"I don't care how it works, I just want it to work!" is the rallying cry of most home desktop users.
Well here I go pissing off another bunch of people with this response:-) It's also my rallying cry, and FWIW, I have degrees in both Electrical and Software Engineering and enough experience doing both that I think my opinions may have some validity. (Knowing how anti-credential some/.ers can be, you may disagree)
Even those of us who love working with computers, often just want them to work as advertised without tinkering. I am not averse to learning new things, but there are some things I shouldn't need to know. I don't know, or care to, how to milk a cow; but I know where to get a gallon of 1% when I need it. Likewise, I don't want to have to remember to issue "rmmod STV680" every time I want Gnome/KDE(I forget which)'s digital camera app to download pictures from my camera. Burning a CD should be a trivial operation with a GUI, not one in which I *have* to remember to do things in just the right sequence or it fails without bothering to even give an error message.
The whole *point* of computers in modern life is to make things easier. Saying that users just need to learn more arcane bits of knowledge (come on, mounting a USB device just to get at some pictures? Are you serious!) just to do trivial tasks completely misses that point.
At some point you have to say "yep, it's easy enough" and move on,
Yes, but the problem with so many Linux apps is that the author said "yep, it's easy enough" waaay too early in the process. I occasionally burn CDs on my Linux box (my primary home machine, BTW) and sometimes it locks the CD in the drive. Doesn't tell me why I can't have my CD back, just gave me a cryptic error and then refuses to work. A few weeks ago when I was investigating how to make VCDs, I learned that one of the parameters to cdrecord (don't flame me if I got the app's name wrong!) is to release the CD if something fails. Why on earth the writer of KonCD or whatever it is I was using can't set this parameter or check error codes and release the CD I have no idea, but things like this irritate me to no end. I'm sure the app was quite satisfactory to the guy who wrote it, and I can't really complain to him. After all, it was free and he's under no obligation to support it, but it reflects poorly on his programming skill to release something so obviously flawed to the public, even if it is Open Source. I've come to the conclusion that I would happily pay for stable, predictable GUI Linux apps that didn't require that I get inside the developer's head just to figure out how to do the most basic tasks. Command line apps generally work quite well; it's just that I find so many of the GUI wrappers suck ass that they're almost worthless.
I really don't want to have to write scripts for every little task that RH8/9/Fedora claims to have a GUI app for just because I can't trust the app to do what it says it can. At that point I may as well switch back to Windows!
Double-click the first image, and it almost automatically starts up a slide show
Now this is how user software should work. Yes, it takes a hell of a lot of work to do a good, robust user interface, but it's what users expect. Give them any less and they think your Linux box is just a primitive, expensive toy.
I don't fathom how people state that they own their homes
Same way they say they own their cars while still making payments. Can you sell it? Yes. Can you make a profit on the sale and keep it? Yes. Sounds to me like you own it!
Someone who would qualify for about a 60K annual income now in the US... here in Bangalore, for example, you can rent a beautiful 3 bedroom apartment
And here in Minnesota in the US that $60k would have you owning your own house on 1/2 to 5 acres of land (depending on how far from work you wanted to live; assuming a workplace in the city), still able to afford a good car, and live well. Seeing the world and living in different places is great, but any American SW engineer who thinks that in order to have a good standard of living they have to move abroad is either insane or stupid.
Not all tech jobs are in the Bay Area where $60k is considered an entry-level salary.
if something goes wrong and I can't fix it, the wose that could happen is that I would only lose my job.
I understand that sentiment, but look at it from a different perspective. I design software for medical instruments. I like my job. I know that if I screw up majorly (actually it would probably take me, the SQA dept and my code reviewers to all screw up), I could cause someone's death. But I don't obsess over it. I try to write good, solid, robust code. When I see defects (even if they're my bugs), I enter problem reports so they will be fixed. If we see process failures, we try to fix the process. And that process emphasizes fixing things, not blaming people! Instead of worrying that I might kill someone (which is unlikely, as we have an enormous number of safeguards designed into the software to detect problems along the way) I take pleasure in knowing that I'm making people's lives better.
We are always trying to improve our software quality. And that's really the key: a group of people -- developers, QA and management alike -- trying to build a good product that could be used on our loved ones if need be.
