You can also make/boot a symlink to the mini-partition. This is what I do; I like the idea of the kernel image being on a separate partition so it doesn't really matter what I do with my root FS.
I also agree totally, the problem with a lot of competition in commercial products, is its not functionality/ideology based. Rather, all to often it is just down to independant companies working independently for their own profit.
Sure, independence has its good points, more competition leads to better software. But as stated, look at Windows and OSX (although technically not forks, the same applies).
Both are virtually indentical to the end user in terms of features etc. yet both have had mountains of duplicate code doing the same thing in each, just a slightly different way.
Furthermore, you've got a pretty good guarantee that the your "supplier" won't disappear or try to force you on to some other product. Think of all the examples of companies being bought out and their product lines tanked. Entire architectures and OSes have died this way (DEC/Alpha/Unix/VMS, for example). Now count up the significant forks in OSS... in general they don't happen that often. Which is the greater danger?
Alternatively, think how fun it must be to run PeopleSoft nowadays, knowing you're a target for forced product migration after a takeover. However running XFree86 I'm not worried at all about its fork; Only if it turns out to be better will I need to do anything.
Any real examples for *Linux*, or just hot air? EGCS is the best one I can think of... and that led to the *horrible* revival of GCC development *gasp*. Most forks are by a few people, which would have quit the project otherwise, so they are effectively zero drain on resources. Negative competition does happen in open source; but its from people reinventing the wheel, not forking.
there's something to be said for "working out of the box".
There's even more to be said for "reasonably secure out of the box". However you're right about most Unix admins being cluefull enough to know to always fix it. It'd be nice if Debian changed their default though, they have far less of an excuse.
Um, so if they win, that just means your next. What makes you think SCO will stop at Linux? They've already mentioned *BSD. SCO has a case? I'm glad you somehow know, when IBM (and the court) have asked SCO to tell us "what this case is actually about". So perhaps you could share this evidence with us, and "grep -R SMP *.[ch]" doesn't count.
You have to enable the mouse driver under "input core". I enable everything under input core as modules. If you want to use a USB mouse you'll need to enable the HID driver under the USB menu. In the last few kernels you'll need to check the "HID input layer" checkbox, which isn't on by default. Hope this helps.
Well, if by win you mean manages to not lose quite yet, then yes. In an IP lawsuit against IBM, the only winning move is not to play; They will lose, and the only question is how long.
Nice quote, though its wrong. Its pretty clear that is not at all the reason Linus started Linux. I've never heard any of the developers on LKML express an opinion like that; Occasionally there are posts with that sentiment but they usually turn out to be newbie users.
I worked at MS on NT in the past, and now I run Linux. I don't hate MS, I just love operating systems. I knew other people at MS who ran Linux at home *while* they were working at MS. Obviously they don't hate MS.
I agree that there are lots of MS-bashing Linux users (as well as the other way), but I think your causality is completely backwards: People that hate MS use Linux (or Mac). This says nothing of the majority of users on either platform.
So why would someone write a quote like that? Here's my guess:
- Some insecure BSD _users_ hate Linux because it is successful.
- BSD _developers_, just like Linux developers, simply love their project.
Somehow I'd guess that's not running XP. Its easy to have a snappy Win98 machine, but what do you do when MS stops releasing vulnerability patches for it? That's why people have to stick to comparing only currently maintained software. Unless you don't care about being a worm-infected drone, of course, but that will definitely cut into the speed of your GUI.
Several current RoboCup teams use the Linx Technologies HP-2, which offers a simple serial interface, supports 8 channels in the 900-928MHz range (so you can avoid interference), and has a data rate up to 50Kb/sec. Unfortunatly its not a transciever, but it definitely wins on cheap, small and lightweight, and doesn't require a 1m antenna;)
To each their own I guess, working with other people I made them play soccer. Some years only working with one other person. We've had college freshmen implmenting interesting things in one semester in their spare time.
No, programming Aibo isn't for everyone, but I think its one of the best ways to learn/teach programming available now. If I didn't do it as a student, I would probably buy one for doing devlopment on my own. Before the OpenR-SDK was open however, the consumer ones were probably a waste unless you had a lot of expendable cash. But progamming ability means a lot... with open source you can share them with other people too.
Also, if you don't buy the absolute latest Aibo and its $1500 or less. The ERS-210 is $1250 right now, at least in Japan. It is pretty much as capable, though it lacks built in WiFi. Did you buy a 3GHz P4 in your last computer, or get something a little older and save money for other add-ons?
