bah. I have seen a few people intercept SSH before but only at demonstrations. I knew all of these guys and they said they have never wanted for accounts - there's enough unencrypted traffic to not bother going after the encrypted traffic. If there is one box that no-one connects to without using ssh, it is almost always connected to from an insecure box and, at present, there is nothing to stop tty sniffing. I wont even bother mentioning people who use windoze ssh clients. On most "secure" networks, ssh is the strong link in the weak chain. As for SSL, I have never seen an intercept of SSL by anyone who didn't have the SSL certificate.
Re:BEAM is the Cold Fusion of Robotics
on
The Robot Diaries
·
· Score: 2
I think this is the point. This guy was saying that BEAM = purely reactive robots, and he was saying that that on it's own wasn't enough. He has said a couple of messages ago that most robots have a purely reactive component to them which does what kit BEAM robots do but it has a highlevel system to guide them. Obviously you two guys are in agreement. Our insulting friend two messages ago however, seems to believe that highlevel systems like pathfinding are useless and that purely reactive robots (esp BEAM robots) can do all the things that their more complex cousins can do. Which is all well and good, but can he point to some examples? No.
I'm sitting on the fence saying, well, I don't see much of a use for either class of robot. Which do I find more interesting? BEAM, because their electronics are simple enough that they appear to be similar to living things. Why is that good? Because we can study living things to figure out new ways to make robots. As for the task of delivering a box in a hospital, well I'm afraid I only know of one level of intelligence that can do than to any significant degree of efficency: human intelligence. Can you train a cockroach to deliver parcels? No, so why would you think you can build a robot to do it. I'd more be looking at things that we get dogs to do, like sniff out bombs and drugs. Can we make a robot with dog level intelligence. No, not yet, but at least it is something to aim for. The dream of domestic robots which vacuum your floor, this is something you could imagine that you might be able to train a dog to do. Not well, but it's not inconceivable.
no I think you totally missed my point, the infastructure of the Internet is more akin to a public bus station than private property. Even if you own said infastructure, by connecting it to the web you really are giving permision to come onto it. So maybe we need an even more elaborate analogy (because they just work so well) like maybe if you have put a sign up at your front gate which says "visitors welcome", thus allowing people onto your property but only allow people into your front living room (akin to your web server) but people start trying the doors to the bathroom and the bedroom. Even if they find one unlocked they're not trespassing until they actually go in there, and even then you're on shaky ground. But if they then start smashing stuff up or sleeping in your bed or using your shower, you've definitely got something to complain about.
Re:BEAM is the Cold Fusion of Robotics
on
The Robot Diaries
·
· Score: 1
well in that case I don't think purely reactive robotics is too much out of the question, although obviously some path finding algorithm on top of a purely reactive system would be more efficient. I think the claim by BEAM enthusiests would be that their robots would not get "stuck" whereas more CPU dominated designs would.
Re:BEAM is the Cold Fusion of Robotics
on
The Robot Diaries
·
· Score: 2
k, so in your example of delivering an object, are there any other systems in place, like say an indicator on the doors which transmit a signal to recognise the room? Or are we asking our robot designer here to recognise arbitary things and learn which room is which?
Re:BEAM is the Cold Fusion of Robotics
on
The Robot Diaries
·
· Score: 2
no really, I've read the previous posts. I don't see any specific examples of what you are toting as useful applications of autonomous robots. BEAM is not about abstract ideas.. Tilden has said many times that we have to stop thinking about what's the best way to do things and just start making robots.. lots of them, and let's start scaling these things up to something that is apreciably intelligent. At the moment BEAM (and all other robots) leaves a lot to be desired. Just the concept of a toy puppy that children would like to play with is difficult at present. That's kind of strange isn't it? How hard is it to make a furry mutt that can bite your fingers, chase a ball, etc. The answer, unfortunately is hard, very hard. These are things that I find useful, so my question to you is: What do you find useful things to make autonomous robots to do?
entering a house even if the door is open is still breaking and entering, just as staggering down the road after you've been drinking is being drunk and disorderly, even if you're not disorderly.
and the police would propably tell you that all they can do is charge him with trespass because he's on your property.. however, if he was say, checking the doors and windows on a bus station (a publicly owned bus station) the police couldn't do a damn thing.
Re:BEAM is the Cold Fusion of Robotics
on
The Robot Diaries
·
· Score: 2
can you give examples of things that existing autonomous robots can do and, if possible, why you think BEAM cant?
that's seriously demented. Does it do anything remotely useful? Like if I throw the ball and the bot dont go running after it can I note the time and look at the diary that night to find out why?
still, I would really love to see a BEAM robot teach it's emergant behaviour to an offspring. And if this behaviour emerged spontaniously, what would that say about our creations?