I'm not 100% sure (no time to check the FDA site), but I believe this phrase is the important one: it means the software would have been classified as a medical device and thus subject to FDA regulation, including requiring FDA approval and following a specific development process. Yes, I know it's not a physical device, but the FDA considers certain software "medical devices" depending on what data they provide and going from memory, I think this would be classified as such. Good thing you left, you could have been shipping illegal product.
Thing is, I can never get the basmati rice to come out right. Anyone know how to cook it persian style and wanna give me a hint?
My standard method for cooking jasmine rice (tip: get a 25lb bag for $9 at an Indian food store. Don't buy rice at grocery store). For 1 cup rice: Wash rice well bring 1 cup rice and 2 cups water to a boil (salt to taste.. 1/2 tsp or so?) When boiling, lower to very low simmer and cover. Remove from heat after 11 minutes. No need to drain or re-wash as all water is absorbed. Perfect rice
You can get fluffier rice if you toast it before cooking, but watch that it doesn't burn. This also removes the need to wash it. It also works well if you stir-fry uncooked rice, chopped onions and a bit of garlic and chile pepper for a few minutes then cook as described above.
The idea that cats are scared of dogs is overrated
To say the least! Our 5lb (or thereabouts) cat used to kick the crap out of our 80+ lb dog until the dog wised up and realized she was much larger than the cat and could fight back. Now I regularly come home to find them curled up together sleeping (cat "spooned" inside the dog).
but if the programming in CS is too tough for this guy, then EE definitely will
Too bad your comment is rated so low as it's probably true. The majority of people in my freshman year EE classes switched majors to Computer Science after encountering Physics and Math 101 -- I remember rooms going from standing room only first semester to mostly empty seats second semester! And a lot of those CS converts switched to Business after taking their first CS Math courses. Find what you really like to do. Life's too short to spend it hating your job(s) constantly.
Because some bean-counter decided that you didn't need an extra test system, or the time to thoroughly debug it BEFORE you put it into production.
I agree with you up to a point: I've argued myself blue in the face that we developers should have more systems for testing on (which is expensive, as our machines cost us close to $100k each to build), but sometimes you have to use the production system (but I still think you have to be patient).
Case in point: we received a customer complaint that should have been simple to reproduce. However, none of our in-house test systems showed that problem. No matter how creative we were, the machines simply shut down gracefully as they were designed to in that situation. The issue is compounded by the fact that this is not just an off the shelf PC, but an embedded system and we have multiple, slightly different versions of hardware in the test lab. So we go to the manufacturing dept and get a systems on the production floor identical in mechanical hardware, electronics and software to the customer's system. Same result: system works perfectly. Call the customer, have them create the situation on their machine that causes a problem while we are connected to it remotely: instant crash!
We did identify a problem and fixed it, but to this day I've never heard of that problem on any other system. It turned out to be a race condition that required a thread being interrupted within a window measured in single-digit microseconds at most, leading to data corruption. Perhaps there was some tiny timing variation on that system's CPU that caused it to fail only on that system, we'll never know.
Sometimes you don't have a choice but to debug on the only system that exhibits the problem.
I'm curious: do you live in the US? If so, do you find that the projects' max bids are reasonable? I ask this because I notice that a lot of the bidders are from India or Eastern Europe and the maximums that buyers are willing to pay are for the most part...insane!
I tried it once and almost won the bid (I was high bidder, but the buyer was comfortable with my experience and knowledge of the field), then the requirements changed, and my new bid put me out of the running (still better than the alternative of me losing my shirt on a fixed-price bid, tho!)
I can see meeting the max bid amounts if the projects are almost copycats of things you've done before, but for just about any new development work, I'd go over the maximum just on documentation costs. RentACoder bidders do document their work, don't they?
Worse than that: it's commercialism with the implied hint that if you don't participate, there's something wrong with you. For the record, I got my partner a box of handmade truffles (partly cause I like chocolate and I knew I'd eat some), but we opened them 3 days early. If I hadn't heard of those chocs before, odds are she wouldn't have gotten anything for Valentine's. It's not my style to think "hey, it's XXXX day, I'd better do what's expected of me." I think most people need a trigger to send presents, do nice gestures, etc. A few weeks ago I sent her a dozen roses at work, just because. Her co workers couldn't get it into their heads that I gave them to her just because I felt like it and insisted that I must be trying to apologize for something???