Sony's latest humanoid is called the SDR-4X2. There are some good videos from the ROBODEX2003 here. You can see some longer videos (and japanese interviews) here. The longer ones don't show the SDR moving much, but there's some good footage of Honda's (5ft) Asimo.
The Aibo 210 and 220 have been able to do wireless from the beginning thanks to a PCMCIA/PC-Card slot, and special WiFi cards from Sony that fit in the limited space. The new model just makes the WiFi built in so you don't need to buy the add on card.
A Python Software Development (SDK) is available for the Evolution Robotics ER1 Personal Robot -- why not for the Aibo?
Um, becuase you didn't write one? Aibo has a free C++ (same language level as the internals, nothing between you and the metal) SDK called Open-R. There is also much code available from aibohack, the Tokketsu project (You don't have to RTFA, but at least RTFS (where S = Summary)). There is also complete source code from two RoboCup teams, Carnegie Mellon and German team. Google for them.
P.S. if you overuse the word AI people know you don't know what you're talking about, as in: "Q: What kind of programs do you write? A: I write computer programs!". In other words, its so incredibly general it means nothing to a robotics person. What isn't AI on a robot? Do you mean vision, locomotion, localization, map building, speech recognition, behaviors, or planning?
Got to love Slashdot: $2500 Gaming rig = great, but a $1500 programmable robot = dumb investment. Being unhappy with an Aibo you don't get new programs for (or write them for free) is a bit like being depressed with a game console that you didn't buy any games for.
Or you can run our code and make it walk twice as fast. The person here ported it into a smaller test program and has a nice demo. If you can't program, the Aibo is probably not a wise investment. If you can, think of it as a portable computer with a camera and legs... 1500-1600 USD isn't so bad for that.
Slashdot Mirror
You can also make /boot a symlink to the mini-partition. This is what I do; I like the idea of the kernel image being on a separate partition so it doesn't really matter what I do with my root FS.
Yes, to many here Software is far more important.
I also agree totally, the problem with a lot of competition in commercial products, is its not functionality/ideology based. Rather, all to often it is just down to independant companies working independently for their own profit.
Sure, independence has its good points, more competition leads to better software. But as stated, look at Windows and OSX (although technically not forks, the same applies).
Both are virtually indentical to the end user in terms of features etc. yet both have had mountains of duplicate code doing the same thing in each, just a slightly different way.
Ok, so tell me again why they are different?
Furthermore, you've got a pretty good guarantee that the your "supplier" won't disappear or try to force you on to some other product. Think of all the examples of companies being bought out and their product lines tanked. Entire architectures and OSes have died this way (DEC/Alpha/Unix/VMS, for example). Now count up the significant forks in OSS... in general they don't happen that often. Which is the greater danger?
Alternatively, think how fun it must be to run PeopleSoft nowadays, knowing you're a target for forced product migration after a takeover. However running XFree86 I'm not worried at all about its fork; Only if it turns out to be better will I need to do anything.
Any real examples for *Linux*, or just hot air? EGCS is the best one I can think of... and that led to the *horrible* revival of GCC development *gasp*. Most forks are by a few people, which would have quit the project otherwise, so they are effectively zero drain on resources. Negative competition does happen in open source; but its from people reinventing the wheel, not forking.
there's something to be said for "working out of the box".
There's even more to be said for "reasonably secure out of the box". However you're right about most Unix admins being cluefull enough to know to always fix it. It'd be nice if Debian changed their default though, they have far less of an excuse.
You mean "go 8086", since real mode is still there ;)
...
5. Profit!
(in the copyright law financial gain sense of course, where getting information back from others is seen as a profit).
All I know is, I'm glad I switched to OS X.
Um, so if they win, that just means your next. What makes you think SCO will stop at Linux? They've already mentioned *BSD. SCO has a case? I'm glad you somehow know, when IBM (and the court) have asked SCO to tell us "what this case is actually about". So perhaps you could share this evidence with us, and "grep -R SMP *.[ch]" doesn't count.
You have to enable the mouse driver under "input core". I enable everything under input core as modules. If you want to use a USB mouse you'll need to enable the HID driver under the USB menu. In the last few kernels you'll need to check the "HID input layer" checkbox, which isn't on by default. Hope this helps.
Maybe he meant the 1670's?
