At $25.000 per CD, this is really getting out of hand. Even if the judge said he could have been much harsher, this is ridiculous. At $15 per CD, that's like saying the industry lost the sale of 1667 copies for each of the CDs. Part of the amount is punitive, I know, but still...
Software piracy cases are judged equally stupid. A couple of local software pirates were convicted of massive piracy (they readily confessed to it), and the software companies simply multiplied the CD count by the suggested retail price to come up with an enormous $400.000.000 claim against two teenagers. Do they really think that each and every pirated CD equals one sale less? Of course it doesn't! From my experience, a good guess would be that maybe 1 in 50 pirates would seriously consider buying the expensive stuff they pirated.
This is not a defense for piracy - not at all. It's merely a call for a "reality check" on the part of the music and software industries to take a realistic approach on these things. It's OK to defend your copyright, but do so in a reasonable fashion, and come up with damage claims that reflect how the actual world works. And I don't need to make comments about clueless judges - their actions speaks volumes on their own.
This is indeed a good idea, and in my native country it happens all the time. Unfortunately, it is severely hindered by the fact that there are essentially two types of plastic. One is the hard breakable kind (like a monitor-case) and the other is the soft bendable kind (like plastic-bags). To most people it's just plastic, and it gets thrown in one big pile. The bad news is that by mixing the two kinds you basically get bendable breakable plastics - bends as easily as a plastic bag, but breaks the minute it bends.
The solution would be to have it sorted, but since this is almost impossible to do automatically, it would require that you sort the plastics yourself. Since this requires a little thinking and effort, Joe Random Sixpack is certainly not going to bother with it.
I've been doing some work with GAs myself, and the tank recognition problem you mention is NOT a problem inherent to GAs - the training set was chosen poorly. They ARE prone to over-adapting to what are, in hindsight, irrelevant details. But most of this can be overcome by a careful selection of the training set. This really is very hard - it needs to be progressively more difficult, but at a slow, steady pace. Want to teach it to recognize tanks? Maybe start with silhouttes of tanks on white paper, move on to tanks with a few distractions thrown in, then on to tanks in simple scenarios (on a road, in a ditch etc.), then on to <a series of harder stuff>.
Teaching GAs is very similar to teaching young children to read - you start out with REALLY simple stuff like a few select letters (say, the ones in the child's name), train that until they get the idea, move on to more letters and perhaps assign a few select words to each letter (like "A as in Apple", "B as in Banana"), then on to REALLY simple two-letter words etc.
Toss a couple of Harry Potter books their way instead and train exclusively and intensively on those, and you get children that can, by and large, recite the entire text but can not read simple newspaper texts. Not because your children are stupid, not because Harry Potter books are bad, not because you didn't try really hard - it just wasn't the right thing to do (read: wrong training set).
IMHO the reason why GAs typically have the problems you mention (and they are very real problems) is that the average GA designer/trainer is a traditional programmer. That's not a bad thing to be, but it doesn't turn you into a good teacher. Choosing a training set is SOOOO difficult, and might I suggest that GA-training require a different mindset than ordinary programming. Maybe we should try having professional teachers have a go at designing training set...
Good points - I think we agree here. I just want to point out that there can easily be a unique relationship between God and mankind, alongside some other planet teeming with a bacteria-like lifeform. Or a dinosaur-like lifeform. Or a [insert favorite lifeform]. Just because He's got a special relationship with mankind, doesn't necessarily imply that we're the only intelligent ones around - He might just like us better (nice thought - always good to have influential friends:-).
You're absolutely right - it does not follow that He must do so. But it does follow that He could do so - we just don't know.
But either way, there's no indication whatsoever that He granted Earth exclusive rights to life, which was just the point I wished to make. You may (or may not) argue that man is the most intelligent life around (the pivotal being or something), but that's beside the point - you don't need that much intelligence to be alive. Bacteria, fungi and plants are very much alive, but possess no intelligence to speak of. Those lifeforms would not show up on SETI - we really need to travel to other planets in order to find out.
There's a piece about kids programming Lego Mindstorms at last years RoboCup in Sweden, where they found that 7-14 year old non-nerd non-geek children were able to program them to score a goal in ~30 minutes or so. Here's the link. If only they had that kind of fun when I was that age...
