If simplicity is the benchmark, space itself is in no way evidence of design.
If there can be no evidence against the existence of the designer, there can be no evidence in support of the designer either.
Let the existence of the designer and making an observation be random variables. We say that x is evidence for y if p[y|x] > p[y]; that is, observing x makes y more likely than y is a priori. It's seen that if x is evidence for y, then x is evidence against (meaning s/>/</) the complement of y.
Let o1..on be the possible outcomes of the observation (all happening with positive probability) and let d be the outcome that the designer exists.
By assumption p[d|oi] >= p[d] for all i. Let e_i = p[d|oi] - p[d] for all i, and observe that e_i >= 0 for all i.
Clearly p[d] = sum(p[d]*p[oi] for all i), since sum(p[oi] for all i) = 1.
Now, we also have p[d] = sum(p[d|oi]*p[oi] for all i). But p[d] = sum(p[d|oi]*p[oi] for all i) >= sum(p[d]*p[oi] for all i) + sum(all e_i) = p[d] + sum(all e_i). Hence e_i must be 0 for all i.
This means there can be no evidence for the designer unless there can also be evidence against the designer.
He sells the same message as RMS, more or less, but he sells it with a different tone. Give him a listen in your next one-hour drive to work (and continue on the way back), or whenever you feel like it.
we need somebody who will make dramatic changes to re-align the law with common sense.
What's common sense, here?
I think the common-sense thing to do is to repeal and enact a set of laws that will be of maximum benefit to the American People, based on the best available evidence and well-tested economic models.
Basing judgments on evidence and theoretical understanding is how we do software, medicine, aviation and construction, to name a few. I think evidence should play a role in governance as well.
I don't have any hard evidence to back up any particular position. You didn't present any---do you have any?
I have a bit of soft evidence: people have been copying music whenever they had the technology to do it. Back in the old days, it was cassettes. Then it was CDs, and now it's mp3s copied over the internet. The only thing that's changed is that we've all gotten a few billion more friends to copy from.
Dragnet litigation and copyright propaganda video before every movie doesn't seem to have changed peoples' inclination to copy media. No DRM scheme has seemed to be effective in stopping people who really want to copy media.
I suggest that (as an experimental policy) it becomes permitted to do verbatim copying of all copyrighted works for non-commercial purposes.
If big chunks of the "copyright industries" die off and people have a shortage of music on the radio, in the tube and in the tubes, and the shortage persists over a long time as people try out new business models, then this would be evidence that copying is bad when we just let people do it.
If people are happy about the amount and quality of new music coming out and the price of concert tickets in the new model, then that's evidence that we as a society will pay musicians enough money to make them satisfy society's needs while still doing the sharing and copying we want.
and play non-violent games (they're both complete Pokemon addicts)
Does the phrase "Poke-battle" mean anything to you? These creatures spew lightning at each other. Team Rocket kidnaps your precious poke-monsters and do horrible things to them. I know this from not even trying to watch the show. My xgf played (some of) the games; I wouldn't exactly call them non-violent. It's cartoon violence, but that still effects kids. I recall a guest on The Skeptics' Guide to the Universe (http://www.theskepticsguide.org) saying something to that effect; with a little digging I'm sure you could look it up. Go listen to the podcast anyways, it's great news for nerds---which includes the hosts*:)
And let's not even talk about "Try to bamb tha harbar" and the primary main objective;)
* Counting an academic neurologist, a software engineer, an IT staffer, a TV/video production guy (i.e. professional card-carrying member of the A/V-club:D) and a hippie lib-arts chick (not libarts or libartsc0-dev, you dork). They talk about physics, medicine, philosophy of science, astronomy, and whatever their guests are into. James Randi is on every once in a while.
He might even be the one providing work to his buddies:)
I'd assume plumbers need to know a little bit about building codes; or at least know through some kind of oracle [let's call that oracle a lawyer] that their plan for laying down pipes and shit is A-OK with the gubbermint.
If his buddies charge reasonable fees, he might want to hire them.
