When I have signed contracts to purchase things, I have had to sign waivers limiting liability. Those waivers certainly covered reasonable expectations and disclaimed certain possible defects. This is a terrible problem for both sides, because it is just completely unexpected. I have never before heard of a steel beam's magnetization causing such difficulty. TFA is pretty slim on the real effects they are experiencing. I wonder if this is just one of those pseudo-scientific problems (magnetism = evil?) or if it is a real problem, or if it's just my reading comprehension. It would be interesting to see what the field measurements actually looked like. You'd need a very strong magnet to affect a TV from any significant distance.
At least with smaller pieces of metal you can whack them a few times to re-randomize the magnetic domains. I don't know if that actually works for something large enough to support a building (you might have to hit it hard enough to damage it or the structure it supports). Depending on the alignment of the magnetic field it might be possible to form an electromagnet to cancel its field ("degauss" it). Or the structural members can be replaced and removed (I've done this in my house). Most of these options are pretty expensive (except for the first one where you hit it a lot with a hammer).
It seems unfair for me, as a homebuyer, to get stuck dealing with a house which was built with nonstandard components (in the form of a magnetic structural support). From the builder's perspective it seems like this would be something that they would have to eat and then go after the material seller for their losses, if they can prove when the magnetization occurred.
In answer to OP's question, I suspect that the movie mafia are going after downloaders at large ISPs because the payoff is bigger - they get one process going through one legal department, and a number of names and addresses of suspected downloaders is produced, after which they send out ransom letters. At smaller ISPs, there are fewer targets to send letters to, so the cost is proportionally higher per target.
So in a sense, you could say that customers at smaller ISPs are safer, depending on the movie organizations' intent - if they want to make money off people and get a lot of big settlements to make news, they would focus on large ISPs and their customers. If they want to thoroughly scare people, they'll go after everyone, independent of the cost of doing so.
As of right now the largest FLASH I can find is a 512 Gb unit from Micron (MT29F512G08CUCABH3-12) in a 100 ball LBGA. Couldn't find that package description but maybe a similar one is 9x15.5mm dimension. You'll need more than 32 of these to get to 2 TB, plus a couple of controller ICs.
In short, with tomorrow's technology (what Micron is still developing), you will need a 6" long stick, covered with ICs on both sides. This will not be an inexpensive device for at least a few years.
Hi Kurtis - or anyone else who has spent more time looking -
Is there a schematic or parts list available for the hardware? I looked over the paper but it was pretty high level. I've been wanting to play with exactly this sort of thing for some time. Sorry if I haven't googled enough...
I've been to a few smaller PCB fabs a few years ago, before the days of 4PCB and PCBExpress and the like - mail order, nearly overnight, you fit it into their process flow shops. Anyway, this is back when a 4 layer board run was a $2k/2 week kind of deal rather than the $500 or so you can get now (or cheaper if you can wait). Those places were FILTHY and smelled like all kinds of hell. Nasty business. It's amazing how far these guys have come.
The value is so much better now too. Ten years ago, to get an overnight board we used to mill out two layer boards using a piece of prepreg with copper on either side. A guy would machine off all the copper we didn't want, then drill holes where we needed vias to connect from one side or the other. Then I had to fill the vias with little pieces of wire and solder each side, then stuff the board, then test and debug it; over repeated rework cycles the board would start to peel apart. On top of that, if you get the board hot enough, the vias (wires) would fall out and that was pretty hard to figure out. It was gravity assisted current limit.
Now, you finish your board design and ship it off to one of these guys. During the time you used to spend getting to square 1 with the milled board, you could order parts and then the board shows up from one of these guys like 4PCB here. A 2 day turn on a 4 layer board is no problem and just a few hundred bucks. The time I spent soldering vias into the milled board cost more than the real PCB I can get now. It's amazing. The way they get the price down is a combination of two things - first, you fit into their process flow, as I mentioned earlier. That means that they don't look at your board, they don't think about your board, they just cram it on a panel with some other guys' boards. If you want slots made in the board, you don't get 'em; if you want internal routs cut out of your board you don't get 'em. You get what their process says it does, and so does everyone else. This leads to the second way they get price down - volume. Lots of guys now order from a couple big shops, rather than these little (pretty dirty, as I mentioned) little mom-n-pop PCB houses. And we all order the same process.
It's amazing to see how some of these basic market principals have worked in the past ten years, and it has made a huge change in the R&D industry. It's much easier to do a pilot run of a board, it's much easier and cheaper to make a limited run, and since you are risking less you can order more and try things out. Truly awesome for an electrical guy.
