I'm not sure what the problem is. If you're making your money selling custom hardware, then the driver code shouldn't matter that much (since it can't be used without the HW anyway). The only time a problem would arise is if you somehow put the entire (reverse-engineerable) footprint & logic within the kernel portion itself, which is simply poor design practice. You should be able to write a relatively generic kernel section to support specific chips, and/or a generic interface that's tapped into by your non-kernel based application code. Then you release the kernel changes, knowing that they're useless to your competitors without your custom app. code.
Actually, IMHO the culprit is the half-@ssed regulation from '96 Telecom Act. Either more or less regulation would have worked better. Instead of laying down solid protective restrictions, or letting the market sort itself out, they tried to play psychologist with the telecom giants --and blew it big time.
The whole idea of having one big (monopoly) company provide the infrastructure, and another the actual service, has failed utterly in the case of broadband. In DSL, the infrastructure guys strangled the service companies because they were in fact competitors. In cable modems, the unregulated growth & consolidation of the service carriers, along with the monopoly status of the 'customer' networks, caused the one big player to eventually fold.
Notice that the individual cable-system owned broadband programs are still going strong (well, ok, still going). That's because the company that controls the market gets a direct benefit from helping those programs make money. In contrast, even the companies that had a stake in the @home service still saw it a joint venture --with potential rivals. Whether or not it made money didn't matter to them, and since they're all local monopolies, they had no incentive to even pay @home on time (who else were they going to 'sell' to?).
Heh, just the other day one of my cable-modeming friends suggested that I could make up the difference in access rates (vs. dialup) by axing my land line. He's now one of the lucky ones who gets to do all his surfing from work from now on.... (ironically, my cable system doesn't use @home, so if I had followed his advice I'd still be ok --for now)
Ditto for open source Linux DVD players. They'll be just like encryption software used to be. You'll have to download them from some European country since they'll never be part of Red Hat or Mandrake. This is a big win for the MPAA.
Oh yeah, and we all know how effective that limitation was! Fifteen minutes on a scandanavian FTP site instead of five on a local one. Oh, the horror! And I think DVD playback is a big enough 'killer app' to motivate Joe Shmoe to go out and download it, don't you? Esp. if the alternative is paying for commercial Windows software to do it.
Re:DeCSS decision now available on EFF site
on
DMCA 2, Freedom 0
·
· Score: 1
I am dissappointed. More attention should have been paid to the fact that DVD playing should not be a cartel.
That's not the point at all. It's not the job of the judiciary to decide whether a law is moral or not; that's for the legistlature. The courts' job is to determine if it's legal, ie. constitutional. This is where the court failed in its duty. The First Ammendment issue is the *only* issue that matters here, and the court apparently blew it off. If simply linking to the code on another site is illegal, shouldn't they have been arresting those guys with the t-shirts? The double-standard here is amazing.
Nah, I've done this even when I didn't have an alarm clock (or I forgot to set it). The bad thing was that for about a year during college, I would consistently wake up about an hour before I was supposed to. It got to the point where I'd deliberately try to convince myself that my appointments were an hour later the next morning, so I wouldn't lose that extra hour of sleep!
Heh, maybe my PocketPC should be called Torgo. I dropped it on its head as a child, and now it limps around and has trouble communicating (and it is a tool of evil)...
Re:My take on the whole thing
on
XBox Released
·
· Score: 1
Not only that, but wasn't the XBOX supposed to be aimed more at the older audience (even compared to PS2)? It's got the PC crossover potential, adult oriented ads (sort of), and even oversized controllers for adult hands.
But now the range of products out for PS2 seems to make it the more interesting choice for adult gamers. Factor in the relatively high correlation between adult gamers and/. geeks, and the XBOX may have a problem here...
Re:too bad it was going to be a big leap forward
on
XBox Released
·
· Score: 1
Well, if M$ was smart, they could have made a version of their Strategic Commander for XBOX, to compensate for the lack of a full keyboard. Then, any serious RTS-er would buy it (like racers buy wheels), while the console-kiddies could make do with less-than-optimal gamepad functions. There's no reason to say that Halo had to be 'dumbed down' for the XBOX.
1)No installation problems; no need to set up a firewall; etc.
2)No unrecoverable outages (unless the entire national network goes down). My current service is quite reliable, unlike the broadband some friends have.
3)No strange usage/port/upload/download restrictions. I can run a LAN, or a Linux box, and access or send whatever I want.