I'll bite the bait - if your God is omnipotent, why do you place boundaries on his work by claiming that we, the lifeforms on Earth, are the only ones he made? It would occur to me that a truly omnipotent God would have created beings in his image lots of places. Everywhere would not be a problem, since he's also omnipresent. He just hasn't seen fit to tell us about it, or something, I don't claim to know how your God works. I see NO reason why believing in the Bible would exclude believing in extraterrestrial life - I rather think it would include extraterrestrial life because of the omnipotency. What happened to being humble before God? Your cocky claim that you know EXACTLY where his work ends is in glaring contrast to the humbleness towards an omnipresent/omnipotent God that the Bible actually teaches us.
Also, suppose you spot a misspelled word (or worse, something like a person's name) that you want fixed. How long does it take to move the cursor there by means of a keyboard and/or mouse and fix it? How long does it take for you to say (and the machine to understand and execute): "third paragraph, eleventh line, replace McIntire with McIntyre"? Hmmm, seems like you will even have to count and specify the individual characters to get it to do that right
True - if no other advance in HCI (Human Computer Interaction) takes place simultaneously. I've seen systems where people who'd lost the use of both arms were able to point on the screen by just looking for a couple of seconds. Imagine seeing your spoken words roll by as you speak, and when you spot an error you just look real hard at it and say "replace with <real word>". I'm assuming a lot of tech here, but the topic was The Computer of 2010 so you'll ahve to live with a little dreaming (and ranting) on my part:-)
I agree - the limit is probably always going to be lower. But comparing elite keyboard typers to dictation rookies is not a fair comparison. Unless the goal is to have zero learning curve - impossible, if you ask me. How is the machine suppose to tell if I'm using sarcastic, ironic tech-speak to talk with my co-workers or I'm using slowly spoken carefully chosen words to communicate with my slightly senile grandmother? The two scenarios are so far apart, that no single system is likely to be able to handle both situations without numerous errors.
Don't get me wrong - I'd be *really* impressed by a zero learning curve speech recognition system, and would probably use it right away. I just think it's an impossible task, due to the context problem exemplified in the above paragraph. If some elite hacker is about to prove me wrong, I'll stand happily corrected:-)
Do you actually use a Palm regularly? I do, and some of my co-workers are true artists in terms of grafitti style writing. But they still write VERY slowly compared to their keyboard typing speed. It's just a slow way to enter data...
Don't get me wrong - I really like using my Palm. But not for hardcore typing. Good for taking notes during meetings. Going over my calendar in the commute train. Catching up with the news that got hotsync'ed while I was busy typing to earn a living.
Like you, I can type at ~100WPM. But it took time to learn how to do that. Years. Wonder how fast I could be dictating words if I had trained that as hard as I did typing. I'm also pretty sure that sore vocal chords would be a reminiscent of old days once you mastered it, just as sore hands from typing when I was still learning to do that...
Why does everybody demand that voice recognition require little or no training whatsoever? I've been able to move my fingers for years before I could type, and I'll happily accept that it'll be a considerable time before I can dictate with any form of precision and speed. Once we can do that (assuming that the current crop of speech recognition software catches up), *THEN* let's discuss if typing or dictating is faster than the other.
Sure they review playability, configurability, stability, and a few other standard review items -- all important, no doubt. Yet they consistently overlook what is going to keep you coming back for more. I think it's time the developers took an objective look at what made those games of yesteryear great, and take a step back from replying on the technological developments of tomorrow for sales.
A thought just occurred to me - maybe they have no replay value because they don't want you to be too happy with any single game. The reason: Then you won't buy as many games! Think about it - if game X can keep you happily gaming for a year, you are less inclined to buy another title. Fewer bought titles at $40-$50 a pop - less money for the publishers. You might argue that having a thoroughly good time with, say, Civilization for years (as I did) will make me buy Civ II. It didn't - it had changed too much. To this date, I have never played Civ II for more than about 10 minutes at a time. I have not even considered Civ:CTP - lots of fun playing Alpha Centauri though. I've bought numerous space sims because I wanted better ones - Wing Commander and X-Wing were kinda cool, but I kept on buying new ones to see if they had "the right stuff". They didn't. Although drawing conclusions from just one individual (especially a sick one like me:-) is always a questionable approach, I do think that a lot of people are like me - buying lots of crap hoping to find a nugget of gold. And $40 for a crappy game is exactly as much money as $40 for a marvelous (sp?) one...
If they are porting Office over right now, when something like StarOffice, or if WordPerfect makes a miracle comeback and starts to eat up their market AND they can trace the loss back to Linux, then they release the product and try to kill the competing office product.
