Users should be able to PAY for a guaranteed minimum amount of bandwidth during times when the available bandwidth is being totally used. During times when network bandwidth is unused the spare bandwidth should be equally available to everybody.
Then perhaps you'd prefer to think of their policy this way: you can pay nothing (beyond the regular cost of dorm housing), and you always have a guaranteed minimum of 56 kbps.
My first year in the dorms (admittedly, it's been 8 years) the total off-campus bandwidth for the several thousand students was just 40 megabits. By the time these policies were started, it was significantly more (60 Mb, if not more), but still not a lot given the number of people that had to share the limited resources.
I think one of the problems with the idea of a guaranteed minimum is the cost; bandwidth can be expensive, which is why we all get crappy DSL lines for our homes instead of a dedicated T1. And keep in mind that the vast majority of people don't really use very much - so if you're a big internet user, you're best off if everybody else pays as much as possible for their crappy DSL, and you pay a little bit more than them for slightly less-crappy DSL. If you start offering people dirt-cheap plans with a very low guaranteed minimum bandwidth, then you're going to have to be willing to pay a lot more to get a guarantee that seems palatable to you. Very few people are actually willing to pay what it would cost to have very good service.
I'm generally willing to trade multitasking for better battery life and apps that don't totally hose the phone. Or a memory card reader for a smaller, cheaper phone (memory cards? what are those anyway?). I couldn't really care less about apps. There's maybe a couple I'd get if I had an iPhone, but not many.
One of my college roommates was responsible for the dorm networks; they definitely had policies that pissed people off (usually the people who were abusing the network the most), but it was done so that the limited resources were usable by everybody. Among them:
P2P traffic was capped at 50% of total bandwidth.
There was a rolling monthly bandwidth cap. Exceed it, and you were capped at 56k modem speeds for about a week until you were under the cap again. (On-campus traffic was not counted, and not limited; many large downloads such as linux distros were mirrored on-campus.)
If you picked up a virus, you were isolated from the network. The only thing you could get to was windowsupdate.com, until you removed the virus and called the helpdesk to promise you had an antivirus installed.
I once encountered a processor bug that affected operation of the cache. As I recall, there were two workarounds: 1. Disable the cache. 2. Don't use the system with more than 112 MB of RAM.
Neither one of those is particularly easy to swallow on a shipping product.
As for the "3G router" - in some areas that would indeed be a market (some of Scandinavia has better wireless than wired connections, as does much of Africa) but it would very likely not be a new product, but more of an additional option for the Airport Extreme.
And Apple is SO well known for making products that are aimed at non-US markets...
</sarcasm>
Since Steve's second coming, I don't think they've made any significant regional adjustments to products other than localization, and legally required changes to be able to sell the product. Before that, there were a few products that were targeted at Japan (like the PowerBook 2400c).
Just to be clear, I'm not saying that 3G won't be added to some products. And I'm not complaining about the lack of international attention, since I've lived within walking distance of Apple HQ for the past 6 months...;-) I'm just saying Apple will work in its mysterious ways - if they think there's a market, they'll sell it. Sometimes that works out well (iPod, iPhone), sometimes not (iPod Hi-Fi), and sometimes it's rather middling (Apple TV).
When I hear "story" and "video game" in the same sentence, I always think of Marathon. It didn't have anything fancy like cut scenes, or three dimensions... But it had an evolving plot. Beyond the "you're human, they're alien, go kill them before they kill you" that most FPSs use. It's certainly not the best, but for a game released in 1994, it was pretty unusual.
I have no idea how much I've saved by not owning a car, but it's a lot. I made a conscious decision to live with only my bicycle for probably at least my first year post-college - after all, I've been biking everywhere for my last 5 years of college.
Eventually, when I have a family to share it with, I'll probably buy a car. But for now, I'm enjoying not having one: I'm losing weight; plus I can max out my IRA, 401k, and stock purchase plans, plus pay for my relatively expensive housing, and still have plenty of money left over.