4)Cheaper. I use my cell as my 2nd (voice) line, and I wouldn't want to cancel it even if I got broadband. (YMMV of course)
5)Cheap bandwidth at work:-) (And if I lose my job, I'll need that extra $50/month anyway!)
Maybe all students should change majors after the first year or two.
I'd agree with that. I realized in my second year that I really wanted to be a CS major, but in my Eng. school, it was *already* too late to switch without taking an extra year (which my 'rents wouldn't pay for *grr*). So now I'm an EE doing lots of software, and wishing I had that extra bit of CS design theory to help me along.
You've got a point here. I once got a assigned a project to add multi-language support to an existing product. The job involved sifting through tons of code to find and convert all the strings to be translated. I could have just gone through it by hand with a search tool, which was simple & obvious but also incredibly tedious. Instead, I spent lots of time writing automated scripts to parse out the strings and reformat them. It probably took just as long if not longer, but it was a helluva lot more fun, and kept my coding skills sharp.
If I had approached this project with less enthusiasm, I would have wound up doing things the boring way, and reinforced feelings of apathy towards the job. By putting a little thought & effort into it, I had fun and still got the job done.
If you like computer systems but not programming, you should really be taking an IT training program, or else a Computer Engineering (CPE) major. That said, it shouldn't be too hard to transition into IT from CS, as long as you've got networking experience from somewhere (job/internship, school IS department, etc.).
BTW, CPEs are a relatively new major, and are usually a EE/CS hybrid program. Unfortunately, my school didn't introduce it until a few years after my time, so the best I could manage was a BSEE with a 4th-year Computer concentration. I still ended up doing hybrid HW/SW work (which I wnated), but I feel outclassed by others who got more balanced training in true CPE programs.
The only problem with this is that I think EE->software works better in the job market than CS->hardware. There's still a perception that 'anybody' can program, and having an EE (hardware) background can be an asset. It's much harder to convince someone that you can design hardware just because you took a couple of classes, and even if you get hired in a EE department, you'll probably end up as the de-facto 'software guy'.
If you're that close to graduation, I'd say muddle through and get your degree, then think about other options. You may find a job that's more interesting than your course-work; otherwise, you could always take a short-term job and start saving for grad school. Lots of people end up taking radically different majors in grad school, so it should be do-able.
Indeed, the Supreme Court has never upheld a prior restraint, even faced with the competing interest of national security or the Sixth Amendment right to a fair trial.
Err, this may be a dumb question, but what about judges' gag orders? Aren't they limiting free speech in exactly the same way? Seems like the Sixth is beating the First in that case. Or have the Supremes never bothered to rule on that?
You're right about this. I worked with a team like this for a little while, and basically everyone in the group would be called 'quirky' by some. But they all understood each other, and the various talents and areas of interest meshed very well. That team did more with five or six people than teams of 20+ managed at other sites.
...to being a successful 'quirky' engineer, is to become an expert in something that no-one else wants to deal with. Case in point: The Analog Guy(tm). These are the old guys with unruly white hair who walk around being grouchy and complaining about politics all day, but they'll always have a job because they're an endangered species. None of the younger engineers (myself included) have the time or patience, much less the experience, to deal with sticky analog issues that crop up in modern circuits.
To solve this, every engineering shop that uses custom PCB designs has at least one old radio engineer on staff. As long as he comes through with an answer when needed (and since he's the only one who knows what he's talking about, that's pretty easy), it doesn't matter if he sleeps at his desk the rest of the day. Of course most of these guys do know what they're talking about, but it's still a pretty cushy job.
Thus, the golden rule of the 'quirky' senior engineer: Never get involved with a task that requires deliverables. Set yourself up in a position where you only have to complain about other people's work. (Sort of like the 'software guru' in old Dilbert cartoons:)
The whole problem is infrastructure. Building new wiring is much more difficult than you seem to realize, and the only existing network that's laid out correctly for broadband is the phone network. The phone companies have proven to be very good at sabotaging competitive efforts that use their networks.
Cable modem is great for those areas that can retrofitted for it, but most cable networks aren't laid out correctly, so you're almost back to square one. BTW, the reason the independent cable modem companies are dying is because they have to deal with the local cable companies, and take what they can get; those cable companies that run their own services are doing pretty well, AFAICT.
