...it will be a huge test for the OSS community, to see if they can really handle these problems as well as they always say they can.
I guess you've never heard of Apache. 68.8% market share ought to be a good test of the OSS development model. Of course that's not to say that all OSS will fare as well as the carefully developed Apache project, but saying the whole model is untested... is simply uninformed.
Any JNI application is essentially running in "unsafe mode" within the VM. The VM has no mechanism to protect itself from a JNI application it is running. I don't find that to be much different from running "unsafe" applications.
With JNI, due to the complexity of the glue pushed on the native application, it's quite easy to mess up the VM. Java is not a bad language, but I find JNI to be one of the most unweildy native interfaces I've seen, even compared with those for functional languages.
No, he's referring to this story: GNU/Hurd Web Server Online I remember that slashdotting the first time around... pretty funny to link a beta OS up like that to be taken down in minutes by the hoard.
Err, you're the one who didn't get it; It's called "humor" on this planet. You won't think its funny though, rather like any non-liberal reading Doonesbury.
"Monkey" isn't an accurate description in either case though; Bush looks more like a chimpanzee, while Kerry looks like an ape or wookie.
Well I won't be reading CDDL code for exactly that reason. It'll just support a claim of "Willfull infringement" on anything you write later that accidentally steps on a patent. I enjoyed using Solaris in the past, but userspace is where I spend my time, and that's where Linux has really taken off compared to Unix.
Well, assuming they have more than 1600 patents, they could just claim the patents they are suing you for aren't in that list. In other words, without an explicit list of patents, this press release won't help anyone sued by Sun.
Well, by ready I would mean living up to the promised design, or anywhere near it. I looked at Debian's Hurd distro a year ago and I while it has come a long way, I can't say I was impressed either.
Of course one of the things made Hurd usable was porting Linux device drivers to GNU/Mach. So how come it's not called GNU/Mach/Linux (well, following RMS' example it should be Linux/GNU/Mach). One cannot say drivers are trivial/easy either; Roughly 50% of the 6 million lines in Linux is drivers.
Which is incidentally something RMS has so far been unable to do. Linus plugged the gaping hole in GNU that was/is the quagmire of Hurd.
While I respect the GNU people for GCC and the GPL, I don't consider the rest of the necessary stuff as all that difficult to write: libc and the unix utilities. I call my systems GNU/Linux mainly because of GCC (and in spite of things like "info"). Calling Linux "only" a kernel is a joke; you can't do anything without kernel (or the compiler). So maybe Linus could give more credit to GNU, but then again maybe he would if RMS didn't try to aggrandize himself at every opportunity...
If you really want to build your own, why not start from a team with open-source software or published robot designs?
RoboRoos - Currently the software release seems to be down, but a friendly email to them would probably fix it. CMDragons - My team (whee shameless plug). Our complete 2002 software is available (runs on Debian). RedZone robotics is currently selling a robot based on our design. RoboRoos - These guys have done very well the last couple of years, and have a fair amount of information online as well.
2 of the 3 RoboCup soccer leauges already have red cards, believe it or not. Probably the craziest thing I've ever seen in 6 years of RoboCup was in 1999, in a game between FU-Fighters (Germany) and RobotIS (Korea) in the small size leauge. Small size robots are up to 180mm in diameter, and up to 150mm tall. Germany had robots where the "kicker" was an 800 RPM spinning aluminum metal plate, while robotIS had tiny robots (from another leauge). At one point Germany's robot kicked one of the RobotIS robots completely into the air! It did a complete flip, and when it landed, it just kept on playing, which was almost more amazing. Not suprisingly there were several red cards in that game.
AAAARRRGHHH! As any member of any RoboCup team would tell you, NO! It's not that we don't like BattleBots, but (1) they aren't autonomous, and (2) we get asked this like 50 times a year. Fully autonomous Soccer is a very different reasearch problem from constructing a RC fighting vehicle.