It doesn't rain much, but when it does, I either wait for it to pass, or just bike through it. If it's heavy enough and I'm going far enough that I'd get soaked, I bring an extra set of clothes along. I go grocery shopping every week or two, and can carry all of it back in a large messenger bag, plus one pannier on the bike. (bonus: I never need plastic bags.)
My total bike maintenance comes to probably $200 per year. A couple times, I've rented a car (like when friends come to visit for a while), but I could probably rent a car for a week out of every month, and still come out ahead. Many communities have car sharing programs available, which can be quite useful for those who can almost live without a car.
I'm not sure I'd call it "poorly engineered." Differently engineered, certainly - most mice don't require you to lift your finger off the left side of the mouse to register a right-click. It's an elegant solution to the problem of balancing the power users' desire to have two mouse buttons with the simplicity of a one-button design for new users.
It's certainly not the mouse for everybody - especially anybody who plays first-person shooters. However, I've never used the right button for gaming, and I sometimes use only the keyboard.
I did a small project a few years ago with multi-core and multi-processor execution. The workload I used was a variety of parallel sorting algorithms. The fastest machines I had access to at the time were a dual-processor dual-core machine, and a dual-processor quad-core. That's 4 or 8 cores total, with two or four L2 caches (one for each pair of cores). I got close to perfect speedups when running on 2 cores; about 1.95x to 2x the speed of running on a single core. Going to 4 cores didn't give nearly as much speedup (about 3 to 3.5x the speed of a single core running), but that's mainly due to the data sharing between threads; since there were separate L2 caches, there's a significant latency hit any time data is moved between cores that don't share the cache.
Now by increasing the size of the dataset, the amount of work to be done increases faster than the amount of communication, so the speedup increases (that's why I gave a range of 3 to 3.5x). My datasets weren't huge - maybe 150 MB or so, I think, small enough that they could finish in under a second. Make the amount of data huge, and the speedup will get even closer to 4x on 4 cores. (Of course, running on all 8 cores of the 8-core machine was even less ideal - a speedup of 4 to 5x. But that incurs the extra penalty of storing data back to RAM when it has to move data between the different chips.)
More modern machines address these communication problems better (for example, by connecting the processors to each other, in addition to connecting them through the northbridge; or by having cache shared by all cores on a single chip), and will do even better.
The main problem, as you mention, is really software. However, you don't need parallel software to take advantage of many cores - running different programs helps as well. I wish I had more than 2 cores on my 4-year-old desktop, as running a video chat (with all the inherent encoding and decoding) at the same time as various other tasks (say browsing on a flash-heavy site) can be quite slow. On the other hand, now that dual-core machines are fairly mainstream, more applications will take advantage of multiple cores. And I can't think of a whole lot of applications that are both CPU-bound AND are inherently limited to 2x parallelism. Progress just keeps marching along...
On the other hand, the high-end product lines make up a significant portion of the profits for chip companies. They charge a huge premium for perfect chips that run at high speeds and low (or relatively low) power and temperature. The chips that aren't as perfect get sold for much less to the masses, running at lower clocks, lower voltages, and even with features disabled (as in the case where a chip has a defect).
Because of this, chipmakers will probably continue to have at least one product for the high-end market, and the benefits will continue to trickle down to the rest of us.
This is absolutely a correct sentiment. That's one reason why a hardware memory tester will do far better than software - it can run the memory at the high end of the voltage spec, or the low end, or bounce back and forth between the two. It can also do operations designed to heat the chips up, as the chips might slow down when hot. And you don't have to worry about running your OS in part of the RAM, either.