I think they make take an even more devious approach. If they wait until the GNOME foundation has spent a lot of its members money, and THEN release Office for Linux when GNOME Office (or whatever it'll be called) is not quite ready yet, they make sure to incur maximum losses on their competitors while still being able to say that they're just reacting to market forces - and that at the crucial time for GNOME Office, just before market entry. It would be completely legal and completely ruthless, so I think Microsoft would love it:-)
developing mantainance code in NT is pain (PErl vs. Win32/MFC.
Have you ever tried WSH (Windows Scripting Host) alongside ADSI 2.5? Everybody thinks its a W2K tech, but it works like a charm on NT, and there's little you can not control through a snippet of JavaScript. It's a free download too:-)
Agree with all the other things you write, though.
OK, I can see how it could be mistaken for a 30 minute boot-time. My mistake. Sorry. It is 30 minutes from the uCsimm comes out of its antistatic bag until it boots Clinux, which takes just 3 seconds (as andersen has been really busy pointing out - thanks man).
As you'll probably see from my name, I submitted the story. It has nothing to do with press releases - I have two of those babies running Clinux in my office (yes, I get paid to work with them:-).
In short, you're just plain wrong. I've been using them and they've been in production for, what, a year? Press releases and early adopters my a**
Maybe within another year, when the new XFree4 DRI supports more video cards, sales will pick up. Right now, you still have to be a bit of a hacker to even get hardware acceleration to work in Linux.
The former part is of course necessary, but in no way enough. The latter part is the essential part. It must be made REALLY easy to install support for whatever kind of hardware is present. If it involves kernel patches, tweaking with X configurations or any such things, no ordinary user will even think about trying. Most slashdotters could do it, I'm sure - but we're not representative of the general game-buyer market.
Something along the lines of putting the video card into the machine, booting Linux and pressing "Yes" in the "install drivers for xxx graphics card". If it needs to download some piece of software to do so, then it should. If it needs to rebuild X, then it should. If it needs to restart the kernel, then it should.
I know that it sounds a whole lot like the process for installation on Win9x
nifty little utility to split/reassemble all those little bits of larger files
Sounds a whole lot like how us "oldies" used to move pr0n (and more serious stuff too - just can't remember what it was) over the Usenet with uuencode. As I said - nothing new - just old stuff another way.
It's 10x16 centimeters, so 4x6 inches is close enough.
Off my head it costs $250 for parts - some (as in all) assembly required. A few days ago I inquired about a price for an assembled system, but I have yet to get an answer. Will probably be a while now that they got/.'ed.
If you're into in-a-box stuff, like at what the CSimm (no affiliations) does. It's smaller (the whole darn thing fits in an old-style 30 pin ram-socket). Just bought two of those for running CLinux. They're pretty little Motorola Dragonballs (also powers the PalmPilot V series) with 8 Megs of RAM and 2 Megs of Flash and an ehternet chip. No sleep for me now:-)
are usually referred to by astronomers and astronauts as "the Earth - Moon system".
That's all good and well, but it sure doesn't scale very well. Jupiter has 16 moons or something, and it would be pretty darn tedious to recite the whole list just to say "Jupiter and its moons". "Jupiter system" contains the same info.
If it is to be called the Solar system, then what is the generic term for a star and its satellites?
Star system - stellar system sounds better, but is probably less common. Also, star system is much more easily translated into the myriad of languages we speak here on Earth. I do know that stellar essentially means star, but star is the common term that every language has a direct mapping of. You don't mind that the non-English speaking majority of the world gets a good term too, do you? (hehe)
By extension, I would call the Earth and the Moon a planet(ary) system and name it 'Earth System'. The Moon, by itself, would be a satellite system, or maybe a moon system - I don't know. By calling it a moon system, I'd kind of connect it with a crackpot 'religion', but I guess I could probably live with that.
How much money do they need to buy every engineer a copy of 'Metrics for Dummies' and give them a day or two to read it? *That* would be a worthwhile effort...
I've just purchased two ucSimm modules with a DragonBall and running CLinux. The whole thing, with 8 Megs of RAM, 2 Megs of Flash ROM and an Ehternet chip fits in a single old-style 30 pin RAM socket!
For those who care, a DragonBall is officially a Motorola 68EZ328 processor running at 16 MHz. CLinux is a version of our favorite OS designed for chips with no MMU (Memory Management Unit), and fits nicely into the 2 Megs of flash ROM.
Slashdot Mirror
At $25.000 per CD, this is really getting out of hand. Even if the judge said he could have been much harsher, this is ridiculous. At $15 per CD, that's like saying the industry lost the sale of 1667 copies for each of the CDs. Part of the amount is punitive, I know, but still...