Remember, only the *best* team has to beat humans by 2050. It's a lofty goal, but with something like 150+ teams pre-registered for 2005 (total for 4 leagues), I think we might be able to reach it. I've been doing RoboCup since 1999, and you would not believe the progress just in the last 6 years. I do think its self-centered BS for any particular team to claim *they* will reach the 2050 goal, but I don't think its so bad to say *some* team will.
Humanoids need to advance in hardware, but look at the other robot and simulation leagues and you will see that strategy is coming along quite well already. When I started working on small size, humans with joysticks could beat a RoboCup team. Now humans get demolished. Speed is exactly the computer's advantage -- they are dumb in high-level strategy, but they outdo you in driving and kicking accuracy. Plus they never blink and have a reaction time of 100ms for almost any sort of change. Humans have *at best* 100ms reaction times, and much slower if they get into an unexpected situation. Check out our videos
Personally we've never really found a use for Realtime extensions; Robots are a pretty soft-realtime application becuase the complex algorithms need to be interruptable for debugging. It's way too easy for vision or path planning to take more than the alloted time.
Plain Linux does pretty well at handling the event loop; Most teams are simply event driven off of vision frames, with the motion control loop handled by custom embedded hardware on the robot.
Laugh all you want, but Windows-based teams dominate 2 out of the 3 RoboCup soccer leagues. For small size, the top three teams are Windows-based:
1 - Fu-Fighters (Germany)
2 - RoboRoos (Australia)
3 - LuckyStar (Singapore)
4 - CMRoboDragons (USA & Japan) - go Linux! In the dogs (Aibos), the Aibo itself is running Sony's Aperios operating system, but then there's the platform on which all the development is based:
1 - German Team (Germany, duh)
2 - UTS Unleashed (Australia)
3 - NuBots (Australia)
4 - UPennalizers (USA) - mixed Win/Linux IIRC The remainder is the midsize, in which Linux does very well. That's because large embedded systems (i.e. 40cm cubic robots) are very well suited for Linux and not so much for Windows (For example, being able to ssh into the robot and run code is nice).
The CMU teams of which I am a part (CMRoboDragons, CMPack) are 100% Linux and proud of it, but Windows does much better than you'd think. Cornell's small size team even gets money from Microsoft (they're a 4-time champion).
Don't forget: Extra gyros which start inactivated and can be used later.
It is too bad they won't bring Hubble down in a shuttle like they originally planned, as it definitely belonds in the Air & Space Museum. Now they are worried about safety - maybe they should just ask for a volunteer shuttle crew. I bet enough astronauts would be willing to go to save a part of history.
Everything you say about blogs applies to news organizations as well. So I guess they have no need for them to be transparent either? I don't think its unreasonable to ask that blogs/organizations to be forthcoming about their funding, just like when CNN says some entity is owned by the same company that owns them. While having a single biased source is not so bad, other people will start quoting it, assuming its "the truth". Pretty soon they all quote each other, and since they agree, they must be right. It usually starts with a crackpot writing an unsubstantiated book, but now blogs make this process far more efficient.
As a robotics researcher, I've thought about this a lot, so here's my take:
The biggest problem is that Asimov's 3 Laws require complete information, which is not possible. Opening a door could hurt a human on the other side, therefore no robot can open a door. If you walk backwards or in the dark, you might step on a baby, so you can't do that either. And so it goes, making any action not possible. At the sime time, a robot is supposed to act to save people from harm. In addition, this all assumes the robot can process its sensors well enough to recognize people in all situations, and all potential sources of harm.
The best I've come up with is the following: A robot must take the action (possibly null) that results in the minimum expected harm to humans given currently available or easily obtainable information.
That's a lot weaker, and results potential accidental deaths due to ignorance (just like human actions). But everything in life involves risk, so that's pretty much inescapable.