I agree about 110% with this. I used to do computer repair, and we had both a hardware memory tester, as well as various software programs that could test RAM. In my experience, the software testers could pick out about 5% of memory that was bad, the hardware tester could pick out about 95% of bad memory, and old-fashioned "swap it with known good memory" troubleshooting was required to find the last 5%. The software testers only pick out the memory that is very, very bad. I probably had better luck picking out bad RAM by looking at the manufacturer stamp on the chip - there are a number of manufacturers that I learned to trust (like Micron and Samsung). On the other hand, when I'd see a chip that had nothing on it but a label saying "512Mbx8" or something like that, it was almost guaranteed to be bad.
Lesson for the day: you get what you pay for. Don't buy cheap RAM, or your computer will crash incessantly.
That's the silliest claim I've ever heard. If they're not tracking each card individually, that would cause at least two major problems: When a card is redeemed, how do they know it hasn't been redeemed before? When a card is manufactured, how do they know it hasn't been manufactured before?
Now that's not to say that they might have difficulty identifying an illegitimate card. Especially if it's a fake that has been manufactured, and has not yet been redeemed. Or maybe their manufacturing folks don't talk with the iTunes folks, and they don't know whether a card has been manufactured - just whether it has been redeemed.
That would be a sensible aim if the iPhone was the market leader.
Now, show us some reference where the iPhone is shown to be leading the market.
According to some research from about 2 months ago, Apple has 17.3% of the smartphone market share, behind only Nokia at 38.9%. Going from 0% to 17.3% in 18 months is pretty impressive. Not only that, but Apple sells one model of iPhone (ignoring capacity and color differences). Nokia has many smartphones. I would not be surprised if the iPhone beats every particular model of Nokia's.
A few years from now, when Apple is selling the iPhone nano, iPhone classic, and several other models, they might well be crushing Nokia the same way they crushed everybody in the MP3 player market. (Or they might lose one for a change, you never know.)
Personally, I'm waiting for the iPhone Shuffle. Nothing more than an earpiece, a microphone, and a button to call one of your friends at random.
I think graphics chips are moving in the direction of saving power like CPUs have tended to do. For example, the MacBook Pro, which will run either the discrete GPU or the integrated one, depending on whether you prefer performance or battery life. Laptops will definitely go in this direction, because it's easier to reduce your power consumption than to increase the capacity of the battery without making it larger and heavier.
Desktops graphics cards will get better at power savings too, but probably not as much, because it's something most people seem to care about - energy is readily available, cooling isn't too difficult, and people buy computers based on initial price and performance, not long-term costs like energy usage.
That's only low until you have to design something like that. Or design its power supply, PCB layout, or heatsink. I designed a board once that had a 25W chip on it, and that had plenty of challenges. Of course, we had some extra restrictions that most systems don't have - extraordinarily limited space for a heatsink, another nearby heat source, very large ambient temperature range, and no fan allowed.
The highest-power Pentium 4 was 115 Watts, which is quite frankly ridiculous. From what I can tell, the Prescott had a 122 mm^2 die; that puts it at a power density of 94 W/cm^2. The E8400's 107 mm^2 die comes in at about 61 W/cm^2. A slide I've seen in many places, originally attributed to Intel, pegs a nuclear reactor at 200 W/cm^2.
With the P4, Intel realized that you eventually reach a point where it's not realistic to cool the CPU. Since then, basically everybody has started towards finding more ways to save power. After all, about half of those 65 Watts are from transistor leakage - the electricity is just wasted, not actually doing any work. As process technologies shrink, the leakage power increases. So, to be able to continue to cool our CPUs, they'll have to figure out ways to do more work with less energy.
Actually, we picked a few bacteria off an existing culture (the "big fuzzy mounds"), then spread them on the new dish. I forget what sort of tool we used, but they were at least spread thinly enough that you couldn't see the colonies with the naked eye.