One great rule of thumb that has really been hammered home---at least that's what I got out of it---by listening to the Skeptics' guide to the Universe (www.theskepticsguide.org) is that you should always say this to yourself:
"Is what this person saying based on their preconceived notions, or is it based on evidence? Trust evidence over faith, opinion and preconceived notions."
You may be right in what you say---I can't confidently say I know you're wrong. But you provide no evidence backing up your assertion.
Are there any scientific studies that examine this node of the decision tree of life?
I'm not saying that science holds the answer to every problem; not even to every problem in its domain. You always have to take the current situation into consideration when you want to apply knowledge obtained through science.
But: I want science rather than opinion, where possible, because it tends to work better.
While I'm all for transparency in spending of my tax money, sooner or later you have to stop micromanaging and let people do their job. Otherwise, you will get less for your money.
I'm paid tax money in my current occupation (CS student). I'm hoping there's enough oversight that you can trust that I'll be thrown out on my ass if I'm not worth the money I'm paid, without me having to be held accountable to other people than my advisors all the way.
The LHC is much more notable than some random guy who hasn't submitted a paper for publication yet, of course it is, but is it qualitatively different in any other way?
they can prevent you from using the domain knowledge you acquired on the job.
I don't understand this.
Say I'm an inexperienced audio programmer and they give me a crash course in sound physics and psychoacoustics. I can't program sound for their competitor now?
You can put yourself on his fan list (http://slashdot.org/~NewYorkCountryLawyer/fans) and give him an instant +6 Friend moderation:)
On it you can read the following bit, in which Ray's words mesh humorously with the text of the slashdot link:
Nothing I say on Slashdot should be construed as legal advice; for that you need to consult with a lawyer with whom you have a one-on-one relationship. Change your relationship with NewYorkCountryLawyer
I'm intrigued by your ideas, and wish to purchase your latest edition of your femaleish-maleish dictionary;)
I'm hoping you tell or have told you boyfriend what you just told us here on slashdot.
And to all the girls, here's how guys communicate: we (almost) always interpret what is said in the most straightforward, say-what-you-mean nothing-between-the-lines way. When I say I have a headache, it's because my head hurts (dammit), not because I sad I didn't get the promotion.
Successful communication requires one or more parties to move out of their comfort zone. It's probably for the best if both do: guys need to read between the lines a bit more, and girls need to tell the guys when and how to read between the lines.
But speaking as a guy, understanding girls isn't made easier by the fact that "I have a headache" sometimes means "my head hurts; end of news bulletin" and sometimes means "I refuse to have sex with you due to something you did within the last month that I won't tell you about" and sometimes "my head hurts really badly; please give me an aspirin and a really big hug."
If you talk to guys the guy way, and tell them flat out what you want, they'll either agree straight away [who doesn't want to hug and comfort their hurting girlfriend?], or at least the two of you will know that you both know that there's an issue to be addressed and what it is. In the latter case, you at least have an easier time talking about solutions than if only one party knows what the problem is.
And also for those who only speak their native language and English. Or their native language, Russian and English. Or the seven different languages spoken natively and locally on their non-Europe continent, and English.
Is/. really so far behind times that it can't handle Unicode?
I'd like to confidently say that parent shows up as "wÃf¼nsche", but I don't know how that string looks to you.
The US soldiers in the 1991 Gulf War were slightly safer than if they'd been home as civilians.
So what you're saying is that if soldiers stay at home and we send robots instead, the decrease in US body count will be negative, right?
So unless robots mean the US goes to war a lot less, it'll mean more loss of human life.
I'm not sure I want to argue that we should people into war for the safety of it; on the other hand, the facts might want to, if they are as you say, and they decide what's real:|
A properly programmed robot has the ability to appear compassionate and yet treat the situation skeptically until they know for sure the target is or is not a threat.
It does, although I have some additions and disagreements.
They characterize the spoofability of DNS replies as a flaw in UDP. I think that's incorrect. UDP isn't marketed as a data integrity protocol, it's marketed as a transport protocol. That job it does fine. TCP is the same thing: a transport protocol.