And I'll oblige. Libertarians believe in state and local rights, far more so than Republicans and Democrats do nowadays. While I might not vote for the fiber system if I lived there, a Libertarian state or federal govt would never tell a city it can't do something like that, nor would they allow a corporation to sponsor a bill to that effect. A true Libertarian citizen would not tell someone in another town what they should and should not do with their local government, beyond debating the merits of various approaches.
A Libertarian government strives to regulate to the minimum extent possible, while still maintaining a working government. For example, such a government would work just fine in the presence of a municipality that voted itself Socialist. Republicans, and to a lesser extent Democrats, would have a fit if this happened.
The two parties have brainwashed the masses into believing that Libertarians only believe in letting corporations go wild. While we support the elimination of most checks on corporate actions, the parties want you to forget that we also are against (1) regulations that help companies, and (2) regulations on state and local governments. Tell me, how much has the government helped protect you against corporations, and how much has it helped them abuse you by passing laws that help them to hurt you (DMCA, Patriot, municipal network bans). Who are *they* working for?
A Libertarian city might be ugly, with every corporation being allowed to lay the wires they could rent on private land, but I bet every home would have more than one strand of fiber to it by now. I feel a municipal network is more a reaction to overly restrictive state and Federal regulations, with a city-sponsored network the only way to deal with that on a local-only basis.
Well, that's a misleading definition of normal in my opinion. I'd say earth's rotation would return to "normal" if the earthquake had no lasting effect. That idea may have not been intended, but was implied by the phrasing of your post. However this is not the case, as earth's rotation as a function of time has be shifted permanently (by ~100 days as you say). From now on the earth will be 100 days behind in its continual slowing, how is that "normal"?
Slashdot Mirror
...it will be a huge test for the OSS community, to see if they can really handle these problems as well as they always say they can.
I guess you've never heard of Apache. 68.8% market share ought to be a good test of the OSS development model. Of course that's not to say that all OSS will fare as well as the carefully developed Apache project, but saying the whole model is untested... is simply uninformed.
Ah, but how do we know we can trust your link? Maybe you're in league with the subversives.
I believe you're thinking of Austin, not Houston.
Any JNI application is essentially running in "unsafe mode" within the VM. The VM has no mechanism to protect itself from a JNI application it is running. I don't find that to be much different from running "unsafe" applications.
With JNI, due to the complexity of the glue pushed on the native application, it's quite easy to mess up the VM. Java is not a bad language, but I find JNI to be one of the most unweildy native interfaces I've seen, even compared with those for functional languages.
No, he's referring to this story: GNU/Hurd Web Server Online
I remember that slashdotting the first time around... pretty funny to link a beta OS up like that to be taken down in minutes by the hoard.
Err, you're the one who didn't get it; It's called "humor" on this planet. You won't think its funny though, rather like any non-liberal reading Doonesbury.
"Monkey" isn't an accurate description in either case though; Bush looks more like a chimpanzee, while Kerry looks like an ape or wookie.
Well I won't be reading CDDL code for exactly that reason. It'll just support a claim of "Willfull infringement" on anything you write later that accidentally steps on a patent. I enjoyed using Solaris in the past, but userspace is where I spend my time, and that's where Linux has really taken off compared to Unix.
Well, assuming they have more than 1600 patents, they could just claim the patents they are suing you for aren't in that list. In other words, without an explicit list of patents, this press release won't help anyone sued by Sun.
The Hurd exists and is usable today.
Well, by ready I would mean living up to the promised design, or anywhere near it. I looked at Debian's Hurd distro a year ago and I while it has come a long way, I can't say I was impressed either.
Of course one of the things made Hurd usable was porting Linux device drivers to GNU/Mach. So how come it's not called GNU/Mach/Linux (well, following RMS' example it should be Linux/GNU/Mach). One cannot say drivers are trivial/easy either; Roughly 50% of the 6 million lines in Linux is drivers.
Oops! Someone pointed out to me that the first link above should read Roobots.