In any case, you're certainly right - it was as rigorous as an 8th grade science class sounds. But still, when I'm backpacking and drinking water from a random stream somewhere, I'll be using the appropriate filter, not exposing the water to a little extra sunlight.;-)
You know, I've never been terribly confident in those since an 8th grade science experiment. I think the point was to try and come up with something that might cause a mutation in some bacteria - so we grew some in a petri dish, picked one colony (to get all the same type of bacteria), then grew it in a petri dish, then picked some out of that uniform batch and put them in another petri dish. We covered half with aluminum foil, then put it in the UV hood designed for disinfecting lab goggles. The bacteria had no problems growing after being supposedly "killed" by the UV (not to mention, no mutations like we wanted).
Of course, that was still an interesting result that our science teacher liked - so we spread more bacteria on a petri dish, put it uncovered into the UV hood (in case the plastic on the petri dish was opaque to UV or something like that), and ran it for much longer than normal. The bacteria still had no problems growing.
Now maybe it's just difficult to kill bacteria when you've put them on top of a nice big pile of food (aka agar)... But I really don't expect UV to kill anything. That doesn't mean I wouldn't drink the water - it just means that the UV shouldn't be the only line of defense if I think there's really a risk of getting sick (which there probably isn't a big one anyway...).
Agreed, talk to your school. Personally, my internships were $16-$18/hour. I got offers as high as $24/hour. It'll vary widely by major - mine (Computer Engineering and CS) were the highest-paid at our school.
You can find other info online, google something like "intern salary" or "engineering intern pay" or something like that. Of course, those will be interns who have a different education than you, so you might get very different numbers.
If you're thinking about trying to negotiate your salary, you might want to be cautious - often, intern and co-op positions are much more competitive than full-time work. The rule I was always told was that if you only get one offer, it's best to just accept it. If you get a few, and your favorite company wasn't the highest-paid, you might ask if they'll match your other offer. Of course, many companies pay all interns the same thing, so you might just get laughed at if you try to negotiate (or they might retract the offer).
Money is nice, but by far the most valuable thing you'll be getting is the experience. You might decide that you hate what you thought you wanted to do for the rest of your life. Or you might find that you want something a little bit different. Or maybe it'll be everything you've ever dreamed of and more. If you like what you're doing, the experience will help you find a real job (or more internships) later. If you don't, it'll save you from a big mistake down the road.
This is how 98% of all fortune 500 companies do this.
you're a nut if you allow a personal PC to connect to the company network.
On the other hand, I'm 98% more efficient on my personal PC than on my work PC, and the cost for my RSA token is 98% less than a laptop. I'm also 98% more likely to log in remotely from my personal PC than I would be from a laptop - it's easy to log on and keep up with E-mails during off-work hours (if I want to, anyway). If I had to drag out a different computer, there's no way I'd be doing it unless I was expecting something.
My previous employer's policy was to have antivirus installed, and disallow split tunneling (to avoid the remote computer becoming a conduit between the public internet and the internal network). My current employer wants antivirus, but allows split tunneling.
Here's the real question: whose time is more valuable, the people who will be using the network remotely, or the people who will be fixing it when one of the users does something bad? The less valuable the users' time is, the better case you can make for locking the VPN connection down to prevent anything from getting fouled up.
I think the basic idea is that the iPhone user will be using less bandwidth than someone with an unlimited data plan on their laptop (or, for that matter, a cell phone + tethering). Most people don't want to do all their internet on their cell phone.
Also of note, the iPhone 3G plan is $30/month, so you're only paying double.;-)
So your options are to 1. get an iPhone, 2. get AT&T DSL (which offers a similar benefit), or 3. pay per-use or per-month.
Of course, you'd think they'd want you to use wifi rather than the cell network for your data, since it probably costs them less for the WiFi.
Even drunk, I can get around any safeguard I set up for myself. I know because I have.
Obviously, you're not drinking enough. One of my past roommates was stumped by the screen on his bedroom window when we confined him there after drinking too much on Halloween.
Slashdot Mirror
Users should be able to PAY for a guaranteed minimum amount of bandwidth during times when the available bandwidth is being totally used. During times when network bandwidth is unused the spare bandwidth should be equally available to everybody.