Hey man.. go grab nano, a clone of pico, it rocks man.. it is so great, it has search and replace, go to line (the two big ones) and a couple of other cool features. It even has a "pico emulation mode" where it looks and acts exactly like pico. So if you don't want a multi-megabyte editor or an editor with obsficated key commands, go grab nano.
no it's real, Be is currently in deliberations as we speak. They are teaming up with Opera to release a free (as in software) distribution that will be sent out with AOL cd's. [LIE MODE OFF]
but what exactly do you have against the CIA/FBI identifying and stopping terrorists and the like? I mean, what you're saying here is that people should use encryption so there are so many encrypted messages that the agencies we pay to protect us can't decrypt all of them. I mean really, what these keyword searches are doing is picking out of the pile all the messages that may be validly scary. Then, assumably, someone looks at these messages and determines whether or not there really is something here. Turn on the encryption for every little thing and that small bit of security disappears. Obviously not a big thing, but it could be a crucial issue in your national security. Now the things that I personally have a problem with, is the companies and marketing jerks who are doing this sort of monitoring. They arn't doing keyword searches, their goal is to build a very big database with every little bit of information they can get on you. Presumably this is to protect their intellectual property or sell you stuff, two things I really don't want them doing. So if there ever was a reason to encrypt everything you send it is to protect you from these ilk (criminals?) and not your own intelligence agencies.
bah.. you would have to release these new changes as a library that people would have to download seperate from the main application. To justify this your changes would have to be something useful outside of this application. For example, if you discovered that Mozilla doesn't support your new graphics file format and you want to allow people to view files in this new format you might write a library and hack the changes into Mozilla which link to this library. People would have to download Mozilla and then download your library seperately. But your library can be used with more things than just Mozilla, any program can support your new file format by including your library. In ways, this is a bad thing, in other ways it is a good thing.
you "obviously" have no argument. Bugs, yes, blatant security flaws, no.. that's irresponsible and inexcusable. We're not talking about "default password is null" bugs here. We're talking about "copies data into stack based buffer without checking source size", the same security bugs that have been reported in code since Zardoz. And I don't hide bugs from my supervisors.. we have a development process that is realistic and structured to handle the fixing of bugs, not the hiding of them. That's enough troll food for one day.
ok, so maybe you're going off the deep end here, but I think it is very possible that this could have an adverse impact on the security industry. Advisory services (and even individual developers) are judged on the timelyness and accuracy of their alerts. When an administrator has to make a decision about how serious an advisory is, they look at the reputation of the advisor. If advisors have the right to change their advisories after the fact and prohibit offsite archiving (sort of like rewriting history) they are beyond retrospective analysis. A security expert may make the claim that someone released an inaccurate or late advisory but without a trustworthy archive to point at, they can't even claim that the advisory has been updated since first release. This seriously undermines peer review which is a cornerstone of software security.
Slashdot Mirror
isn't SSH2 non-free?
bah. I have seen a few people intercept SSH before but only at demonstrations. I knew all of these guys and they said they have never wanted for accounts - there's enough unencrypted traffic to not bother going after the encrypted traffic. If there is one box that no-one connects to without using ssh, it is almost always connected to from an insecure box and, at present, there is nothing to stop tty sniffing. I wont even bother mentioning people who use windoze ssh clients. On most "secure" networks, ssh is the strong link in the weak chain. As for SSL, I have never seen an intercept of SSL by anyone who didn't have the SSL certificate.
I think this is the point. This guy was saying that BEAM = purely reactive robots, and he was saying that that on it's own wasn't enough. He has said a couple of messages ago that most robots have a purely reactive component to them which does what kit BEAM robots do but it has a highlevel system to guide them. Obviously you two guys are in agreement. Our insulting friend two messages ago however, seems to believe that highlevel systems like pathfinding are useless and that purely reactive robots (esp BEAM robots) can do all the things that their more complex cousins can do. Which is all well and good, but can he point to some examples? No.
I'm sitting on the fence saying, well, I don't see much of a use for either class of robot. Which do I find more interesting? BEAM, because their electronics are simple enough that they appear to be similar to living things. Why is that good? Because we can study living things to figure out new ways to make robots. As for the task of delivering a box in a hospital, well I'm afraid I only know of one level of intelligence that can do than to any significant degree of efficency: human intelligence. Can you train a cockroach to deliver parcels? No, so why would you think you can build a robot to do it. I'd more be looking at things that we get dogs to do, like sniff out bombs and drugs. Can we make a robot with dog level intelligence. No, not yet, but at least it is something to aim for. The dream of domestic robots which vacuum your floor, this is something you could imagine that you might be able to train a dog to do. Not well, but it's not inconceivable.
no I think you totally missed my point, the infastructure of the Internet is more akin to a public bus station than private property. Even if you own said infastructure, by connecting it to the web you really are giving permision to come onto it. So maybe we need an even more elaborate analogy (because they just work so well) like maybe if you have put a sign up at your front gate which says "visitors welcome", thus allowing people onto your property but only allow people into your front living room (akin to your web server) but people start trying the doors to the bathroom and the bedroom. Even if they find one unlocked they're not trespassing until they actually go in there, and even then you're on shaky ground. But if they then start smashing stuff up or sleeping in your bed or using your shower, you've definitely got something to complain about.
