you'll notice the case seems to hinge on Microsoft's monopoly status.
If they did not have a monopoly on desktop computer systems, this type of lawsuit wouldn't be a problem for them. Since, due to all sorts of vendor lock-in promoted by Microsoft itself, it is difficult for users to pick a different desktop, the lack of security in their software ( i.e. buffer overflows everywhere )... I don't know. Since I'm not a lawyer, this is where the case falls apart for me.
But maybe a monopolist which continues to abuse it's position _should_ be held to a higher standard than others ? Is it not arguable that MS has the resources required to audit all of it's code and fix such issues ? Maybe not technically true, but arguable in court...
I'm pretty sure that as long as the licensed operator provides the service they're licensed for, anything else they might do is OK, providing they aren't causing extra interference and are operating under all of the limits they're required to ( serving x percentage of population with whatever required coverage, etc ).
To be more explicit, as long as the TV stations which are broadcasting these encrypted signals _also_ broadcast a signal you can pick up on your normal TV, there's no regulation I'm aware of saying they can't charge for the ability to descramble an encoded signal.
Having said that, yes, this does seem like a bastardization of the original intent of PUBLIC airwaves and of course, I Am Not A Lawyer. If you don't like it, start a grass-roots campaign and write your congresscritter.
I agree with other posts that the point is moot. Who is going to pay a monthly fee to have these boxes when you can have NetFlix instead ? Of course, I wonder why you'd buy a Microsoft operating system when there are alternatives, too...
This is just a test, though. Hopefully the results will tell them they have to drop the monthly fee and extend the viewable hours. If they make it cheap enough, they just might have a killer app. What we want is video on demand without montly fees. It must be competitive with DirectTV, cable, NetFlix and BlockBuster, which this is not. Pricing may in fact be their biggest problem with this.
If solar cells were cheap enough, we could cover large, unpopulated desert-climate areas with them.
There is plenty of otherwise unused rooftop space... if solar cells were cheap as roof tiles, it wouldn't matter if efficiency was 10%, it'd still make economic sense to put them in, and the power generated would make a difference.
Given this, and the thrust of the article being about *cost*, not *efficiency*, I fail to see how the parent article is "insightful".
Granted, we have a long way to go before large solar panels are *that* cheap, but still, it's not all about efficiency. You could make super-efficient solar cells, but if the cost was too high, it wouldn't matter for consumer-grid energy production. High-performance, high-cost solar cells are only likely to be used for special-use purposes, like in solar-powered mobile devices ( aircraft, cars, toys, etc ).
I think part of the reason solar tech has been so slow to catch on in terms of rooftops is that a lot of the research money has been going to make cells more efficient, not cheaper.
A contract organization is looking to solve your short-term needs with highly qualified individuals, and is hoping for repeat business. A headhunter wants to fill shoes, and is looking for volume business. A contract organization can act as a headhunter, but I'd only use them as such after giving them a 'trial' period.
One thing to look out for, however- don't use a contractor if they might be able to steal YOUR clients. It's not worth the risk, and the temptation might be too much if your business ( or the client's projects ) are small enough to be stolen.
There are two types of decent headhunter. One is a lone individual, who has a few friends from previous businesses who they know well and can count on. They are rare, since once everyone is placed, the business is over.
The other is a smallish organization who caters to a niche market, specializes, and even occasionally holds training seminars for the people they employ. Companies like this have 401(k) plans an other stuff you might be surprised to hear a contract work businees offer. Also rare.
On the other hand, if you had a really great work environment and paid well, you wouldn't have a hard time attracting good talent, would you ?!? Set those things right, and you won't _need_ a pimp... er, headhunter.
Trust me, when you're taking some mid-level 'weed-out the weak' physics/math/engineering course and EVERYONE you are competing against was in the top 2% of their graduating class with unbelievable SAT scores, it makes a difference.