Then perhaps you'd prefer to think of their policy this way: you can pay nothing (beyond the regular cost of dorm housing), and you always have a guaranteed minimum of 56 kbps.
My first year in the dorms (admittedly, it's been 8 years) the total off-campus bandwidth for the several thousand students was just 40 megabits. By the time these policies were started, it was significantly more (60 Mb, if not more), but still not a lot given the number of people that had to share the limited resources.
I think one of the problems with the idea of a guaranteed minimum is the cost; bandwidth can be expensive, which is why we all get crappy DSL lines for our homes instead of a dedicated T1. And keep in mind that the vast majority of people don't really use very much - so if you're a big internet user, you're best off if everybody else pays as much as possible for their crappy DSL, and you pay a little bit more than them for slightly less-crappy DSL. If you start offering people dirt-cheap plans with a very low guaranteed minimum bandwidth, then you're going to have to be willing to pay a lot more to get a guarantee that seems palatable to you. Very few people are actually willing to pay what it would cost to have very good service.
You say "crippled", I say "design tradeoffs".
I'm generally willing to trade multitasking for better battery life and apps that don't totally hose the phone. Or a memory card reader for a smaller, cheaper phone (memory cards? what are those anyway?). I couldn't really care less about apps. There's maybe a couple I'd get if I had an iPhone, but not many.
Damn it Jim, it's a phone, not a laptop.
One of my college roommates was responsible for the dorm networks; they definitely had policies that pissed people off (usually the people who were abusing the network the most), but it was done so that the limited resources were usable by everybody. Among them:
P2P traffic was capped at 50% of total bandwidth.
There was a rolling monthly bandwidth cap. Exceed it, and you were capped at 56k modem speeds for about a week until you were under the cap again. (On-campus traffic was not counted, and not limited; many large downloads such as linux distros were mirrored on-campus.)
If you picked up a virus, you were isolated from the network. The only thing you could get to was windowsupdate.com, until you removed the virus and called the helpdesk to promise you had an antivirus installed.
+1 for silly hardware bug workarounds.
I once encountered a processor bug that affected operation of the cache. As I recall, there were two workarounds:
1. Disable the cache.
2. Don't use the system with more than 112 MB of RAM.
Neither one of those is particularly easy to swallow on a shipping product.
As for the "3G router" - in some areas that would indeed be a market (some of Scandinavia has better wireless than wired connections, as does much of Africa) but it would very likely not be a new product, but more of an additional option for the Airport Extreme.
And Apple is SO well known for making products that are aimed at non-US markets...
</sarcasm>
Since Steve's second coming, I don't think they've made any significant regional adjustments to products other than localization, and legally required changes to be able to sell the product. Before that, there were a few products that were targeted at Japan (like the PowerBook 2400c).
Just to be clear, I'm not saying that 3G won't be added to some products. And I'm not complaining about the lack of international attention, since I've lived within walking distance of Apple HQ for the past 6 months... ;-) I'm just saying Apple will work in its mysterious ways - if they think there's a market, they'll sell it. Sometimes that works out well (iPod, iPhone), sometimes not (iPod Hi-Fi), and sometimes it's rather middling (Apple TV).
When I hear "story" and "video game" in the same sentence, I always think of Marathon. It didn't have anything fancy like cut scenes, or three dimensions... But it had an evolving plot. Beyond the "you're human, they're alien, go kill them before they kill you" that most FPSs use. It's certainly not the best, but for a game released in 1994, it was pretty unusual.
I have no idea how much I've saved by not owning a car, but it's a lot. I made a conscious decision to live with only my bicycle for probably at least my first year post-college - after all, I've been biking everywhere for my last 5 years of college.
Eventually, when I have a family to share it with, I'll probably buy a car. But for now, I'm enjoying not having one: I'm losing weight; plus I can max out my IRA, 401k, and stock purchase plans, plus pay for my relatively expensive housing, and still have plenty of money left over.