Software piracy cases are judged equally stupid. A couple of local software pirates were convicted of massive piracy (they readily confessed to it), and the software companies simply multiplied the CD count by the suggested retail price to come up with an enormous $400.000.000 claim against two teenagers. Do they really think that each and every pirated CD equals one sale less? Of course it doesn't! From my experience, a good guess would be that maybe 1 in 50 pirates would seriously consider buying the expensive stuff they pirated.
This is not a defense for piracy - not at all. It's merely a call for a "reality check" on the part of the music and software industries to take a realistic approach on these things. It's OK to defend your copyright, but do so in a reasonable fashion, and come up with damage claims that reflect how the actual world works. And I don't need to make comments about clueless judges - their actions speaks volumes on their own.
This is indeed a good idea, and in my native country it happens all the time. Unfortunately, it is severely hindered by the fact that there are essentially two types of plastic. One is the hard breakable kind (like a monitor-case) and the other is the soft bendable kind (like plastic-bags). To most people it's just plastic, and it gets thrown in one big pile. The bad news is that by mixing the two kinds you basically get bendable breakable plastics - bends as easily as a plastic bag, but breaks the minute it bends.
The solution would be to have it sorted, but since this is almost impossible to do automatically, it would require that you sort the plastics yourself. Since this requires a little thinking and effort, Joe Random Sixpack is certainly not going to bother with it.
I've been doing some work with GAs myself, and the tank recognition problem you mention is NOT a problem inherent to GAs - the training set was chosen poorly. They ARE prone to over-adapting to what are, in hindsight, irrelevant details. But most of this can be overcome by a careful selection of the training set. This really is very hard - it needs to be progressively more difficult, but at a slow, steady pace. Want to teach it to recognize tanks? Maybe start with silhouttes of tanks on white paper, move on to tanks with a few distractions thrown in, then on to tanks in simple scenarios (on a road, in a ditch etc.), then on to <a series of harder stuff>.
Teaching GAs is very similar to teaching young children to read - you start out with REALLY simple stuff like a few select letters (say, the ones in the child's name), train that until they get the idea, move on to more letters and perhaps assign a few select words to each letter (like "A as in Apple", "B as in Banana"), then on to REALLY simple two-letter words etc.
Toss a couple of Harry Potter books their way instead and train exclusively and intensively on those, and you get children that can, by and large, recite the entire text but can not read simple newspaper texts. Not because your children are stupid, not because Harry Potter books are bad, not because you didn't try really hard - it just wasn't the right thing to do (read: wrong training set).
IMHO the reason why GAs typically have the problems you mention (and they are very real problems) is that the average GA designer/trainer is a traditional programmer. That's not a bad thing to be, but it doesn't turn you into a good teacher. Choosing a training set is SOOOO difficult, and might I suggest that GA-training require a different mindset than ordinary programming. Maybe we should try having professional teachers have a go at designing training set...
Write-Once Read-Never :-)
Good points - I think we agree here. I just want to point out that there can easily be a unique relationship between God and mankind, alongside some other planet teeming with a bacteria-like lifeform. Or a dinosaur-like lifeform. Or a [insert favorite lifeform]. Just because He's got a special relationship with mankind, doesn't necessarily imply that we're the only intelligent ones around - He might just like us better (nice thought - always good to have influential friends :-).
You're absolutely right - it does not follow that He must do so. But it does follow that He could do so - we just don't know.
But either way, there's no indication whatsoever that He granted Earth exclusive rights to life, which was just the point I wished to make. You may (or may not) argue that man is the most intelligent life around (the pivotal being or something), but that's beside the point - you don't need that much intelligence to be alive. Bacteria, fungi and plants are very much alive, but possess no intelligence to speak of. Those lifeforms would not show up on SETI - we really need to travel to other planets in order to find out.
There's a piece about kids programming Lego Mindstorms at last years RoboCup in Sweden, where they found that 7-14 year old non-nerd non-geek children were able to program them to score a goal in ~30 minutes or so. Here's the link. If only they had that kind of fun when I was that age...
