Using regular industrial-rated parts is standard for CubeSats in LEO, where radiation is far less of a problem than many people seem to think. The orbit is only about 350km altitude, well below the Van Allen belts, and with a mission duration of about 6 months it's not worth using parts that are vastly more expensive. For the small variation in failure probability, it makes far more sense to put up multiple ArduSats using regular parts than to put up a single ArduSat with rad-hardened parts for the same price. The plan is that nanosatisfi will launch up to 150 ArduSats over the next 5 years, and there are bound to be failures along the way but overall the cost:failure ratio works out much better with regular industrial parts.
Common consumer-grade electronics such as cameras and laptop computers has been in use on the ISS for many years. Thermal management and power management are much bigger challenges for CubeSats than the relatively minor radiation they encounter.
RFID has nothing to do with encryption/security. It's a serial number.
...
Now if RFID tags had RSA or something built in, it would be a different story. But they don't.
But some do. If you actually did some research you'd discover there are a wide variety of chipsets used in RFID tags, and some do incorporate encryption as well as being able to "stealth" themselves unless the reader identifies itself correctly. Some also come with R/W memory which could be very useful. The tag I've implanted is read-only but I'm hoping one day to be able to carry my data with me in a tag that works much like a contactless memory stick, and can be mounted on my computer like any other mass storage device.
The fees are just there to ensure that LMI can actually defend the trademark if push comes to shove. What's the point of having it as a registered trademark if the first time it's misused there are no funds to even send a cease-and-desist letter? Linus would have to dig into his own pockets or ask for help. This way organisations which are trading on his work and making money from use of the term "Linux" have to pay a very small fee that's proportional to their own turnover. It seems like a pretty small price to pay to ensure the trademark is maintained!
So you talk about hypothetically releasing a "WatchLinux" - but what if the term "Linux" is totally devalued by inappropriate use and Linus can't defend it? What's the point of using it in the product name then? The whole point of establishing the trademark is to maintain its meaning and value exactly so people like yourself can benefit from that value.
Or another way of thinking about it: imagine you wanted to release "WatchWindows2000". Do you think you'd get away with that as a product name, especially without paying anything for the rights to do so? Then why is that any different to wanting to use Linus' trademark? The fact is that with the trademark established you *will* be able to release a product named "WatchLinux" if you do so according to LMI's (and therefore Linus') licencing conditions, but good luck trying to release "WatchWindows2000" under *any* conditions!
...because I have the trademarked word Linux on my website somewhere I will pull it...
This is ridiculous. You do *not* have to pay money to use the word "Linux" on a website! Do you have to pay to mention the word "Microsoft"? Of course not, but you can bet it's a trademark. And if you wanted to use "Microsoft" in your product or company name you'd be jumped on big time. Sure, you can sell stuff that's branded "Microsoft", as long as it actually comes from Microsoft. And you can sell stuff that's branded "Linux", and as long as it's really Linux. But if for eg you wrote a totally new operating system kernel and tried to release it as "Linux 3.0", that could potentially cause massive confusion. That's the sort of thing the trademark protects against, not someone just using the term "Linux" to refer to the Linux we all know and love.
So will I get sued for using vanilla sources and not removing the Linux mark?
The whole point of this action is just to secure the trademark in Australia so that it can't be exploited or used inappropriately. It's not to stop someone downloading and compiling the Linux kernel!
Hey! What's the idea? linuxchix.com isn't a pr0n site!!!1
It *was* a pr0n site, which is one of the things that sparked the whole move to prevent such misuse of the term. That's the reason this trademark application is in place.
Nope, not quite that exciting: just part of a web server farm full of Linux whiteboxes running Ubuntu and Debian. The only vaguely relevant thing about that pic is that hiding down behind my left leg is a black shape that happens to be one of the Linux Australia servers.
Maybe we can all put Tux icon in place of the word Linux
Ahhhh! This sort of ridiculous statement makes me want to scream! Yes, you can use the Tux icon in fair and reasonable ways because it's protected by copyright and Larry is nice enough to use his rights WRT the copyright to allow people to use it appropriately. But that's *exactly* what the trademark action is also trying to achieve for the term "Linux"! At present it's not a trademark in Australia. This action is to make sure the trademark is controlled by Linus, and do you think he's going to be unreasonable about what people are allowed to do with it? Of course not! On the other hand, what do you think would happen if the trademark was allowed to fall into the hands of a commercial entity with a vested interest? Hmm, scary thought indeed.
Effectively you're saying "I'm not going to use this term because it's going to be under Linus' control through trademark law, so I'll use this image instead because it's under Larry's control under copyright law."