People are strange.
Oh, and if you want to be with someone on Valentine's day next year, post a personal ad. Bet dollars to donuts (what the hell does that mean anyway?) that if you placed a upbeat online ads that you weren't attached and were looking only for someone to hangout with on V'day, you'd get a host of replies from MOTAS. There are lots of people alone out there and the net makes it a lot easier for them to hook up...that's how I met my babe, anyway.
Re:doesn't have to be isolated or small...
on
RFID Tags For The Rich
·
· Score: 4, Insightful
All of this just goes to show that if you want to be successful, it's all about establishing a relationship with the customer, and that's the job of the sales person
Exactly true. Summers in college I worked on the floor at a high-end men's specialty clothing store in NYC taking customers' clothes and credit cards from the salesmen to the cash registers hidden in the back (customers' eyes were not to be sullied by the sight of a plebian cash register, I guess!) and bringing them back out packaged and ready. Even though as temps we were just one step up from the minimum wage employees that did a similar job, we were expected to dress in a suit and tie and had to go through a "training session" which basically consisted of "the customer is always right" and emphasized that they stressed customer service. The kind of people who drop $5,000 on a suit of clothes without a second thought expect that kind of service and usually get it. It means lots of repeat business.
Same thing happens now. At work, a small group of us used to go to lunch to the same two restaurants 4-5 times a week. It got to the point where the owner or waitresses would see us come in the door and usher us to our favorite table with the condiments and free appetizers we liked ready and waiting. They liked the repeat business (sometimes we brought in large groups) and did their best to please us; we liked the service so we tipped very well and kept coming back. Works both ways.
Believe me, a good quality tester is a valuable asset
Second that! I may want to swear at some of our testers when they send my "fixed" bugs right back at me, but they do find important defects (and just as important, they will document the steps & environment necessary to duplicate them). We the developers suggest test cases, but the better testers go well beyond that and design their own test cases that really stress the system.
It's not a job I would want (did enough hardware testing early in my career), but I respect those who are good at it.
I have a Model 100 here -- it's fairly well constructed, but it's still a hunk of consumer plastic. I'm pretty sure I could destroy it by throwing on the floor a few times.
Maybe you are talking about some industrial model that I'm not aware of.
No, he's probably talking about the same one you have. I had one (inherited from my first job when we threw a bunch away) and even though made from cheap looking plastic, it's pretty rugged. We used them as test bench machines and they got banged around, dropped on the floor, poked with soldering irons and worked just fine. Until a few years ago, mine was used to control the AC and heat in my apartment. I tossed it when I moved.
Because it's true. Haven't heard it in the software world, but back when I was an EE, managers had a phrase "time to shoot the engineers and ship the product." Left to their own devices, development engineers will try to make something perfect. QA engineers have to understand the concept of "good enough," which is when the product meets the required level of functionality and quality.
It's nice to talk about perfection, but reality is that optimizing every aspect of the code means it will never ship.
Just how much content can you have being specific about Embedded Systems Programming. Seriously, I'm asking.
Well, having read ESP for at least 10 years, perhaps I can answer.
A lot!! The amount of Desktop CPUs sold is a drop in the bucket compared to the number of embedded CPUs. Look around the average house and compare the number of "computer controlled" items versus the number of desktop (i.e, Windows, Linux, Mac) computers. Just in my living room alone I can think of the thermostat, X10 lighting controllers, VCR, DVD, CD, TV. Now think of all the various types of hardware those CPUs need to interface to and the varying levels of system cost and you begin to see that embedded design spans a huge range of complexity. The skillset of a versatile developer needs to encompass issues ranging from electrical engineering to object-oriented design, to low-level systems programming, to digital signal processing and sometimes even mechanical engineering (e.g., the code shut off the pump, but inertia keeps the liquid moving for a while), depending on what the product is. Remember, we're usually developing code for hardware that isn't even debugged yet!
My own career has spanned the range from medical instruments running on Pentium processors with over 1MLOC to power controllers running on tiny sub-1k (that's kilobyte) processors with 100 lines of assembly code. In one case to e.g., read temperatures, I may have a custom designed hardware interface that cost over $500, in the other, I have to understand and use the electrical properties of a resistor, capacitor and diode (total cost about $.05) to do something similar in order to meet target costs.