Bell-curve grading in such a scenario can be a real bitch, and profs for whatever reason ( lazyness is my guess ) often use it anyway.
In my experience, the teachers at less-difficult-to-enter schools have to work a little harder to explain the course material to students, and thus the classes are much more understandable and the learning experience better... and of course, easier. But not necessarily due to grading- it's easier due to the fact that you don't have to either already know the material or teach it to yourself ( or hire a tutor ), which happens at places like MIT and Stanford because the 'teacher' is some math research grad student who is a wiz but is teaching because he has to, not because it's his job, and can't understand how you wouldn't 'get' such basic material. I know this might be taken the wrong way, but at a big-name research university, you can count yourself lucky if your math secton leader can speak understandable english- and they won't know jack about teaching.
If you really want to learn about Diff Eq's, you might be better off taking the course from Foothill College rather than Leland Stanford, Jr University. Sad but true.
Of course, you'd rather do your research project at one of the big-name schools, and it always looks better on a resume...
The processor upgrade we installed was from Sonnet and they may be available from one other manufacturer as I recall. As for pricing and availability, I'll let you do your own Google search, but I seem to remember the prices beating the heck out of getting a new machine.
The upgrade does, as I implied earlier, involve putting a fan in your cube ( awww ) but it's a rather quiet fan, probably you won't notice the difference in noise. In El Salvador, you might want that extra cooling anyway.
Installing the processor upgrade involves a nearly complete disassembly ( and hopefully, reassembly ) of your machine, so unless you're technically skilled and inclined to do so, you may want to find someone skilled to do the job. Having said that, if you're careful, good at following directions and don't lose any of the many screws you'll have to remove, it's not _that_ hard...
There *is* a way to change the start-button so it causes 'sleep' instead of trying to shut down your machine. Do a search of Apple's support site, you might find something on that. I'd tell you but I wasn't the one to do it and don't know how it's done. It's also a simple matter to disconnect the cable to the switch, which you might not want to do unless you have a keyboard with a power button... but that fixes any start button issues! I know this last bit works OK because someone had a cube where the button went flaky.
Just to bring this post back on-topic, unless I'm reading the technote wrong and *only* the dual-processor machines are affected, I think the cube would potentially have some of the problems described in the Apple Technote on this issue but your network connection might save it, as is likely the case with our cubes, which are of course quite connected to a high-speed switch...
It's pretty clear; it's all networking ( and slow networking at that ), which explains why some people with the described machines have no problems. All of the other problems reported are likely of the typical well-what-did-you-do-to-your-machine variety, not a direct result of the update.
I know of two computers which have had the 10.2.8 update applied. One is a cube with a processor upgrade, the other is a G4/533 single-CPU AGP machine.
No problems to report with either. From what I've read, it may rather specifically be DHCP which causes problems. The update was probably only pulled because, well, screwing up someone's networking is a very serious matter. It could be difficult to download other updates!
I have to say, those Cubes are fantastic machines, especially considering the processor upgrades which are available for them. One interesting thing I learned when we did the processor upgrade: the fanless cube actually had a built-in bracket for a fan, which comes with the CPU upgrade.
Seriously, it's well worth the money to get yourself a processor upgrade card given the machine you've just bought. I'm pretty sure you can upgrade to 800Mhz or 1GHz for a reasonable price. This is what has kept prices for these machines and Cubes so high. It'll more than double your performance considering the added cache, and that'll make a big difference if you're compiling. Also max out your RAM, it's cheap.
Rather than complain about the puck, dual-button+scrollwheel optical USB mice are super-cheap, get one.
Project Builder is already pretty damn sweet, and has been for some time. Try it out now, it'll make the transition to Xcode smoother for you.
I work for an engineering-based consulting business. While the majority of the applications used around here do not push newer G4 machines, there are definitely applications in the 3D modeling space which require well, as much computing power as possible.