Linus only coded a kernel
Which is incidentally something RMS has so far been unable to do. Linus plugged the gaping hole in GNU that was/is the quagmire of Hurd.
While I respect the GNU people for GCC and the GPL, I don't consider the rest of the necessary stuff as all that difficult to write: libc and the unix utilities. I call my systems GNU/Linux mainly because of GCC (and in spite of things like "info"). Calling Linux "only" a kernel is a joke; you can't do anything without kernel (or the compiler). So maybe Linus could give more credit to GNU, but then again maybe he would if RMS didn't try to aggrandize himself at every opportunity...
If you really want to build your own, why not start from a team with open-source software or published robot designs?
RoboRoos - Currently the software release seems to be down, but a friendly email to them would probably fix it.
CMDragons - My team (whee shameless plug). Our complete 2002 software is available (runs on Debian). RedZone robotics is currently selling a robot based on our design.
RoboRoos - These guys have done very well the last couple of years, and have a fair amount of information online as well.
2 of the 3 RoboCup soccer leauges already have red cards, believe it or not. Probably the craziest thing I've ever seen in 6 years of RoboCup was in 1999, in a game between FU-Fighters (Germany) and RobotIS (Korea) in the small size leauge. Small size robots are up to 180mm in diameter, and up to 150mm tall. Germany had robots where the "kicker" was an 800 RPM spinning aluminum metal plate, while robotIS had tiny robots (from another leauge). At one point Germany's robot kicked one of the RobotIS robots completely into the air! It did a complete flip, and when it landed, it just kept on playing, which was almost more amazing. Not suprisingly there were several red cards in that game.
Most RoboCup mid-size robots run Linux. This team isn't at all unusual. In other leagues it's a different story, see my earlier post on the subject.
AAAARRRGHHH!
As any member of any RoboCup team would tell you, NO! It's not that we don't like BattleBots, but (1) they aren't autonomous, and (2) we get asked this like 50 times a year. Fully autonomous Soccer is a very different reasearch problem from constructing a RC fighting vehicle.
(sorry, I'm a bit touchy on this subject)
Remember, only the *best* team has to beat humans by 2050. It's a lofty goal, but with something like 150+ teams pre-registered for 2005 (total for 4 leagues), I think we might be able to reach it. I've been doing RoboCup since 1999, and you would not believe the progress just in the last 6 years. I do think its self-centered BS for any particular team to claim *they* will reach the 2050 goal, but I don't think its so bad to say *some* team will.
Humanoids need to advance in hardware, but look at the other robot and simulation leagues and you will see that strategy is coming along quite well already. When I started working on small size, humans with joysticks could beat a RoboCup team. Now humans get demolished. Speed is exactly the computer's advantage -- they are dumb in high-level strategy, but they outdo you in driving and kicking accuracy. Plus they never blink and have a reaction time of 100ms for almost any sort of change. Humans have *at best* 100ms reaction times, and much slower if they get into an unexpected situation. Check out our videos
That's why my team uses Debian :P
Personally we've never really found a use for Realtime extensions; Robots are a pretty soft-realtime application becuase the complex algorithms need to be interruptable for debugging. It's way too easy for vision or path planning to take more than the alloted time.
Plain Linux does pretty well at handling the event loop; Most teams are simply event driven off of vision frames, with the motion control loop handled by custom embedded hardware on the robot.
Laugh all you want, but Windows-based teams dominate 2 out of the 3 RoboCup soccer leagues. For small size, the top three teams are Windows-based:
1 - Fu-Fighters (Germany)
2 - RoboRoos (Australia)
3 - LuckyStar (Singapore)
4 - CMRoboDragons (USA & Japan) - go Linux!