Yes, it's worth having an entire magazine devoted to the subject, even as that magazine gets thinner because more and more of its content is moving onto their website. In fact, I wish there were more like it. I have years of backissues because there is so much valuable information I would hate to lose.
For me, its the psychological aspect that gets me. I "can't" just pick up and go. I "can't" tell the boss where he can stick this job.
I used to worry about this also. But what finally made me decide to buy a house (now on my second, having made a tidy profit on the first in just 5 years) was: 1) I own a nice subwoofer. Subwoofers are not apartment friendly:-( 2) I like growing vegetables and happily dig up large amounts of lawn to turn into garden space. 3) I can mod it however the hell I like (well within reason, but you get the idea). If I decide to run Cat6 to each room and put a WAP in a tree in the backyard, I don't have to clear it with anyone.
For me, the advantages far outweighed the disadvantages. Obviously YMMV.
Airplanes can cause loss of signal. I used to live on one of the departure/approach paths at MSP and occasionally the signal was lost for a few seconds when an airplane was climbing out more or less in the beam path. But I think it's more likely that the receiver was being saturated by transmissions from the plane than the aircraft itself blocking the signal.
One advantage over cable is that when you move you can have service immediately if you transfer the dish yourself, unlike cable where it may need to be "turned on" at the new location. But in our case, we used DishMover since it was already snowing as we were moving and I really didn't feel like climbing up on the icy roof to install the antenna -- the install tech took the time to do a really good job, better than I would have done with subzero windchills around me.
Get the degree that interests you; they're about equal in marketability. Take the "sky is falling" moaning around here with a grain of salt. Yes, jobs are hard to find in technical fields right now, especially for entry-level personnel. But this happens every few years, though it has been particularly bad for a while. It will pick up: you can't push technology away, we'll always need skilled engineers.
But I would caution you to aim whatever degree you get at a specialty. I'd look at a biology or bioinformatics minor if possible (but being in the medical field, I may be biased). The "ordinary business coder" work is going the way of the dodo. Likewise, the kind of EE work I started my career doing can be done by a motivated hobbyist these days. The fact is that as time progresses, technology becomes available to a lower and lower skill level, that's really why so many of these jobs are being sent overseas. The overseas flow will stop when it's cheaper to automate the work than to farm it out. But by then, the smart people will have moved on to a different specialty.
Now if you want a different kind of advice on job finding;-) Learn to be a great communicator: take writing classes and speech classes if you can and try to be socially active with non-technical people (something I really wish I had done as an undergraduate!). Most people you will meet will have no interest in technology, but you need to be able to talk to them effectively. Make the HR person previewing you feel like you're a nice, personable guy, and you're 75% of the way to getting that job!
Slashdot Mirror
Well here I go pissing off another bunch of people with this response
It's also my rallying cry, and FWIW, I have degrees in both Electrical and Software Engineering and enough experience doing both that I think my opinions may have some validity. (Knowing how anti-credential some
Even those of us who love working with computers, often just want them to work as advertised without tinkering. I am not averse to learning new things, but there are some things I shouldn't need to know. I don't know, or care to, how to milk a cow; but I know where to get a gallon of 1% when I need it. Likewise, I don't want to have to remember to issue "rmmod STV680" every time I want Gnome/KDE(I forget which)'s digital camera app to download pictures from my camera. Burning a CD should be a trivial operation with a GUI, not one in which I *have* to remember to do things in just the right sequence or it fails without bothering to even give an error message.
The whole *point* of computers in modern life is to make things easier. Saying that users just need to learn more arcane bits of knowledge (come on, mounting a USB device just to get at some pictures? Are you serious!) just to do trivial tasks completely misses that point.
Yes, but the problem with so many Linux apps is that the author said "yep, it's easy enough" waaay too early in the process. I occasionally burn CDs on my Linux box (my primary home machine, BTW) and sometimes it locks the CD in the drive. Doesn't tell me why I can't have my CD back, just gave me a cryptic error and then refuses to work. A few weeks ago when I was investigating how to make VCDs, I learned that one of the parameters to cdrecord (don't flame me if I got the app's name wrong!) is to release the CD if something fails. Why on earth the writer of KonCD or whatever it is I was using can't set this parameter or check error codes and release the CD I have no idea, but things like this irritate me to no end. I'm sure the app was quite satisfactory to the guy who wrote it, and I can't really complain to him. After all, it was free and he's under no obligation to support it, but it reflects poorly on his programming skill to release something so obviously flawed to the public, even if it is Open Source.