A blind attacker against UDP has to guess a source port and a transaction ID. A blind attacker against TCP has to guess an initial sequence number as well. If you use SYN cookies (http://cr.yp.to/syncookies.html), that means 24 extra bits of randomness, for 56 in total. It's better than 32 (which takes 10 hours), but again: this is against blind attackers [those not in the middle].
If you worry about men in the middle (as the paper does), TCP offers you no bonus: the man in the middle simply spoofs the receiving server's SYN+ACK and DNS reply.
As an addition: with any DNS, you have to trust not only that the servers you get replies from are those you should be getting replies from, but also that they act how their customers want them to.
That is, you have to trust them to give the right answers. DNSSEC doesn't ensure that when.com gives you the key for google.com, it really is the key google wanted.com to give you. It only gives you a key for google.com which you can check has been used to sign the records for google.com;.com could easily give you their key and the altered google.com records signed with th.com key.
That taken into consideration, we have to trust the DNS servers. If we have authentic communication with them (through DNSCurve), we don't need the records to be stored in signed form, since they are signed during the transit and we trust the servers to give us the correct records.
However, we still need some way of knowing the public key for the DNS servers we wish to contact.
In summary:
Using UDP currently makes the easiest form of attack not too easy; moving to TCP makes it quite hard, but ultimately won't fix that problem.
Moving to TCP won't fix the man-in-the-middle
Using DNSCurve doesn't fixed the most important issue: that if we only have one CA (the root), we have all the problems of only having a single CA
I think the latter point is more problematic.
The attacks worth carrying out to a degree where it causes people real problems are either against single big targets (DDos or break-in), or those easy to carry out in big numbers where you can extract money from your victim (Web Bank Phishing). The first doesn't affect individual citizens, and the second is handled through SSL. What's left to fix in DNS again?
A cute side-note about DNSCurve: it uses the disemvoweled base32 alphabet (see http://en.wikipedia.org/wiki/Base32, grep for NVRAM and nintendo), but it allows 'u' instead of, say, '-' (which is a valid DNS character). I would like to use the public key fuckfuckfuckfuckfuckfuckfuck.jonaskoelker.some-dns.com;)
How does DNSSEC stop the browser from giving Joe User a warning box that the https cert is not signed by a recognized CA?
That's the only real reason why you pay CAs to sign your certs - to stop Joe User from being bothered it.
You don't need the CAs, once applications are rewritten to grab keys from the DNS instead.
Using DNS as a PKI means that my DNS provider is now my CA. If I grab jonaskoelker.free-dns.com and I start out with only a trusted root key, I can learn free-dns's key and trust them. I can then securely send them my key, which they sign for free, along with my signed records.
Then, when you go to jonas.free-dns.com with a modified firefox, that firefox will trust the DNS key for jonas.free-dns.com as an SSL key for jonas.free-dns.com as well, and you'll trust that the guy whose server you're talking to is the same guy as the one who got the name in the first place.
With a changed Firefox, you won't need a CA.
Now, changing how "we" (meaning our browsers) decide whether to trust a site may not be easy, but it can be done.
If your DNS parent is com, all I can say is "Meet your new CA, same as the old CA";)
Slashdot Mirror
If simplicity is the benchmark, space itself is in no way evidence of design.
If there can be no evidence against the existence of the designer, there can be no evidence in support of the designer either.
Let the existence of the designer and making an observation be random variables. We say that x is evidence for y if p[y|x] > p[y]; that is, observing x makes y more likely than y is a priori. It's seen that if x is evidence for y, then x is evidence against (meaning s/>/</) the complement of y.
Let o1..on be the possible outcomes of the observation (all happening with positive probability) and let d be the outcome that the designer exists.
By assumption p[d|oi] >= p[d] for all i. Let e_i = p[d|oi] - p[d] for all i, and observe that e_i >= 0 for all i.
Clearly p[d] = sum(p[d]*p[oi] for all i), since sum(p[oi] for all i) = 1.