Well, if by win you mean manages to not lose quite yet, then yes. In an IP lawsuit against IBM, the only winning move is not to play; They will lose, and the only question is how long.
Unfortunately a verbal pun combined with a Monty Python reference is far beyond the average moderator. +1 funny if I had it.
Nice quote, though its wrong. Its pretty clear that is not at all the reason Linus started Linux. I've never heard any of the developers on LKML express an opinion like that; Occasionally there are posts with that sentiment but they usually turn out to be newbie users.
I worked at MS on NT in the past, and now I run Linux. I don't hate MS, I just love operating systems. I knew other people at MS who ran Linux at home *while* they were working at MS. Obviously they don't hate MS.
I agree that there are lots of MS-bashing Linux users (as well as the other way), but I think your causality is completely backwards: People that hate MS use Linux (or Mac). This says nothing of the majority of users on either platform.
So why would someone write a quote like that? Here's my guess:
- Some insecure BSD _users_ hate Linux because it is successful.
- BSD _developers_, just like Linux developers, simply love their project.
Hey that's not fair; you're supposed to horribly misconfigure XFree before you do the comparison!
Somehow I'd guess that's not running XP. Its easy to have a snappy Win98 machine, but what do you do when MS stops releasing vulnerability patches for it? That's why people have to stick to comparing only currently maintained software. Unless you don't care about being a worm-infected drone, of course, but that will definitely cut into the speed of your GUI.
Several current RoboCup teams use the Linx Technologies HP-2, which offers a simple serial interface, supports 8 channels in the 900-928MHz range (so you can avoid interference), and has a data rate up to 50Kb/sec. Unfortunatly its not a transciever, but it definitely wins on cheap, small and lightweight, and doesn't require a 1m antenna ;)
- Jim from CMDragons 2001,02,03
Why has OpenSSH seemingly become the new WuFTPd? I guess on the bright side it *is* getting more secure now.
Sure, but that's not the parent's point. It really takes too much formalism for people to write things baldly.
To each their own I guess, working with other people I made them play soccer. Some years only working with one other person. We've had college freshmen implmenting interesting things in one semester in their spare time.
No, programming Aibo isn't for everyone, but I think its one of the best ways to learn/teach programming available now. If I didn't do it as a student, I would probably buy one for doing devlopment on my own. Before the OpenR-SDK was open however, the consumer ones were probably a waste unless you had a lot of expendable cash. But progamming ability means a lot... with open source you can share them with other people too.
Also, if you don't buy the absolute latest Aibo and its $1500 or less. The ERS-210 is $1250 right now, at least in Japan. It is pretty much as capable, though it lacks built in WiFi. Did you buy a 3GHz P4 in your last computer, or get something a little older and save money for other add-ons?
Sony's latest humanoid is called the SDR-4X2. There are some good videos from the ROBODEX2003 here. You can see some longer videos (and japanese interviews) here. The longer ones don't show the SDR moving much, but there's some good footage of Honda's (5ft) Asimo.
The Aibo 210 and 220 have been able to do wireless from the beginning thanks to a PCMCIA/PC-Card slot, and special WiFi cards from Sony that fit in the limited space. The new model just makes the WiFi built in so you don't need to buy the add on card.
A Python Software Development (SDK) is available for the Evolution Robotics ER1 Personal Robot -- why not for the Aibo?
Um, becuase you didn't write one? Aibo has a free C++ (same language level as the internals, nothing between you and the metal) SDK called Open-R. There is also much code available from aibohack, the Tokketsu project (You don't have to RTFA, but at least RTFS (where S = Summary)). There is also complete source code from two RoboCup teams, Carnegie Mellon and German team. Google for them.
P.S. if you overuse the word AI people know you don't know what you're talking about, as in: "Q: What kind of programs do you write? A: I write computer programs!". In other words, its so incredibly general it means nothing to a robotics person. What isn't AI on a robot? Do you mean vision, locomotion, localization, map building, speech recognition, behaviors, or planning?
Got to love Slashdot: $2500 Gaming rig = great, but a $1500 programmable robot = dumb investment. Being unhappy with an Aibo you don't get new programs for (or write them for free) is a bit like being depressed with a game console that you didn't buy any games for.
Or you can run our code and make it walk twice as fast. The person here ported it into a smaller test program and has a nice demo. If you can't program, the Aibo is probably not a wise investment. If you can, think of it as a portable computer with a camera and legs... 1500-1600 USD isn't so bad for that.