I've come to the conclusion that I would happily pay for stable, predictable GUI Linux apps that didn't require that I get inside the developer's head just to figure out how to do the most basic tasks. Command line apps generally work quite well; it's just that I find so many of the GUI wrappers suck ass that they're almost worthless.
I really don't want to have to write scripts for every little task that RH8/9/Fedora claims to have a GUI app for just because I can't trust the app to do what it says it can. At that point I may as well switch back to Windows!
Now this is how user software should work. Yes, it takes a hell of a lot of work to do a good, robust user interface, but it's what users expect. Give them any less and they think your Linux box is just a primitive, expensive toy.
Same way they say they own their cars while still making payments.
Can you sell it? Yes. Can you make a profit on the sale and keep it? Yes. Sounds to me like you own it!
And here in Minnesota in the US that $60k would have you owning your own house on 1/2 to 5 acres of land (depending on how far from work you wanted to live; assuming a workplace in the city), still able to afford a good car, and live well.
Seeing the world and living in different places is great, but any American SW engineer who thinks that in order to have a good standard of living they have to move abroad is either insane or stupid.
Not all tech jobs are in the Bay Area where $60k is considered an entry-level salary.
I understand that sentiment, but look at it from a different perspective. I design software for medical instruments. I like my job. I know that if I screw up majorly (actually it would probably take me, the SQA dept and my code reviewers to all screw up), I could cause someone's death. But I don't obsess over it. I try to write good, solid, robust code. When I see defects (even if they're my bugs), I enter problem reports so they will be fixed. If we see process failures, we try to fix the process. And that process emphasizes fixing things, not blaming people!
Instead of worrying that I might kill someone (which is unlikely, as we have an enormous number of safeguards designed into the software to detect problems along the way) I take pleasure in knowing that I'm making people's lives better.
We are always trying to improve our software quality. And that's really the key: a group of people -- developers, QA and management alike -- trying to build a good product that could be used on our loved ones if need be.
I'm not 100% sure (no time to check the FDA site), but I believe this phrase is the important one: it means the software would have been classified as a medical device and thus subject to FDA regulation, including requiring FDA approval and following a specific development process. Yes, I know it's not a physical device, but the FDA considers certain software "medical devices" depending on what data they provide and going from memory, I think this would be classified as such. Good thing you left, you could have been shipping illegal product.
Having just finished Bancroft/Arnesen's No Horizon is so Far, I'm impressed it made it. The Antarctic ice is anything but flat for large areas.
My standard method for cooking jasmine rice (tip: get a 25lb bag for $9 at an Indian food store. Don't buy rice at grocery store).
For 1 cup rice:
Wash rice well
bring 1 cup rice and 2 cups water to a boil (salt to taste
When boiling, lower to very low simmer and cover.
Remove from heat after 11 minutes. No need to drain or re-wash as all water is absorbed.
Perfect rice
You can get fluffier rice if you toast it before cooking, but watch that it doesn't burn. This also removes the need to wash it. It also works well if you stir-fry uncooked rice, chopped onions and a bit of garlic and chile pepper for a few minutes then cook as described above.
Having paid a plumber $350 for less than one hour's work, I really wish that "irrational exhuberance "[sic] was just a software-only thing.
To say the least! Our 5lb (or thereabouts) cat used to kick the crap out of our 80+ lb dog until the dog wised up and realized she was much larger than the cat and could fight back. Now I regularly come home to find them curled up together sleeping (cat "spooned" inside the dog).
Too bad your comment is rated so low as it's probably true. The majority of people in my freshman year EE classes switched majors to Computer Science after encountering Physics and Math 101 -- I remember rooms going from standing room only first semester to mostly empty seats second semester! And a lot of those CS converts switched to Business after taking their first CS Math courses. Find what you really like to do. Life's too short to spend it hating your job(s) constantly.