It doesn't rain much, but when it does, I either wait for it to pass, or just bike through it. If it's heavy enough and I'm going far enough that I'd get soaked, I bring an extra set of clothes along. I go grocery shopping every week or two, and can carry all of it back in a large messenger bag, plus one pannier on the bike. (bonus: I never need plastic bags.)
My total bike maintenance comes to probably $200 per year. A couple times, I've rented a car (like when friends come to visit for a while), but I could probably rent a car for a week out of every month, and still come out ahead. Many communities have car sharing programs available, which can be quite useful for those who can almost live without a car.
I'm not sure I'd call it "poorly engineered." Differently engineered, certainly - most mice don't require you to lift your finger off the left side of the mouse to register a right-click. It's an elegant solution to the problem of balancing the power users' desire to have two mouse buttons with the simplicity of a one-button design for new users.
It's certainly not the mouse for everybody - especially anybody who plays first-person shooters. However, I've never used the right button for gaming, and I sometimes use only the keyboard.
I did a small project a few years ago with multi-core and multi-processor execution. The workload I used was a variety of parallel sorting algorithms. The fastest machines I had access to at the time were a dual-processor dual-core machine, and a dual-processor quad-core. That's 4 or 8 cores total, with two or four L2 caches (one for each pair of cores). I got close to perfect speedups when running on 2 cores; about 1.95x to 2x the speed of running on a single core. Going to 4 cores didn't give nearly as much speedup (about 3 to 3.5x the speed of a single core running), but that's mainly due to the data sharing between threads; since there were separate L2 caches, there's a significant latency hit any time data is moved between cores that don't share the cache.
Now by increasing the size of the dataset, the amount of work to be done increases faster than the amount of communication, so the speedup increases (that's why I gave a range of 3 to 3.5x). My datasets weren't huge - maybe 150 MB or so, I think, small enough that they could finish in under a second. Make the amount of data huge, and the speedup will get even closer to 4x on 4 cores. (Of course, running on all 8 cores of the 8-core machine was even less ideal - a speedup of 4 to 5x. But that incurs the extra penalty of storing data back to RAM when it has to move data between the different chips.)
More modern machines address these communication problems better (for example, by connecting the processors to each other, in addition to connecting them through the northbridge; or by having cache shared by all cores on a single chip), and will do even better.
The main problem, as you mention, is really software. However, you don't need parallel software to take advantage of many cores - running different programs helps as well. I wish I had more than 2 cores on my 4-year-old desktop, as running a video chat (with all the inherent encoding and decoding) at the same time as various other tasks (say browsing on a flash-heavy site) can be quite slow. On the other hand, now that dual-core machines are fairly mainstream, more applications will take advantage of multiple cores. And I can't think of a whole lot of applications that are both CPU-bound AND are inherently limited to 2x parallelism. Progress just keeps marching along...
On the other hand, the high-end product lines make up a significant portion of the profits for chip companies. They charge a huge premium for perfect chips that run at high speeds and low (or relatively low) power and temperature. The chips that aren't as perfect get sold for much less to the masses, running at lower clocks, lower voltages, and even with features disabled (as in the case where a chip has a defect).
Because of this, chipmakers will probably continue to have at least one product for the high-end market, and the benefits will continue to trickle down to the rest of us.
Sure, and let us know when you buy a real computer.
<grin>
-another OmniWeb user
This is absolutely a correct sentiment. That's one reason why a hardware memory tester will do far better than software - it can run the memory at the high end of the voltage spec, or the low end, or bounce back and forth between the two. It can also do operations designed to heat the chips up, as the chips might slow down when hot. And you don't have to worry about running your OS in part of the RAM, either.