I'll bite the bait - if your God is omnipotent, why do you place boundaries on his work by claiming that we, the lifeforms on Earth, are the only ones he made? It would occur to me that a truly omnipotent God would have created beings in his image lots of places. Everywhere would not be a problem, since he's also omnipresent. He just hasn't seen fit to tell us about it, or something, I don't claim to know how your God works. I see NO reason why believing in the Bible would exclude believing in extraterrestrial life - I rather think it would include extraterrestrial life because of the omnipotency. What happened to being humble before God? Your cocky claim that you know EXACTLY where his work ends is in glaring contrast to the humbleness towards an omnipresent/omnipotent God that the Bible actually teaches us.
Also, suppose you spot a misspelled word (or worse, something like a person's name) that you want fixed. How long does it take to move the cursor there by means of a keyboard and/or mouse and fix it? How long does it take for you to say (and the machine to understand and execute): "third paragraph, eleventh line, replace McIntire with McIntyre"? Hmmm, seems like you will even have to count and specify the individual characters to get it to do that right
True - if no other advance in HCI (Human Computer Interaction) takes place simultaneously. I've seen systems where people who'd lost the use of both arms were able to point on the screen by just looking for a couple of seconds. Imagine seeing your spoken words roll by as you speak, and when you spot an error you just look real hard at it and say "replace with <real word>". I'm assuming a lot of tech here, but the topic was The Computer of 2010 so you'll ahve to live with a little dreaming (and ranting) on my part :-)
I agree - the limit is probably always going to be lower. But comparing elite keyboard typers to dictation rookies is not a fair comparison. Unless the goal is to have zero learning curve - impossible, if you ask me. How is the machine suppose to tell if I'm using sarcastic, ironic tech-speak to talk with my co-workers or I'm using slowly spoken carefully chosen words to communicate with my slightly senile grandmother?
The two scenarios are so far apart, that no single system is likely to be able to handle both situations without numerous errors.
Don't get me wrong - I'd be *really* impressed by a zero learning curve speech recognition system, and would probably use it right away. I just think it's an impossible task, due to the context problem exemplified in the above paragraph. If some elite hacker is about to prove me wrong, I'll stand happily corrected :-)
Do you actually use a Palm regularly? I do, and some of my co-workers are true artists in terms of grafitti style writing. But they still write VERY slowly compared to their keyboard typing speed. It's just a slow way to enter data...
Don't get me wrong - I really like using my Palm. But not for hardcore typing. Good for taking notes during meetings. Going over my calendar in the commute train. Catching up with the news that got hotsync'ed while I was busy typing to earn a living.
Like you, I can type at ~100WPM. But it took time to learn how to do that. Years. Wonder how fast I could be dictating words if I had trained that as hard as I did typing. I'm also pretty sure that sore vocal chords would be a reminiscent of old days once you mastered it, just as sore hands from typing when I was still learning to do that...
Why does everybody demand that voice recognition require little or no training whatsoever? I've been able to move my fingers for years before I could type, and I'll happily accept that it'll be a considerable time before I can dictate with any form of precision and speed. Once we can do that (assuming that the current crop of speech recognition software catches up), *THEN* let's discuss if typing or dictating is faster than the other.
Sure they review playability, configurability, stability, and a few other standard review items -- all important, no doubt. Yet they consistently overlook what is going to keep you coming back for more. I think it's time the developers took an objective look at what made those games of yesteryear great, and take a step back from replying on the technological developments of tomorrow for sales.
A thought just occurred to me - maybe they have no replay value because they don't want you to be too happy with any single game. The reason: Then you won't buy as many games! Think about it - if game X can keep you happily gaming for a year, you are less inclined to buy another title. Fewer bought titles at $40-$50 a pop - less money for the publishers. You might argue that having a thoroughly good time with, say, Civilization for years (as I did) will make me buy Civ II. It didn't - it had changed too much. To this date, I have never played Civ II for more than about 10 minutes at a time. I have not even considered Civ:CTP - lots of fun playing Alpha Centauri though. :-) is always a questionable approach, I do think that a lot of people are like me - buying lots of crap hoping to find a nugget of gold. And $40 for a crappy game is exactly as much money as $40 for a marvelous (sp?) one...
I've bought numerous space sims because I wanted better ones - Wing Commander and X-Wing were kinda cool, but I kept on buying new ones to see if they had "the right stuff". They didn't.
Although drawing conclusions from just one individual (especially a sick one like me
If they are porting Office over right now, when something like StarOffice, or if WordPerfect makes a miracle comeback and starts to eat up their market AND they can trace the loss back to Linux, then they release the product and try to kill the competing office product.