If an engineer can do in 5 minutes what would have taken 15 minutes, that is a *big* deal. Just because *you* don't need the extra computing power doesn't mean it's not useful to someone else. Having this type of power available in an easy-to-use environment like OS X which also allows direct access to a bunch of unix-based engineering tools is the dream that a lot of engineering shops, which now often employ Windoze boxen with shells/X to unix boxen, would love if and when they try it.
This comment of mine doesn't even touch on the demands editing digital video makes of a machine, which is something home users do increasingly. You wouldn't ask the question "is there really a need" if you had ever done any extensive DV editing.
Of course, there is the class of applications which most taxes any computer- the 3D video game. You could, right now, write a game which would max out even the dual CPU G5. Nobody has written it, but that's just because it'd be hard to market. Look for it in a couple of years, though.
Let's not forget that these systems are running nearly identical binaries, which, while it seems fair, is not.
The binaries are optimized for the G4. Optimization for the G5 will create quite different binaries which could run _much_ faster on the G5.
While these tests are a great comparison for performance we'll see today, apps compiled with newer G5 optimizing compilers will push the top numbers even a bit farther, as will future OS updates. Users with G5 a year from now might look back on these numbers and wonder why they were so low...
No, you need the ltteres, it jsut deos not mtaetr waht odrer tehy are in.
At least, that's what the submission says, and from what I can tell, it's true.
Your post is _much_ harder to read than :
So deos tihs maen taht we don't need the mdilde leterts at all?
Anyone have a link to an article yet? I'd love to have a scientific journal article to throw at people when they point out spelling/typos in casual email.
Looking over these comments, it seems the solutions available, from best to worst, are:
1) You're worried about how it looks on OS X? Get a machine which runs OS X ! You'll be glad you did.
2) Check your page on both Mozilla and something else which uses the KHTML open-source engine in Safari, as those are open-source cross-platform rendering engines.
3) use W3C validation engines and tell users any problem is with the browser.
Slashdot Mirror
you'll notice the case seems to hinge on Microsoft's monopoly status.
... I don't know. Since I'm not a lawyer, this is where the case falls apart for me.
If they did not have a monopoly on desktop computer systems, this type of lawsuit wouldn't be a problem for them. Since, due to all sorts of vendor lock-in promoted by Microsoft itself, it is difficult for users to pick a different desktop, the lack of security in their software ( i.e. buffer overflows everywhere )
But maybe a monopolist which continues to abuse it's position _should_ be held to a higher standard than others ? Is it not arguable that MS has the resources required to audit all of it's code and fix such issues ? Maybe not technically true, but arguable in court...
I'm pretty sure that as long as the licensed operator provides the service they're licensed for, anything else they might do is OK, providing they aren't causing extra interference and are operating under all of the limits they're required to ( serving x percentage of population with whatever required coverage, etc ).
To be more explicit, as long as the TV stations which are broadcasting these encrypted signals _also_ broadcast a signal you can pick up on your normal TV, there's no regulation I'm aware of saying they can't charge for the ability to descramble an encoded signal.
Having said that, yes, this does seem like a bastardization of the original intent of PUBLIC airwaves and of course, I Am Not A Lawyer. If you don't like it, start a grass-roots campaign and write your congresscritter.
I agree with other posts that the point is moot. Who is going to pay a monthly fee to have these boxes when you can have NetFlix instead ? Of course, I wonder why you'd buy a Microsoft operating system when there are alternatives, too...
This is just a test, though. Hopefully the results will tell them they have to drop the monthly fee and extend the viewable hours. If they make it cheap enough, they just might have a killer app. What we want is video on demand without montly fees. It must be competitive with DirectTV, cable, NetFlix and BlockBuster, which this is not. Pricing may in fact be their biggest problem with this.
If solar cells were cheap enough, we could cover large, unpopulated desert-climate areas with them.
There is plenty of otherwise unused rooftop space... if solar cells were cheap as roof tiles, it wouldn't matter if efficiency was 10%, it'd still make economic sense to put them in, and the power generated would make a difference.