well in that case I don't think purely reactive robotics is too much out of the question, although obviously some path finding algorithm on top of a purely reactive system would be more efficient. I think the claim by BEAM enthusiests would be that their robots would not get "stuck" whereas more CPU dominated designs would.
k, so in your example of delivering an object, are there any other systems in place, like say an indicator on the doors which transmit a signal to recognise the room? Or are we asking our robot designer here to recognise arbitary things and learn which room is which?
no really, I've read the previous posts. I don't see any specific examples of what you are toting as useful applications of autonomous robots. BEAM is not about abstract ideas.. Tilden has said many times that we have to stop thinking about what's the best way to do things and just start making robots.. lots of them, and let's start scaling these things up to something that is apreciably intelligent. At the moment BEAM (and all other robots) leaves a lot to be desired. Just the concept of a toy puppy that children would like to play with is difficult at present. That's kind of strange isn't it? How hard is it to make a furry mutt that can bite your fingers, chase a ball, etc. The answer, unfortunately is hard, very hard. These are things that I find useful, so my question to you is: What do you find useful things to make autonomous robots to do?
heh.. that's leet.. love the @ thang.
entering a house even if the door is open is still breaking and entering, just as staggering down the road after you've been drinking is being drunk and disorderly, even if you're not disorderly.
and the police would propably tell you that all they can do is charge him with trespass because he's on your property.. however, if he was say, checking the doors and windows on a bus station (a publicly owned bus station) the police couldn't do a damn thing.
can you give examples of things that existing autonomous robots can do and, if possible, why you think BEAM cant?
that's seriously demented. Does it do anything remotely useful? Like if I throw the ball and the bot dont go running after it can I note the time and look at the diary that night to find out why?
still, I would really love to see a BEAM robot teach it's emergant behaviour to an offspring. And if this behaviour emerged spontaniously, what would that say about our creations?
why not? Your TV and the Internet babysit your kids now..
fucking werd up
my favorite missing feature in pine ;)
Hey man.. go grab nano, a clone of pico, it rocks man.. it is so great, it has search and replace, go to line (the two big ones) and a couple of other cool features. It even has a "pico emulation mode" where it looks and acts exactly like pico. So if you don't want a multi-megabyte editor or an editor with obsficated key commands, go grab nano.
no it's real, Be is currently in deliberations as we speak. They are teaming up with Opera to release a free (as in software) distribution that will be sent out with AOL cd's. [LIE MODE OFF]
no actually that's exactly what I ment.. I read a lot of scientific papers that totally own Bruce Perens.
but what exactly do you have against the CIA/FBI identifying and stopping terrorists and the like? I mean, what you're saying here is that people should use encryption so there are so many encrypted messages that the agencies we pay to protect us can't decrypt all of them. I mean really, what these keyword searches are doing is picking out of the pile all the messages that may be validly scary. Then, assumably, someone looks at these messages and determines whether or not there really is something here. Turn on the encryption for every little thing and that small bit of security disappears. Obviously not a big thing, but it could be a crucial issue in your national security. Now the things that I personally have a problem with, is the companies and marketing jerks who are doing this sort of monitoring. They arn't doing keyword searches, their goal is to build a very big database with every little bit of information they can get on you. Presumably this is to protect their intellectual property or sell you stuff, two things I really don't want them doing. So if there ever was a reason to encrypt everything you send it is to protect you from these ilk (criminals?) and not your own intelligence agencies.
I read a lot of scientific paper's Bruce, I really think it is similar.
bah.. you would have to release these new changes as a library that people would have to download seperate from the main application. To justify this your changes would have to be something useful outside of this application. For example, if you discovered that Mozilla doesn't support your new graphics file format and you want to allow people to view files in this new format you might write a library and hack the changes into Mozilla which link to this library. People would have to download Mozilla and then download your library seperately. But your library can be used with more things than just Mozilla, any program can support your new file format by including your library. In ways, this is a bad thing, in other ways it is a good thing.
you "obviously" have no argument. Bugs, yes, blatant security flaws, no.. that's irresponsible and inexcusable. We're not talking about "default password is null" bugs here. We're talking about "copies data into stack based buffer without checking source size", the same security bugs that have been reported in code since Zardoz. And I don't hide bugs from my supervisors.. we have a development process that is realistic and structured to handle the fixing of bugs, not the hiding of them. That's enough troll food for one day.
they know it so "inside out" that they wrote big security bugs into it.. thus the reason why we are talking about this!
yes.. and Sprite sponsors 90210, they didn't make the show but they "bring" it to us.
ok, so maybe you're going off the deep end here, but I think it is very possible that this could have an adverse impact on the security industry. Advisory services (and even individual developers) are judged on the timelyness and accuracy of their alerts. When an administrator has to make a decision about how serious an advisory is, they look at the reputation of the advisor. If advisors have the right to change their advisories after the fact and prohibit offsite archiving (sort of like rewriting history) they are beyond retrospective analysis. A security expert may make the claim that someone released an inaccurate or late advisory but without a trustworthy archive to point at, they can't even claim that the advisory has been updated since first release. This seriously undermines peer review which is a cornerstone of software security.