Now, we also have p[d] = sum(p[d|oi]*p[oi] for all i). But p[d] = sum(p[d|oi]*p[oi] for all i) >= sum(p[d]*p[oi] for all i) + sum(all e_i) = p[d] + sum(all e_i). Hence e_i must be 0 for all i.
This means there can be no evidence for the designer unless there can also be evidence against the designer.
On http://audio-video.gnu.org/audio/ you can find speeches from the FSF. Most of them by RMS, but there are two by Bradley M. Kuhn.
He sells the same message as RMS, more or less, but he sells it with a different tone. Give him a listen in your next one-hour drive to work (and continue on the way back), or whenever you feel like it.
we need somebody who will make dramatic changes to re-align the law with common sense.
What's common sense, here?
I think the common-sense thing to do is to repeal and enact a set of laws that will be of maximum benefit to the American People, based on the best available evidence and well-tested economic models.
Basing judgments on evidence and theoretical understanding is how we do software, medicine, aviation and construction, to name a few. I think evidence should play a role in governance as well.
I don't have any hard evidence to back up any particular position. You didn't present any---do you have any?
I have a bit of soft evidence: people have been copying music whenever they had the technology to do it. Back in the old days, it was cassettes. Then it was CDs, and now it's mp3s copied over the internet. The only thing that's changed is that we've all gotten a few billion more friends to copy from.
Dragnet litigation and copyright propaganda video before every movie doesn't seem to have changed peoples' inclination to copy media. No DRM scheme has seemed to be effective in stopping people who really want to copy media.
I suggest that (as an experimental policy) it becomes permitted to do verbatim copying of all copyrighted works for non-commercial purposes.
If big chunks of the "copyright industries" die off and people have a shortage of music on the radio, in the tube and in the tubes, and the shortage persists over a long time as people try out new business models, then this would be evidence that copying is bad when we just let people do it.
If people are happy about the amount and quality of new music coming out and the price of concert tickets in the new model, then that's evidence that we as a society will pay musicians enough money to make them satisfy society's needs while still doing the sharing and copying we want.
and play non-violent games (they're both complete Pokemon addicts)
Does the phrase "Poke-battle" mean anything to you? These creatures spew lightning at each other. Team Rocket kidnaps your precious poke-monsters and do horrible things to them. I know this from not even trying to watch the show. My xgf played (some of) the games; I wouldn't exactly call them non-violent. It's cartoon violence, but that still effects kids. I recall a guest on The Skeptics' Guide to the Universe (http://www.theskepticsguide.org) saying something to that effect; with a little digging I'm sure you could look it up. Go listen to the podcast anyways, it's great news for nerds---which includes the hosts* :)
And let's not even talk about "Try to bamb tha harbar" and the primary main objective ;)
* Counting an academic neurologist, a software engineer, an IT staffer, a TV/video production guy (i.e. professional card-carrying member of the A/V-club :D) and a hippie lib-arts chick (not libarts or libartsc0-dev, you dork). They talk about physics, medicine, philosophy of science, astronomy, and whatever their guests are into. James Randi is on every once in a while.
Give a man a fish, and you have fed him for today. Take him to Frys and you have fed him Fish Frys for a lifetime.
He might even be the one providing work to his buddies :)
I'd assume plumbers need to know a little bit about building codes; or at least know through some kind of oracle [let's call that oracle a lawyer] that their plan for laying down pipes and shit is A-OK with the gubbermint.
If his buddies charge reasonable fees, he might want to hire them.
[IBM Model M keyboard. I bet you could vomit into those and just clean them]
My boss doesn't think they need to be cleaned ):
One great rule of thumb that has really been hammered home---at least that's what I got out of it---by listening to the Skeptics' guide to the Universe (www.theskepticsguide.org) is that you should always say this to yourself:
"Is what this person saying based on their preconceived notions, or is it based on evidence? Trust evidence over faith, opinion and preconceived notions."
You may be right in what you say---I can't confidently say I know you're wrong. But you provide no evidence backing up your assertion.
Are there any scientific studies that examine this node of the decision tree of life?