No, not in Australia. That's the whole point. Linux Mark Institute was created to oversee management of the trademark worldwide but the registration has not yet gone through in Australia.
So please people, get a grip: this is not about scamming money out of people by taking control of the word "Linux". It's about making sure Linus himself has control of it.
That's not quite the point. What you're talking about is a usage of the term "Linux" that would be considered acceptable, but this is about securing the trademark so that Linus has the right to take action against uses he considers unacceptable, and to prevent others securing the trademark instead of him. Given his very inclusive, community-oriented nature it would take something pretty serious to make him take action to stop someone using the term, but the point is to secure the trademark so that he *can* take action if he needs to.
As I pointed out in a follow-up story a couple of weeks ago and discussed in the latest Linux Australia podcast, this all came out of a situation where a commercial organisation attempted to register the trademark in Australia. If that had been allowed to occur we could have been in a very difficult position indeed, with the trademark holder potentially either requiring large licence fees from anyone who uses the term in a commercial context or just straight out denying use of it at all. Imagine the mess we'd be in then!
So the action to secure the trademark on behalf of Linus is to ensure that sort of thing can't happen.
Thanks for pointing out the facts of this application! It looks like a lot of people are getting very upset for no particular reason, which is dissappointing.
This trademark application is being done to ensure that Linus has control of use of the "Linux" trademark in Australia. It's *not* a money-grabbing exercise by a lone scam artist. For reference, the relevant players are:
LMI (Linux Mark Institute) which was created as an organisation to administer the "Linux" trademark on behalf of Linus.
Linux Australia, the national representative body for Linux users and developers in Australia which is acting on behalf of LMI to secure the trademark in Australia for Linus.
Jeremy Malcolm, who is Linux Australia's legal counsel and has been working for more than a year to shepherd the application through IP Australia and despite the claims of previous posts is *not* a scientologist loon, but rather the victim of a character assassination attempt by a certain individual who for several years has been mischaracterising him online.
One other thing that needs clearing up is that trademarks are *not* patents! They are totally different things. Many earlier posts have made ridiculous statements about the GPL protecting the name "Linux" and generally confusing the two.
A session on profiling web apps was poor. Yes, I *would* expect a web-mail app to spend a lot of time in regexs, that's how you look for injected content.
You're right, I did a pretty crappy job of doing that presentation. I was stressed, I had lots of trouble getting my laptop to work with the projector, I got the timing wrong, it all just fell in a heap. Oh well, sometimes that happens. At least I learned from my mistakes and revised the presentation: I've since done an extended version of it at 3 other confs and did a dramatically better job of it each time.
Regarding the regex comment though, it seems you missed the point I was trying to make: sure, regex is appropriate in that situation but I didn't really care about the low-level details. My point was that running the profile showed that all the expensive regex checks were being done on *every* *single* *message* in the mailbox, even when there were only a few being displayed. It's an issue of program flow and structure that becomes blindingly obvious when you do profiling, but may not be obvious when you're working in the guts of the code.
FWIW, wearing Speedos was part of the joke in terms of the auction for the rights to throw the balls to dunk him. Rusty ran the auction at the conference dinner the night before, and another little incentive he used was that he (Rusty) would wear a g-string unless the bidding went over a certain figure. Luckily it did;-)
Then, what is probably one of the more complex things for people who would whine about the installer, disk partitioning, is left to the same crappy program. Same with partition mounting and filesystem selection.
Yes, and I made mention of that very thing in the article, but luckily it looks like it'll be taken care of by the proper release: if you boot the installer in a more verbose mode (such as by typing "linux DEBCONF_PRIORITY=medium") there is actually a tool to handle all the partitioning for you automatically. You just don't see it if you do a normal install. The developers intend to have that option more generally accessible in future builds.
The emperor is currently in a boutique clothing store surrounded by so many beautiful clothes he's got no idea which ones to put on first.
Seriously, there have been heaps of papers submitted, and selecting which ones get the nod is a hard task being finalised as I type. I'm just hoping I'm one of them! (I submitted a paper, and I'm expecting to hear in the next few days what the outcome is).
Judging by previous LCAs, the sessions will be great (with the possible exception of mine). Trust me.
As far as installation time goes, well, that varies;-)
One of the earlier articles covers running the basic installer, but you may have trouble getting to it right now since the linmagau server is slashdotted:
As for time to invest in updating security patches etc, that depends which distro (Stable, Testing, Unstable) you go with. For a server, use Stable. Then as long as you have security.debian.org in your sources.list (the default if you use the Woody installer) and do a regular "apt-get update; apt-get upgrade" you'll be set as far as security patches are concerned.