In the dogs (Aibos), the Aibo itself is running Sony's Aperios operating system, but then there's the platform on which all the development is based:
1 - German Team (Germany, duh)
2 - UTS Unleashed (Australia)
3 - NuBots (Australia)
4 - UPennalizers (USA) - mixed Win/Linux IIRC
The remainder is the midsize, in which Linux does very well. That's because large embedded systems (i.e. 40cm cubic robots) are very well suited for Linux and not so much for Windows (For example, being able to ssh into the robot and run code is nice).
The CMU teams of which I am a part (CMRoboDragons, CMPack) are 100% Linux and proud of it, but Windows does much better than you'd think. Cornell's small size team even gets money from Microsoft (they're a 4-time champion).
Yeah they better not forget to check the mirror next time. How bad would that be...
Don't forget: Extra gyros which start inactivated and can be used later.
It is too bad they won't bring Hubble down in a shuttle like they originally planned, as it definitely belonds in the Air & Space Museum. Now they are worried about safety - maybe they should just ask for a volunteer shuttle crew. I bet enough astronauts would be willing to go to save a part of history.
Everything you say about blogs applies to news organizations as well. So I guess they have no need for them to be transparent either? I don't think its unreasonable to ask that blogs/organizations to be forthcoming about their funding, just like when CNN says some entity is owned by the same company that owns them. While having a single biased source is not so bad, other people will start quoting it, assuming its "the truth". Pretty soon they all quote each other, and since they agree, they must be right. It usually starts with a crackpot writing an unsubstantiated book, but now blogs make this process far more efficient.
As a robotics researcher, I've thought about this a lot, so here's my take:
The biggest problem is that Asimov's 3 Laws require complete information, which is not possible. Opening a door could hurt a human on the other side, therefore no robot can open a door. If you walk backwards or in the dark, you might step on a baby, so you can't do that either. And so it goes, making any action not possible. At the sime time, a robot is supposed to act to save people from harm. In addition, this all assumes the robot can process its sensors well enough to recognize people in all situations, and all potential sources of harm.
The best I've come up with is the following: A robot must take the action (possibly null) that results in the minimum expected harm to humans given currently available or easily obtainable information.
That's a lot weaker, and results potential accidental deaths due to ignorance (just like human actions). But everything in life involves risk, so that's pretty much inescapable.
And I'll oblige. Libertarians believe in state and local rights, far more so than Republicans and Democrats do nowadays. While I might not vote for the fiber system if I lived there, a Libertarian state or federal govt would never tell a city it can't do something like that, nor would they allow a corporation to sponsor a bill to that effect. A true Libertarian citizen would not tell someone in another town what they should and should not do with their local government, beyond debating the merits of various approaches.
A Libertarian government strives to regulate to the minimum extent possible, while still maintaining a working government. For example, such a government would work just fine in the presence of a municipality that voted itself Socialist. Republicans, and to a lesser extent Democrats, would have a fit if this happened.
The two parties have brainwashed the masses into believing that Libertarians only believe in letting corporations go wild. While we support the elimination of most checks on corporate actions, the parties want you to forget that we also are against (1) regulations that help companies, and (2) regulations on state and local governments. Tell me, how much has the government helped protect you against corporations, and how much has it helped them abuse you by passing laws that help them to hurt you (DMCA, Patriot, municipal network bans). Who are *they* working for?
A Libertarian city might be ugly, with every corporation being allowed to lay the wires they could rent on private land, but I bet every home would have more than one strand of fiber to it by now. I feel a municipal network is more a reaction to overly restrictive state and Federal regulations, with a city-sponsored network the only way to deal with that on a local-only basis.
P.S. At least I can spell *your* party correctly.
Well, that's a misleading definition of normal in my opinion. I'd say earth's rotation would return to "normal" if the earthquake had no lasting effect. That idea may have not been intended, but was implied by the phrasing of your post. However this is not the case, as earth's rotation as a function of time has be shifted permanently (by ~100 days as you say). From now on the earth will be 100 days behind in its continual slowing, how is that "normal"?
Apparently you haven't heard that a 9.0 earthquake is so powerful it can change the spelling of words...