
5) unshort the wires
6) reset the breaker.
This only applies for lower voltages.
For the HV stuff you better be more careful - even if stuff is off the residual charge can be enough to kill you.
That's mostly due to people fleeing other currencies for the "safe" haven of US dollars during the financial crisis of 2008-2009.
Yeah I didn't really understand that bit.
1) A lot of the "exploding" stuff was in the USA.
2) The federal reserve was "printing" trillions of US dollars, and refusing to say where it was going. http://www.bloomberg.com/apps/news?pid=newsarchive&sid=aatlky_cH.tY ).
Maybe they also realized they wanted something that they'd be sure would work on Feb 29 too.
Perhaps next year Microsoft will launch Office 365.2425.
Yeah some time back I had a crazy bad idea for a law.
Basically a motorcyclist that was provably not wearing a helmet and ended up in a fatal accident is automatically regarded as having consented to donating his/her organs.
I suppose to make it less bad we could add the clause, "unless said person has already explicitly and officially specified nonconsent to organ donation". But who'd want to live in a society with such ruthless laws?
In particular, writing: It's designed that right-handers are dragging the writing implement behind their hand in a smooth gliding motion
Left handers may prefer Arabic or Hebrew (right to left)... Or old style Chinese...(top to down, right to left).
That's specifically the reason why my uncle (for example) was forced to switch by my grandparents tying his left hand behind his back.
Trivia: Rafael Nadal a top tennis player is right-handed but was encouraged by his uncle to play left-handed (an advantage in such sports)... Seems to be doing OK.
We may be better at learning to do stuff with our dominant hand, but it's definitely possible for most to learn how to do complex stuff with the "other hand". After all when you're doing one thing with your dominant hand, your other hand ends up having to do what's left... I learned to drive stick-shift in UK style cars - I doubt I'd be good at driving stick-shift with my right hand. And a car had the clutch pedals swapped with the accelerator, I'm sure my right foot would be bad at clutch-control till I got enough practice.
Conventional fighting game controllers have the joystick on the left and the buttons on the right. At the beginner/intermediate levels the complex and quick motions have to be done mainly with the joystick - so the left-handers may find it easier than the right handers. But for some games at the advanced levels of play it gets complex on the buttons too (e.g. streetfighter: plinking, double tapping, "pianoing"). That said there's a guy (Mike Begum) who plays Streetfighter with his _mouth_ at a rather high level...
Really? Then doesn't that make anyone wonder why the Libertarians are so enthusiastic about getting rid of Government in so many areas?
Seems like all those nice rules are not applicable once it's a corporation and not the government stomping on you.
I have a gun, and I am capable of acting unilaterally. Should I be restricted from speaking my mind as a result?
Of course not. Because if you go crazy with your guns till even the cops can't stop you, the Military does the job.
The Military are supposed to be the "Biggest Dog" in the Country. And that's why they should be on a "leash".
They are supposed to be the "Gun" that the elected leaders aim at targets. They are not supposed to do the "big picture" aiming by themselves.
A government subsiding on low income would be limited in the damage it could do, further helping its reputation.
The good it could do would also be limited. Even Hong Kong and Singapore tax rates aren't that low (I doubt those oil nations can sustain their low tax rates when they run out of oil). Is the USA really rich enough to do run an effective Government on 2% tax? What sort of Government or country do you expect to get? http://en.wikipedia.org/wiki/List_of_countries_by_tax_rates
Are you one of those libertarians who foolishly keep obsessing over quantity instead of quality?
As someone here already said:
Libertarians seem to believe that the US has become a corporate dictatorship so as a solution they want to remove the middle man and go straight to a corporate dictatorship.
Because a corporate dictatorship is what you'd get if your "2%" Government is too feeble to regulate the corporations. A weak Gov might say to the Corporations "Stop that or else", and the Corporations could ask "who is going to make us? You and WHAT army?".
Apple and gang won't hold elections every few years to let the pesky US voters decide who gets to be the bosses.
The difference is in the degree and the extent.
I believe right now you need to be rich enough so that it becomes worth the hassle.
But once you say corporations don't get taxed at all, nearly everyone that currently pays significant income tax will join/start a corporation (individually or with partners), and stop paying any tax.
I did not dismiss it. I addressed it see 1) and 2).
Read this too: http://articles.latimes.com/2010/sep/26/opinion/la-oe-orent-pox-20100926
Congo is not Uganda, but it's not far away enough for diseases like this (they share a border).