I agree with you up to a point: I've argued myself blue in the face that we developers should have more systems for testing on (which is expensive, as our machines cost us close to $100k each to build), but sometimes you have to use the production system (but I still think you have to be patient).
Case in point: we received a customer complaint that should have been simple to reproduce. However, none of our in-house test systems showed that problem. No matter how creative we were, the machines simply shut down gracefully as they were designed to in that situation. The issue is compounded by the fact that this is not just an off the shelf PC, but an embedded system and we have multiple, slightly different versions of hardware in the test lab. So we go to the manufacturing dept and get a systems on the production floor identical in mechanical hardware, electronics and software to the customer's system. Same result: system works perfectly. Call the customer, have them create the situation on their machine that causes a problem while we are connected to it remotely: instant crash!
We did identify a problem and fixed it, but to this day I've never heard of that problem on any other system. It turned out to be a race condition that required a thread being interrupted within a window measured in single-digit microseconds at most, leading to data corruption. Perhaps there was some tiny timing variation on that system's CPU that caused it to fail only on that system, we'll never know.
Sometimes you don't have a choice but to debug on the only system that exhibits the problem.
I'm curious: do you live in the US? If so, do you find that the projects' max bids are reasonable? I ask this because I notice that a lot of the bidders are from India or Eastern Europe and the maximums that buyers are willing to pay are for the most part...insane!
I tried it once and almost won the bid (I was high bidder, but the buyer was comfortable with my experience and knowledge of the field), then the requirements changed, and my new bid put me out of the running (still better than the alternative of me losing my shirt on a fixed-price bid, tho!)
I can see meeting the max bid amounts if the projects are almost copycats of things you've done before, but for just about any new development work, I'd go over the maximum just on documentation costs. RentACoder bidders do document their work, don't they?
Worse than that: it's commercialism with the implied hint that if you don't participate, there's something wrong with you. For the record, I got my partner a box of handmade truffles (partly cause I like chocolate and I knew I'd eat some), but we opened them 3 days early. If I hadn't heard of those chocs before, odds are she wouldn't have gotten anything for Valentine's. It's not my style to think "hey, it's XXXX day, I'd better do what's expected of me." I think most people need a trigger to send presents, do nice gestures, etc. A few weeks ago I sent her a dozen roses at work, just because. Her co workers couldn't get it into their heads that I gave them to her just because I felt like it and insisted that I must be trying to apologize for something???
People are strange.
Oh, and if you want to be with someone on Valentine's day next year, post a personal ad. Bet dollars to donuts (what the hell does that mean anyway?) that if you placed a upbeat online ads that you weren't attached and were looking only for someone to hangout with on V'day, you'd get a host of replies from MOTAS. There are lots of people alone out there and the net makes it a lot easier for them to hook up...that's how I met my babe, anyway.
Exactly true. Summers in college I worked on the floor at a high-end men's specialty clothing store in NYC taking customers' clothes and credit cards from the salesmen to the cash registers hidden in the back (customers' eyes were not to be sullied by the sight of a plebian cash register, I guess!) and bringing them back out packaged and ready. Even though as temps we were just one step up from the minimum wage employees that did a similar job, we were expected to dress in a suit and tie and had to go through a "training session" which basically consisted of "the customer is always right" and emphasized that they stressed customer service. The kind of people who drop $5,000 on a suit of clothes without a second thought expect that kind of service and usually get it. It means lots of repeat business.
Same thing happens now. At work, a small group of us used to go to lunch to the same two restaurants 4-5 times a week. It got to the point where the owner or waitresses would see us come in the door and usher us to our favorite table with the condiments and free appetizers we liked ready and waiting. They liked the repeat business (sometimes we brought in large groups) and did their best to please us; we liked the service so we tipped very well and kept coming back. Works both ways.
Second that! I may want to swear at some of our testers when they send my "fixed" bugs right back at me, but they do find important defects (and just as important, they will document the steps & environment necessary to duplicate them). We the developers suggest test cases, but the better testers go well beyond that and design their own test cases that really stress the system.
It's not a job I would want (did enough hardware testing early in my career), but I respect those who are good at it.