Given this, and the thrust of the article being about *cost*, not *efficiency*, I fail to see how the parent article is "insightful".
Granted, we have a long way to go before large solar panels are *that* cheap, but still, it's not all about efficiency. You could make super-efficient solar cells, but if the cost was too high, it wouldn't matter for consumer-grid energy production. High-performance, high-cost solar cells are only likely to be used for special-use purposes, like in solar-powered mobile devices ( aircraft, cars, toys, etc ).
I think part of the reason solar tech has been so slow to catch on in terms of rooftops is that a lot of the research money has been going to make cells more efficient, not cheaper.
A contract organization is looking to solve your short-term needs with highly qualified individuals, and is hoping for repeat business. A headhunter wants to fill shoes, and is looking for volume business. A contract organization can act as a headhunter, but I'd only use them as such after giving them a 'trial' period.
One thing to look out for, however- don't use a contractor if they might be able to steal YOUR clients. It's not worth the risk, and the temptation might be too much if your business ( or the client's projects ) are small enough to be stolen.
There are two types of decent headhunter. One is a lone individual, who has a few friends from previous businesses who they know well and can count on. They are rare, since once everyone is placed, the business is over.
The other is a smallish organization who caters to a niche market, specializes, and even occasionally holds training seminars for the people they employ. Companies like this have 401(k) plans an other stuff you might be surprised to hear a contract work businees offer. Also rare.
On the other hand, if you had a really great work environment and paid well, you wouldn't have a hard time attracting good talent, would you ?!? Set those things right, and you won't _need_ a pimp... er, headhunter.
Trust me,
when you're taking some mid-level 'weed-out the weak' physics/math/engineering course and EVERYONE you are competing against was in the top 2% of their graduating class with unbelievable SAT scores, it makes a difference.
Bell-curve grading in such a scenario can be a real bitch, and profs for whatever reason ( lazyness is my guess ) often use it anyway.
In my experience, the teachers at less-difficult-to-enter schools have to work a little harder to explain the course material to students, and thus the classes are much more understandable and the learning experience better... and of course, easier. But not necessarily due to grading- it's easier due to the fact that you don't have to either already know the material or teach it to yourself ( or hire a tutor ), which happens at places like MIT and Stanford because the 'teacher' is some math research grad student who is a wiz but is teaching because he has to, not because it's his job, and can't understand how you wouldn't 'get' such basic material. I know this might be taken the wrong way, but at a big-name research university, you can count yourself lucky if your math secton leader can speak understandable english- and they won't know jack about teaching.
If you really want to learn about Diff Eq's, you might be better off taking the course from Foothill College rather than Leland Stanford, Jr University. Sad but true.
Of course, you'd rather do your research project at one of the big-name schools, and it always looks better on a resume...
I finished my BS CS degree, for what it's worth.
Hilarious!
Seriously, this MIT project is a great resource.
The upgrade does, as I implied earlier, involve putting a fan in your cube ( awww ) but it's a rather quiet fan, probably you won't notice the difference in noise. In El Salvador, you might want that extra cooling anyway.
Installing the processor upgrade involves a nearly complete disassembly ( and hopefully, reassembly ) of your machine, so unless you're technically skilled and inclined to do so, you may want to find someone skilled to do the job. Having said that, if you're careful, good at following directions and don't lose any of the many screws you'll have to remove, it's not _that_ hard...
There *is* a way to change the start-button so it causes 'sleep' instead of trying to shut down your machine. Do a search of Apple's support site, you might find something on that. I'd tell you but I wasn't the one to do it and don't know how it's done. It's also a simple matter to disconnect the cable to the switch, which you might not want to do unless you have a keyboard with a power button... but that fixes any start button issues! I know this last bit works OK because someone had a cube where the button went flaky.