In hindsight you can say it was nothing. But it's all based on the information they had at that time. When a potentially infected passenger is in flight, you'll just have to quarantine everyone on the flight till you know things are OK.
What do you propose they do instead?
a) Let everyone including her go?
b) Check her while letting the rest of the passengers go?
What if it turns out she is infected with something nasty and contagious? You now have even more people to quarantine - since the other passengers would have travelled "everywhere" and come into contact with even more people.
They violated their own protocols.
Can you provide actual evidence that they violated their own protocols?
No reason? Based on the information they had, they had good reason to suspect the passenger might have a serious contagious disease from Uganda.
And that passenger was on a plane that will land in the USA. They hadn't seen the patient to be sure that she didn't have monkeypox OR something worse yet.
So do you actually propose letting the passenger and all of them head off to their various destinations without examining any of them? Yeah maybe the fire trucks were unnecessary (not sure why they were there) but the rest is reasonable assuming shit actually happened- cops to handle any uncooperative passengers, ambulances to take the cooperative infected away.
You'll be glad to know that I do not work in the medical field. However I also hope you do not work in the medical field either. You seem even more incompetent than I am when it comes to medical matters.
1) I didn't claim monkeypox was a good candidate for a pandemic.
2) Do you have any believable citations for your claim on how monkeypox spreads?
The following contradicts your claim:
http://www.who.int/mediacentre/factsheets/fs161/en/
Secondary transmission is human-to-human, resulting from close contact with infected respiratory tract excretions, with the skin lesions of an infected person or with recently contaminated objects. Transmission via droplet respiratory particles has also been documented. Transmission can also occur by inoculation or via the placenta (congenital monkeypox). There is no evidence to date that person-to-person transmission alone can sustain monkeypox in the human population.
Despite the last sentence, based on the rest of the paragraph I doubt you'd still want anyone who might have monkeypox to roam freely.
1) It's still a nasty disease to get
2) Some other animal might get it from the human, and monkeypox might be sustainable in that species.
Seems that Prairie Dogs can get it.
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5224a1.htm
http://wwwnc.cdc.gov/eid/article/10/3/03-0878_article.htm
And I doubt all the 87 infected humans were exposed to the blood of prairie dogs, or bitten.
Huh, overreaction? Do you really want monkey pox and other nasty diseases endemic in the Africa to not be rare in your country?
It's not overreaction at all. It's appropriate reaction. When the next super pandemic breaks out, the main solution is likely to be quarantine. The hospitals won't cope, and when the doctors and nurses start getting sick themselves it all falls apart and you'd be better off not going to the hospital for anything. Stay at home, and wait for the disease to kill all it can kill and/or evolve to be less lethal (that can happen in a short time - it can't spread if it kills too fast[1]).
[1] Quarantine in theory can actually work very well in breeding diseases to be less nasty and lethal.
Most of that "elections" game theory reasoning is only true if you are playing the game by assume there is only one election, and that's it. Then yes you vote for the 2nd place finisher. But if there's going to be more than one election it's a stupid idea, unless you genuinely would be more happy with the 2nd place finisher than the 3r place or other.
Because if people voted for the candidate they actually wanted rather than their second least hated candidate, then even if the 1st candidate still wins, the results could show everyone how many voters really wanted the other candidates. That sends a signal.
Then the voters in the next election might figure that one of the alternative candidates actually might have a chance, isn't so bad, and so vote for that candidate.
Alternatively the 2 Parties might realize that more and mroe people really want something different and change accordingly to try to ensure the other candidates don't win. While this does mean the 2 Parties might still win, the voters could still get the changes they wanted.
If you bunch are trying all that "game theory" crap when voting and are NOT succeeding, perhaps it's time to realize that the Two Parties are better at playing the game than you are. So stop doing that stupid shit which is not working.
Lastly: the first past the post system also works the same way for the "3rd candidate" if that candidate happens to get enough votes. Then suddenly the incumbent gets thrown out. That doesn't happen because either a) the voters stupidly think they are so smart at gaming the system or b) the voters actually prefer the Two Parties. Which may be true despite what people here might prefer to believe - there are certainly very many Obama fans (same goes for the R fans). Often it's as if it's their religion or similar. So it'll take a lot to convince them to vote differently.
I heard nowadays it's "pick one".