I also think you overestimate the number of people who really care enough about broadband to go through the hassle of setup and then pay extra each month. If you work with other geeks, as I do, you may get a distorted impression of the overall market. Trust me, Joe Sixpack isn't going to get broadband until it's cheaper and easier (or the porn gets better than what's at the sleazy video store down the street;) Not everyone has $50 to burn each month.
There's no question that the 75GXP has serious problems. The only question left is whether the newer 60GXP shares those problems. Initially, I think it was sort of assumed that the new drives would fix the problems; but now I'm not so sure. I've heard rumors that the 60's are starting to fail as well, but by the time anyone can sort out the statistics it'll be too late for most people.
All I know is that I wouldn't buy an IBM drive right now. As long as they're keeping people in the dark, why should I take the risk? I'm used to company marketoids denying stuff like this; but if they won't acknowledge the problem, how can we ever be sure that it's fixed?
The LD 50 isn't really relevant. The issue is long term damage, not immediate death. Something that is lethal in the short-term at high doses may be relatively harmless when taken long-term in smaller doses, or vice versa.
The dosage makes all the difference. Several types of venom and toxic substances are used therapeuticaly in small doses; the best example is botchulinum(sp?) toxin, which I'm sure has an LD 50 *way* below caffeine's, yet is used as a localized muscle relaxant in a number of medical treatments.
I hate to play devil's advocate here, but addiction and dependence are not the same thing. Caffeine dependence is pretty common, but true caffeine addiction (where the user keeps increasing the dose until death or illness stops them) is fairly rare. Most of the so-called 'hard' drugs are the ones that cause high rates of addiction; they may not even produce that strong of a dependence, but the craving and build-up of tolerance are always there. This is part of the reason for the confusion between different definitions of 'addictive' substances.
I'll agree that THC products shouldn't be classified in the way as narcotics, but some drugs really are dangerous.
True, but they only seem to mention other teams as an afterthought, when it's convenient or unavoidable. Since we never see the other teams, nor does their behind-the-scenes work ever seem to affect the story-arc (while SG-1's discoveries regularly change *everything*), it's easy to conclude that they don't really do much.
The only plausible explanation I can think is that SG-1 is the primary team, and most of the other teams are just backups. Even saying that they're advance scouts doesn't explain it; they go on missions that wouldn't be appropriate for that role (including diplomatic missions --where's SG-9!?;) So the other teams must be cleanup crews who do the mundane stuff, while SG-1 is the only real ecounter group.
Slashdot Mirror
I'm not sure what the problem is. If you're making your money selling custom hardware, then the driver code shouldn't matter that much (since it can't be used without the HW anyway). The only time a problem would arise is if you somehow put the entire (reverse-engineerable) footprint & logic within the kernel portion itself, which is simply poor design practice. You should be able to write a relatively generic kernel section to support specific chips, and/or a generic interface that's tapped into by your non-kernel based application code. Then you release the kernel changes, knowing that they're useless to your competitors without your custom app. code.
The whole idea of having one big (monopoly) company provide the infrastructure, and another the actual service, has failed utterly in the case of broadband. In DSL, the infrastructure guys strangled the service companies because they were in fact competitors. In cable modems, the unregulated growth & consolidation of the service carriers, along with the monopoly status of the 'customer' networks, caused the one big player to eventually fold.
Notice that the individual cable-system owned broadband programs are still going strong (well, ok, still going). That's because the company that controls the market gets a direct benefit from helping those programs make money. In contrast, even the companies that had a stake in the @home service still saw it a joint venture --with potential rivals. Whether or not it made money didn't matter to them, and since they're all local monopolies, they had no incentive to even pay @home on time (who else were they going to 'sell' to?).
Heh, just the other day one of my cable-modeming friends suggested that I could make up the difference in access rates (vs. dialup) by axing my land line. He's now one of the lucky ones who gets to do all his surfing from work from now on.... (ironically, my cable system doesn't use @home, so if I had followed his advice I'd still be ok --for now)
Oh yeah, and we all know how effective that limitation was! Fifteen minutes on a scandanavian FTP site instead of five on a local one. Oh, the horror! And I think DVD playback is a big enough 'killer app' to motivate Joe Shmoe to go out and download it, don't you? Esp. if the alternative is paying for commercial Windows software to do it.