For Testing and Unstable the situation is slightly more complex, but for a server they are irrelevant unless you start doing things like backporting recent packages - not something you are likely to do until you learn more about Debian.
I'm intending to do a future article on the process Debian use for security patches, advisories etc. It's in my TOC on www.debianuniverse.com anyway;-)
Well, it's sad that people say that now that apt has been ported to other distributions, Debian has no advantages anymore
It wasn't at all my intention to give that impression, sorry if I wasn't explicit enough. The point I was making is that there are *lots* more good things about Debian than Apt, just that Apt is the most obvious one (and the one most people are aware of).
Calling APT the main and only advantage of Debian is plain ignorance
Umm, I didn't. In fact I specifically said that Debian is not a one-idea project. It's just that Apt is the most obvious feature to most people who aren't as aware of equally (IMHO) important issues as amazingly broad multi-platform support, strict adherence to an Open Source-only policy, or the outstanding work of the security team to backport fixes. Maybe as maintainer of apt-cacher I focused too much on Apt, but what prompted me to write it in the first place was the requests from Red Hat users for assistance with getting apt-cacher running on their RH boxes.
And I think there is *nothing* wrong with that, I'm really glad to have been able to help them, it's just something that started off the train of thought that ended with that article. Hence the Apt focus.
At Linuxconf2002 Wayne Piekarski showed video of it in action. He also did live demos of the fiduciary marker system used to track objects in view (like gun, etc).
Some of the photos may be mockups, but the system is real.
I was lucky enough to go to Linuxconf2002 in Brisbane a few months ago where Wayne Piekarski (the principle developer of the augmented reality system) did a presentation, showed movies of quake sessions and did live demos of the fiduciary markers they use for augmented reality. Absolutely fascinating.
Anyway, the point is that he said they had to make major changes to the bots to make them much much slower, because a person limited by real world physics couldn't even come close to matching their speed. They also found there were funky moves you get used to doing in Quake (like spinning 180 deg in a split second, or strafing, or jumping and spinning) that are almost impossible IRL.
And players could only do it for a short time before stopping from exhaustion;-)
Slashdot Mirror
Using regular industrial-rated parts is standard for CubeSats in LEO, where radiation is far less of a problem than many people seem to think. The orbit is only about 350km altitude, well below the Van Allen belts, and with a mission duration of about 6 months it's not worth using parts that are vastly more expensive. For the small variation in failure probability, it makes far more sense to put up multiple ArduSats using regular parts than to put up a single ArduSat with rad-hardened parts for the same price. The plan is that nanosatisfi will launch up to 150 ArduSats over the next 5 years, and there are bound to be failures along the way but overall the cost:failure ratio works out much better with regular industrial parts. Common consumer-grade electronics such as cameras and laptop computers has been in use on the ISS for many years. Thermal management and power management are much bigger challenges for CubeSats than the relatively minor radiation they encounter.
RFID has nothing to do with encryption/security. It's a serial number.
...
Now if RFID tags had RSA or something built in, it would be a different story. But they don't.
But some do. If you actually did some research you'd discover there are a wide variety of chipsets used in RFID tags, and some do incorporate encryption as well as being able to "stealth" themselves unless the reader identifies itself correctly. Some also come with R/W memory which could be very useful. The tag I've implanted is read-only but I'm hoping one day to be able to carry my data with me in a tag that works much like a contactless memory stick, and can be mounted on my computer like any other mass storage device.
The fees are just there to ensure that LMI can actually defend the trademark if push comes to shove. What's the point of having it as a registered trademark if the first time it's misused there are no funds to even send a cease-and-desist letter? Linus would have to dig into his own pockets or ask for help. This way organisations which are trading on his work and making money from use of the term "Linux" have to pay a very small fee that's proportional to their own turnover. It seems like a pretty small price to pay to ensure the trademark is maintained!
So you talk about hypothetically releasing a "WatchLinux" - but what if the term "Linux" is totally devalued by inappropriate use and Linus can't defend it? What's the point of using it in the product name then? The whole point of establishing the trademark is to maintain its meaning and value exactly so people like yourself can benefit from that value.
Or another way of thinking about it: imagine you wanted to release "WatchWindows2000". Do you think you'd get away with that as a product name, especially without paying anything for the rights to do so? Then why is that any different to wanting to use Linus' trademark? The fact is that with the trademark established you *will* be able to release a product named "WatchLinux" if you do so according to LMI's (and therefore Linus') licencing conditions, but good luck trying to release "WatchWindows2000" under *any* conditions!