I'm not saying that science holds the answer to every problem; not even to every problem in its domain. You always have to take the current situation into consideration when you want to apply knowledge obtained through science.
But: I want science rather than opinion, where possible, because it tends to work better.
Go read http://en.wikipedia.org/wiki/Internet_Control_Message_Protocol
Each ICMP message is encapsulated directly within a single IP datagram
There's no TCP. There's no ACK. If you tried using TCP, as your child said, you'd get an RST.
The correct reply would be an ICMP Echo Reply; an ICMP packet of type 0 (also on the wikipedia page).
I hope what I'm saying isn't news to nerds... ;)
You need the system status to be shown in a clear and most importantly live way.
Grep for "No, that's the last one. What is all this junk?".
And the caps really aren't that bad
$ units
you have: 250 gibibytes per month
you want: mebibit per second
0.77878308
"Really not that bad"? It's barely enough just for my porn feed! :(
Karma whoring link
Mod parent Funny :p
This is a work funded by taxpaid dollars, so it should be kept open and transparent.
What, you expect to read about day-to-day work on http:///planet.lhc.cern.ch/? ;)
While I'm all for transparency in spending of my tax money, sooner or later you have to stop micromanaging and let people do their job. Otherwise, you will get less for your money.
I'm paid tax money in my current occupation (CS student). I'm hoping there's enough oversight that you can trust that I'll be thrown out on my ass if I'm not worth the money I'm paid, without me having to be held accountable to other people than my advisors all the way.
The LHC is much more notable than some random guy who hasn't submitted a paper for publication yet, of course it is, but is it qualitatively different in any other way?
2.1) Will free-dns sign any key for jonaskoelker.free-dns.com from anyone claiming to be Jonas Koelker? How do they know it's the right Jonas?
Ah... we need to be able to store an "identity verified" bit in DNS...
they can prevent you from using the domain knowledge you acquired on the job.
I don't understand this.
Say I'm an inexperienced audio programmer and they give me a crash course in sound physics and psychoacoustics. I can't program sound for their competitor now?
Subject
Subject yourself to what? ;)
You can put yourself on his fan list (http://slashdot.org/~NewYorkCountryLawyer/fans) and give him an instant +6 Friend moderation :)
On it you can read the following bit, in which Ray's words mesh humorously with the text of the slashdot link:
Nothing I say on Slashdot should be construed as legal advice; for that you need to consult with a lawyer with whom you have a one-on-one relationship.
Change your relationship with NewYorkCountryLawyer
girlspeak translation: ...
Thank you.
I'm intrigued by your ideas, and wish to purchase your latest edition of your femaleish-maleish dictionary ;)
I'm hoping you tell or have told you boyfriend what you just told us here on slashdot.
And to all the girls, here's how guys communicate: we (almost) always interpret what is said in the most straightforward, say-what-you-mean nothing-between-the-lines way. When I say I have a headache, it's because my head hurts (dammit), not because I sad I didn't get the promotion.
Successful communication requires one or more parties to move out of their comfort zone. It's probably for the best if both do: guys need to read between the lines a bit more, and girls need to tell the guys when and how to read between the lines.
But speaking as a guy, understanding girls isn't made easier by the fact that "I have a headache" sometimes means "my head hurts; end of news bulletin" and sometimes means "I refuse to have sex with you due to something you did within the last month that I won't tell you about" and sometimes "my head hurts really badly; please give me an aspirin and a really big hug."
If you talk to guys the guy way, and tell them flat out what you want, they'll either agree straight away [who doesn't want to hug and comfort their hurting girlfriend?], or at least the two of you will know that you both know that there's an issue to be addressed and what it is. In the latter case, you at least have an easier time talking about solutions than if only one party knows what the problem is.
if you give me the choice between hanging out with a bunch of kids that are interested in math or playing video games, the choice is obvious.
Cool, let's do some math together!
Oh, you... oh... :(
For all the monoglots out there
And also for those who only speak their native language and English. Or their native language, Russian and English. Or the seven different languages spoken natively and locally on their non-Europe continent, and English.