I agree about 110% with this. I used to do computer repair, and we had both a hardware memory tester, as well as various software programs that could test RAM. In my experience, the software testers could pick out about 5% of memory that was bad, the hardware tester could pick out about 95% of bad memory, and old-fashioned "swap it with known good memory" troubleshooting was required to find the last 5%. The software testers only pick out the memory that is very, very bad. I probably had better luck picking out bad RAM by looking at the manufacturer stamp on the chip - there are a number of manufacturers that I learned to trust (like Micron and Samsung). On the other hand, when I'd see a chip that had nothing on it but a label saying "512Mbx8" or something like that, it was almost guaranteed to be bad.
Lesson for the day: you get what you pay for. Don't buy cheap RAM, or your computer will crash incessantly.
That's the silliest claim I've ever heard. If they're not tracking each card individually, that would cause at least two major problems:
When a card is redeemed, how do they know it hasn't been redeemed before?
When a card is manufactured, how do they know it hasn't been manufactured before?
Now that's not to say that they might have difficulty identifying an illegitimate card. Especially if it's a fake that has been manufactured, and has not yet been redeemed. Or maybe their manufacturing folks don't talk with the iTunes folks, and they don't know whether a card has been manufactured - just whether it has been redeemed.
That would be a sensible aim if the iPhone was the market leader.
Now, show us some reference where the iPhone is shown to be leading the market.
According to some research from about 2 months ago, Apple has 17.3% of the smartphone market share, behind only Nokia at 38.9%. Going from 0% to 17.3% in 18 months is pretty impressive. Not only that, but Apple sells one model of iPhone (ignoring capacity and color differences). Nokia has many smartphones. I would not be surprised if the iPhone beats every particular model of Nokia's.
A few years from now, when Apple is selling the iPhone nano, iPhone classic, and several other models, they might well be crushing Nokia the same way they crushed everybody in the MP3 player market. (Or they might lose one for a change, you never know.)
Personally, I'm waiting for the iPhone Shuffle. Nothing more than an earpiece, a microphone, and a button to call one of your friends at random.
I think graphics chips are moving in the direction of saving power like CPUs have tended to do. For example, the MacBook Pro, which will run either the discrete GPU or the integrated one, depending on whether you prefer performance or battery life. Laptops will definitely go in this direction, because it's easier to reduce your power consumption than to increase the capacity of the battery without making it larger and heavier.
Desktops graphics cards will get better at power savings too, but probably not as much, because it's something most people seem to care about - energy is readily available, cooling isn't too difficult, and people buy computers based on initial price and performance, not long-term costs like energy usage.
That's only low until you have to design something like that. Or design its power supply, PCB layout, or heatsink. I designed a board once that had a 25W chip on it, and that had plenty of challenges. Of course, we had some extra restrictions that most systems don't have - extraordinarily limited space for a heatsink, another nearby heat source, very large ambient temperature range, and no fan allowed.
The highest-power Pentium 4 was 115 Watts, which is quite frankly ridiculous. From what I can tell, the Prescott had a 122 mm^2 die; that puts it at a power density of 94 W/cm^2. The E8400's 107 mm^2 die comes in at about 61 W/cm^2. A slide I've seen in many places, originally attributed to Intel, pegs a nuclear reactor at 200 W/cm^2.
With the P4, Intel realized that you eventually reach a point where it's not realistic to cool the CPU. Since then, basically everybody has started towards finding more ways to save power. After all, about half of those 65 Watts are from transistor leakage - the electricity is just wasted, not actually doing any work. As process technologies shrink, the leakage power increases. So, to be able to continue to cool our CPUs, they'll have to figure out ways to do more work with less energy.
Well, you might not be able to find the VTO model as-is, but it's a nice DIY project for a hobbyist.
Actually, we picked a few bacteria off an existing culture (the "big fuzzy mounds"), then spread them on the new dish. I forget what sort of tool we used, but they were at least spread thinly enough that you couldn't see the colonies with the naked eye.
In any case, you're certainly right - it was as rigorous as an 8th grade science class sounds. But still, when I'm backpacking and drinking water from a random stream somewhere, I'll be using the appropriate filter, not exposing the water to a little extra sunlight. ;-)
A Reliant Robin, and you should be able to find one in England somewhere.