...it sounds like they have already target one guy...
No, that was just someone who received the same letter as everyone else but misinterpreted it as a demand for fees. It wasn't.
...because I have the trademarked word Linux on my website somewhere I will pull it...
This is ridiculous. You do *not* have to pay money to use the word "Linux" on a website! Do you have to pay to mention the word "Microsoft"? Of course not, but you can bet it's a trademark. And if you wanted to use "Microsoft" in your product or company name you'd be jumped on big time. Sure, you can sell stuff that's branded "Microsoft", as long as it actually comes from Microsoft. And you can sell stuff that's branded "Linux", and as long as it's really Linux. But if for eg you wrote a totally new operating system kernel and tried to release it as "Linux 3.0", that could potentially cause massive confusion. That's the sort of thing the trademark protects against, not someone just using the term "Linux" to refer to the Linux we all know and love.
So will I get sued for using vanilla sources and not removing the Linux mark?
The whole point of this action is just to secure the trademark in Australia so that it can't be exploited or used inappropriately. It's not to stop someone downloading and compiling the Linux kernel!
More information is available at the LMI site: http://www.linuxmark.org/
Hey! What's the idea? linuxchix.com isn't a pr0n site!!!1
It *was* a pr0n site, which is one of the things that sparked the whole move to prevent such misuse of the term. That's the reason this trademark application is in place.
Nope, not quite that exciting: just part of a web server farm full of Linux whiteboxes running Ubuntu and Debian. The only vaguely relevant thing about that pic is that hiding down behind my left leg is a black shape that happens to be one of the Linux Australia servers.
Maybe we can all put Tux icon in place of the word Linux
Ahhhh! This sort of ridiculous statement makes me want to scream! Yes, you can use the Tux icon in fair and reasonable ways because it's protected by copyright and Larry is nice enough to use his rights WRT the copyright to allow people to use it appropriately. But that's *exactly* what the trademark action is also trying to achieve for the term "Linux"! At present it's not a trademark in Australia. This action is to make sure the trademark is controlled by Linus, and do you think he's going to be unreasonable about what people are allowed to do with it? Of course not! On the other hand, what do you think would happen if the trademark was allowed to fall into the hands of a commercial entity with a vested interest? Hmm, scary thought indeed.
Effectively you're saying "I'm not going to use this term because it's going to be under Linus' control through trademark law, so I'll use this image instead because it's under Larry's control under copyright law."
No, not in Australia. That's the whole point. Linux Mark Institute was created to oversee management of the trademark worldwide but the registration has not yet gone through in Australia.
So please people, get a grip: this is not about scamming money out of people by taking control of the word "Linux". It's about making sure Linus himself has control of it.
That's not quite the point. What you're talking about is a usage of the term "Linux" that would be considered acceptable, but this is about securing the trademark so that Linus has the right to take action against uses he considers unacceptable, and to prevent others securing the trademark instead of him. Given his very inclusive, community-oriented nature it would take something pretty serious to make him take action to stop someone using the term, but the point is to secure the trademark so that he *can* take action if he needs to.
As I pointed out in a follow-up story a couple of weeks ago and discussed in the latest Linux Australia podcast, this all came out of a situation where a commercial organisation attempted to register the trademark in Australia. If that had been allowed to occur we could have been in a very difficult position indeed, with the trademark holder potentially either requiring large licence fees from anyone who uses the term in a commercial context or just straight out denying use of it at all. Imagine the mess we'd be in then!
So the action to secure the trademark on behalf of Linus is to ensure that sort of thing can't happen.
Thanks for pointing out the facts of this application! It looks like a lot of people are getting very upset for no particular reason, which is dissappointing.
This trademark application is being done to ensure that Linus has control of use of the "Linux" trademark in Australia. It's *not* a money-grabbing exercise by a lone scam artist. For reference, the relevant players are:
LMI (Linux Mark Institute) which was created as an organisation to administer the "Linux" trademark on behalf of Linus.
Linux Australia, the national representative body for Linux users and developers in Australia which is acting on behalf of LMI to secure the trademark in Australia for Linus.
Jeremy Malcolm, who is Linux Australia's legal counsel and has been working for more than a year to shepherd the application through IP Australia and despite the claims of previous posts is *not* a scientologist loon, but rather the victim of a character assassination attempt by a certain individual who for several years has been mischaracterising him online.
One other thing that needs clearing up is that trademarks are *not* patents! They are totally different things. Many earlier posts have made ridiculous statements about the GPL protecting the name "Linux" and generally confusing the two.