Is /. really so far behind times that it can't handle Unicode?
I'd like to confidently say that parent shows up as "wÃf¼nsche", but I don't know how that string looks to you.
The US soldiers in the 1991 Gulf War were slightly safer than if they'd been home as civilians.
So what you're saying is that if soldiers stay at home and we send robots instead, the decrease in US body count will be negative, right?
So unless robots mean the US goes to war a lot less, it'll mean more loss of human life.
I'm not sure I want to argue that we should people into war for the safety of it; on the other hand, the facts might want to, if they are as you say, and they decide what's real :|
A properly programmed robot has the ability to appear compassionate and yet treat the situation skeptically until they know for sure the target is or is not a threat.
It still makes sense...
Thanks for reading,
M B Dyson
You're a mindless jerk who'll be first up against the wall when the revolution comes. Oh wait, you're not in the marketing division? ;)
seemed well thought out
It does, although I have some additions and disagreements.
They characterize the spoofability of DNS replies as a flaw in UDP. I think that's incorrect. UDP isn't marketed as a data integrity protocol, it's marketed as a transport protocol. That job it does fine. TCP is the same thing: a transport protocol.
A blind attacker against UDP has to guess a source port and a transaction ID. A blind attacker against TCP has to guess an initial sequence number as well. If you use SYN cookies (http://cr.yp.to/syncookies.html), that means 24 extra bits of randomness, for 56 in total. It's better than 32 (which takes 10 hours), but again: this is against blind attackers [those not in the middle].
If you worry about men in the middle (as the paper does), TCP offers you no bonus: the man in the middle simply spoofs the receiving server's SYN+ACK and DNS reply.
As an addition: with any DNS, you have to trust not only that the servers you get replies from are those you should be getting replies from, but also that they act how their customers want them to.
That is, you have to trust them to give the right answers. DNSSEC doesn't ensure that when .com gives you the key for google.com, it really is the key google wanted .com to give you. It only gives you a key for google.com which you can check has been used to sign the records for google.com; .com could easily give you their key and the altered google.com records signed with th .com key.
That taken into consideration, we have to trust the DNS servers. If we have authentic communication with them (through DNSCurve), we don't need the records to be stored in signed form, since they are signed during the transit and we trust the servers to give us the correct records.
However, we still need some way of knowing the public key for the DNS servers we wish to contact.
In summary:
I think the latter point is more problematic.
The attacks worth carrying out to a degree where it causes people real problems are either against single big targets (DDos or break-in), or those easy to carry out in big numbers where you can extract money from your victim (Web Bank Phishing). The first doesn't affect individual citizens, and the second is handled through SSL. What's left to fix in DNS again?
A cute side-note about DNSCurve: it uses the disemvoweled base32 alphabet (see http://en.wikipedia.org/wiki/Base32, grep for NVRAM and nintendo), but it allows 'u' instead of, say, '-' (which is a valid DNS character). I would like to use the public key fuckfuckfuckfuckfuckfuckfuck.jonaskoelker.some-dns.com ;)
You'll still need CAs.
How does DNSSEC stop the browser from giving Joe User a warning box that the https cert is not signed by a recognized CA?
That's the only real reason why you pay CAs to sign your certs - to stop Joe User from being bothered it.
You don't need the CAs, once applications are rewritten to grab keys from the DNS instead.
Using DNS as a PKI means that my DNS provider is now my CA. If I grab jonaskoelker.free-dns.com and I start out with only a trusted root key, I can learn free-dns's key and trust them. I can then securely send them my key, which they sign for free, along with my signed records.
Then, when you go to jonas.free-dns.com with a modified firefox, that firefox will trust the DNS key for jonas.free-dns.com as an SSL key for jonas.free-dns.com as well, and you'll trust that the guy whose server you're talking to is the same guy as the one who got the name in the first place.
With a changed Firefox, you won't need a CA.
Now, changing how "we" (meaning our browsers) decide whether to trust a site may not be easy, but it can be done.
If your DNS parent is com, all I can say is "Meet your new CA, same as the old CA" ;)