No, he's probably talking about the same one you have. I had one (inherited from my first job when we threw a bunch away) and even though made from cheap looking plastic, it's pretty rugged. We used them as test bench machines and they got banged around, dropped on the floor, poked with soldering irons and worked just fine. Until a few years ago, mine was used to control the AC and heat in my apartment. I tossed it when I moved.
By who? Space Ghost?
Because it's true. Haven't heard it in the software world, but back when I was an EE, managers had a phrase "time to shoot the engineers and ship the product." Left to their own devices, development engineers will try to make something perfect. QA engineers have to understand the concept of "good enough," which is when the product meets the required level of functionality and quality.
It's nice to talk about perfection, but reality is that optimizing every aspect of the code means it will never ship.
Yup. There'd be no software.
Well, having read ESP for at least 10 years, perhaps I can answer.
A lot!! The amount of Desktop CPUs sold is a drop in the bucket compared to the number of embedded CPUs. Look around the average house and compare the number of "computer controlled" items versus the number of desktop (i.e, Windows, Linux, Mac) computers. Just in my living room alone I can think of the thermostat, X10 lighting controllers, VCR, DVD, CD, TV. Now think of all the various types of hardware those CPUs need to interface to and the varying levels of system cost and you begin to see that embedded design spans a huge range of complexity. The skillset of a versatile developer needs to encompass issues ranging from electrical engineering to object-oriented design, to low-level systems programming, to digital signal processing and sometimes even mechanical engineering (e.g., the code shut off the pump, but inertia keeps the liquid moving for a while), depending on what the product is. Remember, we're usually developing code for hardware that isn't even debugged yet!
My own career has spanned the range from medical instruments running on Pentium processors with over 1MLOC to power controllers running on tiny sub-1k (that's kilobyte) processors with 100 lines of assembly code. In one case to e.g., read temperatures, I may have a custom designed hardware interface that cost over $500, in the other, I have to understand and use the electrical properties of a resistor, capacitor and diode (total cost about $.05) to do something similar in order to meet target costs.
Yes, it's worth having an entire magazine devoted to the subject, even as that magazine gets thinner because more and more of its content is moving onto their website. In fact, I wish there were more like it. I have years of backissues because there is so much valuable information I would hate to lose.
I used to worry about this also. But what finally made me decide to buy a house (now on my second, having made a tidy profit on the first in just 5 years) was:
1) I own a nice subwoofer. Subwoofers are not apartment friendly
2) I like growing vegetables and happily dig up large amounts of lawn to turn into garden space.
3) I can mod it however the hell I like (well within reason, but you get the idea). If I decide to run Cat6 to each room and put a WAP in a tree in the backyard, I don't have to clear it with anyone.
For me, the advantages far outweighed the disadvantages. Obviously YMMV.
kinda OT, but I'm bored today
With Dish Mover you don't hire anyone, just call and they make an appt to send a service tech out to your house. No charge.
Airplanes can cause loss of signal. I used to live on one of the departure/approach paths at MSP and occasionally the signal was lost for a few seconds when an airplane was climbing out more or less in the beam path. But I think it's more likely that the receiver was being saturated by transmissions from the plane than the aircraft itself blocking the signal.
One advantage over cable is that when you move you can have service immediately if you transfer the dish yourself, unlike cable where it may need to be "turned on" at the new location. But in our case, we used DishMover since it was already snowing as we were moving and I really didn't feel like climbing up on the icy roof to install the antenna -- the install tech took the time to do a really good job, better than I would have done with subzero windchills around me.
Get the degree that interests you; they're about equal in marketability. Take the "sky is falling" moaning around here with a grain of salt. Yes, jobs are hard to find in technical fields right now, especially for entry-level personnel. But this happens every few years, though it has been particularly bad for a while. It will pick up: you can't push technology away, we'll always need skilled engineers.
But I would caution you to aim whatever degree you get at a specialty. I'd look at a biology or bioinformatics minor if possible (but being in the medical field, I may be biased). The "ordinary business coder" work is going the way of the dodo. Likewise, the kind of EE work I started my career doing can be done by a motivated hobbyist these days. The fact is that as time progresses, technology becomes available to a lower and lower skill level, that's really why so many of these jobs are being sent overseas. The overseas flow will stop when it's cheaper to automate the work than to farm it out. But by then, the smart people will have moved on to a different specialty.
Now if you want a different kind of advice on job finding