Don't forget the sanitizing UV light...
You know, I've never been terribly confident in those since an 8th grade science experiment. I think the point was to try and come up with something that might cause a mutation in some bacteria - so we grew some in a petri dish, picked one colony (to get all the same type of bacteria), then grew it in a petri dish, then picked some out of that uniform batch and put them in another petri dish. We covered half with aluminum foil, then put it in the UV hood designed for disinfecting lab goggles. The bacteria had no problems growing after being supposedly "killed" by the UV (not to mention, no mutations like we wanted).
Of course, that was still an interesting result that our science teacher liked - so we spread more bacteria on a petri dish, put it uncovered into the UV hood (in case the plastic on the petri dish was opaque to UV or something like that), and ran it for much longer than normal. The bacteria still had no problems growing.
Now maybe it's just difficult to kill bacteria when you've put them on top of a nice big pile of food (aka agar)... But I really don't expect UV to kill anything. That doesn't mean I wouldn't drink the water - it just means that the UV shouldn't be the only line of defense if I think there's really a risk of getting sick (which there probably isn't a big one anyway...).
Agreed, talk to your school. Personally, my internships were $16-$18/hour. I got offers as high as $24/hour. It'll vary widely by major - mine (Computer Engineering and CS) were the highest-paid at our school.
You can find other info online, google something like "intern salary" or "engineering intern pay" or something like that. Of course, those will be interns who have a different education than you, so you might get very different numbers.
If you're thinking about trying to negotiate your salary, you might want to be cautious - often, intern and co-op positions are much more competitive than full-time work. The rule I was always told was that if you only get one offer, it's best to just accept it. If you get a few, and your favorite company wasn't the highest-paid, you might ask if they'll match your other offer. Of course, many companies pay all interns the same thing, so you might just get laughed at if you try to negotiate (or they might retract the offer).
Money is nice, but by far the most valuable thing you'll be getting is the experience. You might decide that you hate what you thought you wanted to do for the rest of your life. Or you might find that you want something a little bit different. Or maybe it'll be everything you've ever dreamed of and more. If you like what you're doing, the experience will help you find a real job (or more internships) later. If you don't, it'll save you from a big mistake down the road.
This is how 98% of all fortune 500 companies do this.
you're a nut if you allow a personal PC to connect to the company network.
On the other hand, I'm 98% more efficient on my personal PC than on my work PC, and the cost for my RSA token is 98% less than a laptop. I'm also 98% more likely to log in remotely from my personal PC than I would be from a laptop - it's easy to log on and keep up with E-mails during off-work hours (if I want to, anyway). If I had to drag out a different computer, there's no way I'd be doing it unless I was expecting something.
My previous employer's policy was to have antivirus installed, and disallow split tunneling (to avoid the remote computer becoming a conduit between the public internet and the internal network). My current employer wants antivirus, but allows split tunneling.
Here's the real question: whose time is more valuable, the people who will be using the network remotely, or the people who will be fixing it when one of the users does something bad? The less valuable the users' time is, the better case you can make for locking the VPN connection down to prevent anything from getting fouled up.
I think the basic idea is that the iPhone user will be using less bandwidth than someone with an unlimited data plan on their laptop (or, for that matter, a cell phone + tethering). Most people don't want to do all their internet on their cell phone.
Also of note, the iPhone 3G plan is $30/month, so you're only paying double. ;-)
So your options are to 1. get an iPhone, 2. get AT&T DSL (which offers a similar benefit), or 3. pay per-use or per-month.
Of course, you'd think they'd want you to use wifi rather than the cell network for your data, since it probably costs them less for the WiFi.
Even drunk, I can get around any safeguard I set up for myself. I know because I have.
Obviously, you're not drinking enough. One of my past roommates was stumped by the screen on his bedroom window when we confined him there after drinking too much on Halloween.