A session on profiling web apps was poor. Yes, I *would* expect a web-mail app to spend a lot of time in regexs, that's how you look for injected content.
You're right, I did a pretty crappy job of doing that presentation. I was stressed, I had lots of trouble getting my laptop to work with the projector, I got the timing wrong, it all just fell in a heap. Oh well, sometimes that happens. At least I learned from my mistakes and revised the presentation: I've since done an extended version of it at 3 other confs and did a dramatically better job of it each time.
Regarding the regex comment though, it seems you missed the point I was trying to make: sure, regex is appropriate in that situation but I didn't really care about the low-level details. My point was that running the profile showed that all the expensive regex checks were being done on *every* *single* *message* in the mailbox, even when there were only a few being displayed. It's an issue of program flow and structure that becomes blindingly obvious when you do profiling, but may not be obvious when you're working in the guts of the code.
FWIW, wearing Speedos was part of the joke in terms of the auction for the rights to throw the balls to dunk him. Rusty ran the auction at the conference dinner the night before, and another little incentive he used was that he (Rusty) would wear a g-string unless the bidding went over a certain figure. Luckily it did ;-)
Yes, and I made mention of that very thing in the article, but luckily it looks like it'll be taken care of by the proper release: if you boot the installer in a more verbose mode (such as by typing "linux DEBCONF_PRIORITY=medium") there is actually a tool to handle all the partitioning for you automatically. You just don't see it if you do a normal install. The developers intend to have that option more generally accessible in future builds.
The emperor is currently in a boutique clothing store surrounded by so many beautiful clothes he's got no idea which ones to put on first.
Seriously, there have been heaps of papers submitted, and selecting which ones get the nod is a hard task being finalised as I type. I'm just hoping I'm one of them! (I submitted a paper, and I'm expecting to hear in the next few days what the outcome is).
Judging by previous LCAs, the sessions will be great (with the possible exception of mine). Trust me.
Oops, Murdock. ;-)
...and himself (his name is Ian Murdoch) - DebIan.
As far as installation time goes, well, that varies ;-)
m e=Sections&file=index&req=viewarticle&artid=46&pag e=1
;-)
One of the earlier articles covers running the basic installer, but you may have trouble getting to it right now since the linmagau server is slashdotted:
http://www.linmagau.org/modules.php?op=modload&na
As for time to invest in updating security patches etc, that depends which distro (Stable, Testing, Unstable) you go with. For a server, use Stable. Then as long as you have security.debian.org in your sources.list (the default if you use the Woody installer) and do a regular "apt-get update; apt-get upgrade" you'll be set as far as security patches are concerned.
For Testing and Unstable the situation is slightly more complex, but for a server they are irrelevant unless you start doing things like backporting recent packages - not something you are likely to do until you learn more about Debian.
I'm intending to do a future article on the process Debian use for security patches, advisories etc. It's in my TOC on www.debianuniverse.com anyway
It wasn't at all my intention to give that impression, sorry if I wasn't explicit enough. The point I was making is that there are *lots* more good things about Debian than Apt, just that Apt is the most obvious one (and the one most people are aware of).
Umm, I didn't. In fact I specifically said that Debian is not a one-idea project. It's just that Apt is the most obvious feature to most people who aren't as aware of equally (IMHO) important issues as amazingly broad multi-platform support, strict adherence to an Open Source-only policy, or the outstanding work of the security team to backport fixes. Maybe as maintainer of apt-cacher I focused too much on Apt, but what prompted me to write it in the first place was the requests from Red Hat users for assistance with getting apt-cacher running on their RH boxes.
And I think there is *nothing* wrong with that, I'm really glad to have been able to help them, it's just something that started off the train of thought that ended with that article. Hence the Apt focus.
At Linuxconf2002 Wayne Piekarski showed video of it in action. He also did live demos of the fiduciary marker system used to track objects in view (like gun, etc). Some of the photos may be mockups, but the system is real.
I was lucky enough to go to Linuxconf2002 in Brisbane a few months ago where Wayne Piekarski (the principle developer of the augmented reality system) did a presentation, showed movies of quake sessions and did live demos of the fiduciary markers they use for augmented reality. Absolutely fascinating.
;-)
Anyway, the point is that he said they had to make major changes to the bots to make them much much slower, because a person limited by real world physics couldn't even come close to matching their speed. They also found there were funky moves you get used to doing in Quake (like spinning 180 deg in a split second, or strafing, or jumping and spinning) that are almost impossible IRL.
And players could only do it for a short time before stopping from exhaustion