That's not the point at all. It's not the job of the judiciary to decide whether a law is moral or not; that's for the legistlature. The courts' job is to determine if it's legal, ie. constitutional. This is where the court failed in its duty. The First Ammendment issue is the *only* issue that matters here, and the court apparently blew it off. If simply linking to the code on another site is illegal, shouldn't they have been arresting those guys with the t-shirts? The double-standard here is amazing.
Nah, I've done this even when I didn't have an alarm clock (or I forgot to set it). The bad thing was that for about a year during college, I would consistently wake up about an hour before I was supposed to. It got to the point where I'd deliberately try to convince myself that my appointments were an hour later the next morning, so I wouldn't lose that extra hour of sleep!
Yes, but they were ill-tempered, mutated sea-bass^H^H^H sharks!
(with las-ers on their foreheads...)
Darnit, SlashCode 2.x strikes again! This was a reply to one of the comments it's now sitting next to. Doh!
Exactly. The DMCA outlaws innovation unless it's sponsored by a major corporation. And how much real innovation do get out of them anyway?
Heh, maybe my PocketPC should be called Torgo. I dropped it on its head as a child, and now it limps around and has trouble communicating (and it is a tool of evil)...
Not only that, but wasn't the XBOX supposed to be aimed more at the older audience (even compared to PS2)? It's got the PC crossover potential, adult oriented ads (sort of), and even oversized controllers for adult hands.
/. geeks, and the XBOX may have a problem here...
But now the range of products out for PS2 seems to make it the more interesting choice for adult gamers. Factor in the relatively high correlation between adult gamers and
Well, if M$ was smart, they could have made a version of their Strategic Commander for XBOX, to compensate for the lack of a full keyboard. Then, any serious RTS-er would buy it (like racers buy wheels), while the console-kiddies could make do with less-than-optimal gamepad functions. There's no reason to say that Halo had to be 'dumbed down' for the XBOX.
Reasons dialup is better than broadband:
:-) (And if I lose my job, I'll need that extra $50/month anyway!)
1)No installation problems; no need to set up a firewall; etc.
2)No unrecoverable outages (unless the entire national network goes down). My current service is quite reliable, unlike the broadband some friends have.
3)No strange usage/port/upload/download restrictions. I can run a LAN, or a Linux box, and access or send whatever I want.
4)Cheaper. I use my cell as my 2nd (voice) line, and I wouldn't want to cancel it even if I got broadband. (YMMV of course)
5)Cheap bandwidth at work
Maybe all students should change majors after the first year or two.
I'd agree with that. I realized in my second year that I really wanted to be a CS major, but in my Eng. school, it was *already* too late to switch without taking an extra year (which my 'rents wouldn't pay for *grr*). So now I'm an EE doing lots of software, and wishing I had that extra bit of CS design theory to help me along.
You've got a point here. I once got a assigned a project to add multi-language support to an existing product. The job involved sifting through tons of code to find and convert all the strings to be translated. I could have just gone through it by hand with a search tool, which was simple & obvious but also incredibly tedious. Instead, I spent lots of time writing automated scripts to parse out the strings and reformat them. It probably took just as long if not longer, but it was a helluva lot more fun, and kept my coding skills sharp.
If I had approached this project with less enthusiasm, I would have wound up doing things the boring way, and reinforced feelings of apathy towards the job. By putting a little thought & effort into it, I had fun and still got the job done.
If you like computer systems but not programming, you should really be taking an IT training program, or else a Computer Engineering (CPE) major. That said, it shouldn't be too hard to transition into IT from CS, as long as you've got networking experience from somewhere (job/internship, school IS department, etc.).
BTW, CPEs are a relatively new major, and are usually a EE/CS hybrid program. Unfortunately, my school didn't introduce it until a few years after my time, so the best I could manage was a BSEE with a 4th-year Computer concentration. I still ended up doing hybrid HW/SW work (which I wnated), but I feel outclassed by others who got more balanced training in true CPE programs.
The only problem with this is that I think EE->software works better in the job market than CS->hardware. There's still a perception that 'anybody' can program, and having an EE (hardware) background can be an asset. It's much harder to convince someone that you can design hardware just because you took a couple of classes, and even if you get hired in a EE department, you'll probably end up as the de-facto 'software guy'.
If you're that close to graduation, I'd say muddle through and get your degree, then think about other options. You may find a job that's more interesting than your course-work; otherwise, you could always take a short-term job and start saving for grad school. Lots of people end up taking radically different majors in grad school, so it should be do-able.
Indeed, the Supreme Court has never upheld a prior restraint, even faced with the competing interest of national security or the Sixth Amendment right to a fair trial.