Just to bring this post back on-topic, unless I'm reading the technote wrong and *only* the dual-processor machines are affected, I think the cube would potentially have some of the problems described in the Apple Technote on this issue but your network connection might save it, as is likely the case with our cubes, which are of course quite connected to a high-speed switch...
It's pretty clear; it's all networking ( and slow networking at that ), which explains why some people with the described machines have no problems. All of the other problems reported are likely of the typical well-what-did-you-do-to-your-machine variety, not a direct result of the update.
I know of two computers which have had the 10.2.8 update applied. One is a cube with a processor upgrade, the other is a G4/533 single-CPU AGP machine.
No problems to report with either. From what I've read, it may rather specifically be DHCP which causes problems. The update was probably only pulled because, well, screwing up someone's networking is a very serious matter. It could be difficult to download other updates!
I have to say, those Cubes are fantastic machines, especially considering the processor upgrades which are available for them. One interesting thing I learned when we did the processor upgrade: the fanless cube actually had a built-in bracket for a fan, which comes with the CPU upgrade.
Actually, according to your analysis, a Red Hat win would not shut down the SCO FUD machine regarding Linux.
They'd just have to be _really_ careful to avoid mentioning Red Hat when they spread their FUD. Which is a start, I suppose.
Seriously, it's well worth the money to get yourself a processor upgrade card given the machine you've just bought. I'm pretty sure you can upgrade to 800Mhz or 1GHz for a reasonable price. This is what has kept prices for these machines and Cubes so high. It'll more than double your performance considering the added cache, and that'll make a big difference if you're compiling. Also max out your RAM, it's cheap.
Rather than complain about the puck, dual-button+scrollwheel optical USB mice are super-cheap, get one.
Project Builder is already pretty damn sweet, and has been for some time. Try it out now, it'll make the transition to Xcode smoother for you.
I work for an engineering-based consulting business. While the majority of the applications used around here do not push newer G4 machines, there are definitely applications in the 3D modeling space which require well, as much computing power as possible.
If an engineer can do in 5 minutes what would have taken 15 minutes, that is a *big* deal. Just because *you* don't need the extra computing power doesn't mean it's not useful to someone else. Having this type of power available in an easy-to-use environment like OS X which also allows direct access to a bunch of unix-based engineering tools is the dream that a lot of engineering shops, which now often employ Windoze boxen with shells/X to unix boxen, would love if and when they try it.
This comment of mine doesn't even touch on the demands editing digital video makes of a machine, which is something home users do increasingly. You wouldn't ask the question "is there really a need" if you had ever done any extensive DV editing.
Of course, there is the class of applications which most taxes any computer- the 3D video game. You could, right now, write a game which would max out even the dual CPU G5. Nobody has written it, but that's just because it'd be hard to market. Look for it in a couple of years, though.
Let's not forget that these systems are running nearly identical binaries, which, while it seems fair, is not.
The binaries are optimized for the G4. Optimization for the G5 will create quite different binaries which could run _much_ faster on the G5.
While these tests are a great comparison for performance we'll see today, apps compiled with newer G5 optimizing compilers will push the top numbers even a bit farther, as will future OS updates. Users with G5 a year from now might look back on these numbers and wonder why they were so low...
No, you need the ltteres, it jsut deos not mtaetr waht odrer tehy are in. At least, that's what the submission says, and from what I can tell, it's true. Your post is _much_ harder to read than : So deos tihs maen taht we don't need the mdilde leterts at all? Anyone have a link to an article yet? I'd love to have a scientific journal article to throw at people when they point out spelling/typos in casual email.
Looking over these comments, it seems the solutions available, from best to worst, are :
1) You're worried about how it looks on OS X? Get a machine which runs OS X ! You'll be glad you did.
2) Check your page on both Mozilla and something else which uses the KHTML open-source engine in Safari, as those are open-source cross-platform rendering engines.
3) use W3C validation engines and tell users any problem is with the browser.