I think they make take an even more devious approach. If they wait until the GNOME foundation has spent a lot of its members money, and THEN release Office for Linux when GNOME Office (or whatever it'll be called) is not quite ready yet, they make sure to incur maximum losses on their competitors while still being able to say that they're just reacting to market forces - and that at the crucial time for GNOME Office, just before market entry. It would be completely legal and completely ruthless, so I think Microsoft would love it :-)
developing mantainance code in NT is pain (PErl vs. Win32/MFC.
Have you ever tried WSH (Windows Scripting Host) alongside ADSI 2.5? Everybody thinks its a W2K tech, but it works like a charm on NT, and there's little you can not control through a snippet of JavaScript. It's a free download too :-)
Agree with all the other things you write, though.
OK, I can see how it could be mistaken for a 30 minute boot-time. My mistake. Sorry. It is 30 minutes from the uCsimm comes out of its antistatic bag until it boots Clinux, which takes just 3 seconds (as andersen has been really busy pointing out - thanks man).
As you'll probably see from my name, I submitted the story. It has nothing to do with press releases - I have two of those babies running Clinux in my office (yes, I get paid to work with them :-).
In short, you're just plain wrong. I've been using them and they've been in production for, what, a year? Press releases and early adopters my a**
How about going in the opposite direction, people that can live off electricity?
I've always considered the idea of putting chlorophyl (sp?) in skin a nice thought. Feeling hungry? Drink a glass of water and sunbathe for an hour :-)
Maybe within another year, when the new XFree4 DRI supports more video cards, sales will pick up. Right now, you still have to be a bit of a hacker to even get hardware acceleration to work in Linux.
The former part is of course necessary, but in no way enough. The latter part is the essential part. It must be made REALLY easy to install support for whatever kind of hardware is present. If it involves kernel patches, tweaking with X configurations or any such things, no ordinary user will even think about trying. Most slashdotters could do it, I'm sure - but we're not representative of the general game-buyer market.
Something along the lines of putting the video card into the machine, booting Linux and pressing "Yes" in the "install drivers for xxx graphics card". If it needs to download some piece of software to do so, then it should. If it needs to rebuild X, then it should. If it needs to restart the kernel, then it should.
I know that it sounds a whole lot like the process for installation on Win9x
nifty little utility to split/reassemble all those little bits of larger files
Sounds a whole lot like how us "oldies" used to move pr0n (and more serious stuff too - just can't remember what it was) over the Usenet with uuencode. As I said - nothing new - just old stuff another way.
If only i spoke danish :) I do...
It's 10x16 centimeters, so 4x6 inches is close enough.
Off my head it costs $250 for parts - some (as in all) assembly required. A few days ago I inquired about a price for an assembled system, but I have yet to get an answer. Will probably be a while now that they got /.'ed.
If you're into in-a-box stuff, like at what the CSimm (no affiliations) does. It's smaller (the whole darn thing fits in an old-style 30 pin ram-socket). Just bought two of those for running CLinux. They're pretty little Motorola Dragonballs (also powers the PalmPilot V series) with 8 Megs of RAM and 2 Megs of Flash and an ehternet chip. No sleep for me now :-)
are usually referred to by astronomers and astronauts as "the Earth - Moon system".
That's all good and well, but it sure doesn't scale very well. Jupiter has 16 moons or something, and it would be pretty darn tedious to recite the whole list just to say "Jupiter and its moons". "Jupiter system" contains the same info.
If it is to be called the Solar system, then what is the generic term for a star and its satellites?
Star system - stellar system sounds better, but is probably less common. Also, star system is much more easily translated into the myriad of languages we speak here on Earth. I do know that stellar essentially means star, but star is the common term that every language has a direct mapping of. You don't mind that the non-English speaking majority of the world gets a good term too, do you? (hehe)
By extension, I would call the Earth and the Moon a planet(ary) system and name it 'Earth System'. The Moon, by itself, would be a satellite system, or maybe a moon system - I don't know. By calling it a moon system, I'd kind of connect it with a crackpot 'religion', but I guess I could probably live with that.
How much money do they need to buy every engineer a copy of 'Metrics for Dummies' and give them a day or two to read it? *That* would be a worthwhile effort...
I've just purchased two ucSimm modules with a DragonBall and running CLinux. The whole thing, with 8 Megs of RAM, 2 Megs of Flash ROM and an Ehternet chip fits in a single old-style 30 pin RAM socket!
For those who care, a DragonBall is officially a Motorola 68EZ328 processor running at 16 MHz. CLinux is a version of our favorite OS designed for chips with no MMU (Memory Management Unit), and fits nicely into the 2 Megs of flash ROM.