Err, this may be a dumb question, but what about judges' gag orders? Aren't they limiting free speech in exactly the same way? Seems like the Sixth is beating the First in that case. Or have the Supremes never bothered to rule on that?
You're right about this. I worked with a team like this for a little while, and basically everyone in the group would be called 'quirky' by some. But they all understood each other, and the various talents and areas of interest meshed very well. That team did more with five or six people than teams of 20+ managed at other sites.
...to being a successful 'quirky' engineer, is to become an expert in something that no-one else wants to deal with. Case in point: The Analog Guy(tm). These are the old guys with unruly white hair who walk around being grouchy and complaining about politics all day, but they'll always have a job because they're an endangered species. None of the younger engineers (myself included) have the time or patience, much less the experience, to deal with sticky analog issues that crop up in modern circuits.
:)
To solve this, every engineering shop that uses custom PCB designs has at least one old radio engineer on staff. As long as he comes through with an answer when needed (and since he's the only one who knows what he's talking about, that's pretty easy), it doesn't matter if he sleeps at his desk the rest of the day. Of course most of these guys do know what they're talking about, but it's still a pretty cushy job.
Thus, the golden rule of the 'quirky' senior engineer: Never get involved with a task that requires deliverables. Set yourself up in a position where you only have to complain about other people's work. (Sort of like the 'software guru' in old Dilbert cartoons
The whole problem is infrastructure. Building new wiring is much more difficult than you seem to realize, and the only existing network that's laid out correctly for broadband is the phone network. The phone companies have proven to be very good at sabotaging competitive efforts that use their networks.
;) Not everyone has $50 to burn each month.
Cable modem is great for those areas that can retrofitted for it, but most cable networks aren't laid out correctly, so you're almost back to square one. BTW, the reason the independent cable modem companies are dying is because they have to deal with the local cable companies, and take what they can get; those cable companies that run their own services are doing pretty well, AFAICT.
I also think you overestimate the number of people who really care enough about broadband to go through the hassle of setup and then pay extra each month. If you work with other geeks, as I do, you may get a distorted impression of the overall market. Trust me, Joe Sixpack isn't going to get broadband until it's cheaper and easier (or the porn gets better than what's at the sleazy video store down the street
There's no question that the 75GXP has serious problems. The only question left is whether the newer 60GXP shares those problems. Initially, I think it was sort of assumed that the new drives would fix the problems; but now I'm not so sure. I've heard rumors that the 60's are starting to fail as well, but by the time anyone can sort out the statistics it'll be too late for most people.
All I know is that I wouldn't buy an IBM drive right now. As long as they're keeping people in the dark, why should I take the risk? I'm used to company marketoids denying stuff like this; but if they won't acknowledge the problem, how can we ever be sure that it's fixed?
The LD 50 isn't really relevant. The issue is long term damage, not immediate death. Something that is lethal in the short-term at high doses may be relatively harmless when taken long-term in smaller doses, or vice versa.
The dosage makes all the difference. Several types of venom and toxic substances are used therapeuticaly in small doses; the best example is botchulinum(sp?) toxin, which I'm sure has an LD 50 *way* below caffeine's, yet is used as a localized muscle relaxant in a number of medical treatments.
I hate to play devil's advocate here, but addiction and dependence are not the same thing. Caffeine dependence is pretty common, but true caffeine addiction (where the user keeps increasing the dose until death or illness stops them) is fairly rare. Most of the so-called 'hard' drugs are the ones that cause high rates of addiction; they may not even produce that strong of a dependence, but the craving and build-up of tolerance are always there. This is part of the reason for the confusion between different definitions of 'addictive' substances.
I'll agree that THC products shouldn't be classified in the way as narcotics, but some drugs really are dangerous.
True, but they only seem to mention other teams as an afterthought, when it's convenient or unavoidable. Since we never see the other teams, nor does their behind-the-scenes work ever seem to affect the story-arc (while SG-1's discoveries regularly change *everything*), it's easy to conclude that they don't really do much.
;) So the other teams must be cleanup crews who do the mundane stuff, while SG-1 is the only real ecounter group.
The only plausible explanation I can think is that SG-1 is the primary team, and most of the other teams are just backups. Even saying that they're advance scouts doesn't explain it; they go on missions that wouldn't be appropriate for that role (including diplomatic missions --where's SG-9!?