Man, I am so bummed right now. These guys were some of the greatest in the field. Both were very well known in industry and had done a lot in their time to advance the state of the art. I always figured that if I ever got back into the semiconductor industry I'd try to work wherever they were - of course, I don't want to move to California very much, and Pease had sort of retired, but still.
I had no idea Jim Williams had died either. Williams' app notes were both clever and clear, just masterpieces of design and communications. One of my favorites was his AN45, which he worked on while up late with a baby; each circuit was preceded by a number of baby bottles indicating how many bottles he fed his son while working on the circuit. One of them - a CCFL supply built around a Royer oscillator - took more than 30 bottles, drawn lined up three rows deep, convinced me not to use that design in a related project. Just a few months ago Electronic Design or EDN published a paper by Williams describing how to build an ultrasonic thermometer - a technique for measuring temperature using the speed of sound in an olive jar full of dry air. Just neat stuff.
Pease was an interesting guy who I felt I knew better. You could call him up when he was at National and ask him questions - if you had a hard enough question. He had some crafty designs for VFC's and references, but I really remember him for his magazine articles. He once described a proposed highway as a mistake and showed a circuit model for traffic to describe how its construction would make things worse overall. Pease was a neat guy who I knew only through his articles and app notes, and boy would I have liked to go for a ride with him in his old VW. Course, I would have worn a seatbelt (did his car even have them?).
One of the commenters on the FF extension suggested that the extension introduces a serious vulnerability into your browser - by downloading the XML file containing the list of sites to be redirected, you are basically offering that website the ability to redirect "youtube.com" to "nastysexxxxxychix.com" or whatever. Certainly this would be unpleasant on a work computer, but it could also be used to send you to a malicious site. He also pointed out that every 15 times the extension is actually called you are sent to a "Help Us" page where they probably ask for donations.
The same commenter forked the extension to another called FireIce which has a hardcoded list of sites. I think the ideal way would be with a user-configurable list which the user can easily update from a website as desired, rather than automatically downloading an XML file without user input.
What you are saying is that you shouldn't trust people who know the field best. Certainly it makes sense to keep in mind that the people on there may have hidden agendas, and you should use your brain. A forum populated by people informed on the topic because they actually work in the field is the place I want to go to become informed. These are guys who will present or point you towards raw data and help you learn to interpret it yourself. It is possible to work in a field yet retain integrity and ethics.
I agree with your final point - anything you read on the internet should be taken with a grain of salt.
I don't think the point is quite so simple as that. The marketing people are increasing the perceived value of a product by bundling things together and selling for a higher price. The inherent value of the goods is only what consumers are willing to pay. In this case, consumers (at least one consumer, the parent poster) want to buy the unbundled product without all the extra fixings. Normally the invisible hand of the free market will serve up the product I actually desire, but since we're not dealing with fungible goods here like gasoline or apples, only one vendor sells the product and he can sell it however he wants to. And so the parent poster does not purchase the product.
I find this quite frustrating as well. Sometimes I only want two peppers but they come in a pack of three. Dang.
Well, at least when I got out of the semi business a few years ago you could still do a lot with old 5" and 6" wafers. There's a lot to be said for having your own process line, despite the fabless trend, especially in the analog world. An in-house analog process enables a semi manufacturer to build unique parts that a competitor can't as easily replicate. If you can get a higher voltage or current in a similar sized driver IC you can outsell on features, or you can shrink the die and match features and outsell on price. But if you're both buying the same process from the foundry, what advantage do you have that the competition can't get by offering your engineers more money?
You shouldn't be. You can't do digital without analog, despite what every pointy-headed manager puts in his powerpoint slides. Power is analog and that's a sizeable fraction of your computer budget. Motor control (hard drives), sensors (you name it), a lot of user interface, are all analog. Even signal transmission is analog, although if you set your thresholds just right you can pretend it's digital. In fact this is where a lot of semi companies make their money, by encapsulating the messy analog into the chip so all you have to do is put down two capacitors and hook up the digital interface, because people are escared of analog.
It's not EMP. It's high energy particles affecting the operation of the ICs inside an electronics box. You get soft errors like what happens when you are probing a chip with the lights on. Or you can get permanent damage from particles stressing the lattice, and there's no Faraday cage that can shield a robot from gammas or neutrons. It's tough to build a complicated system (with numerous chips) which is rad hard.
Modern digital ICs with very small feature sizes are highly susceptible to this damage, so you're talking about older technology. I've heard of paralleling functions (computers, reset circuits, etc.) up and making them vote on each decision, which might be why they did that on the moon missions.
I will make the same comment I make every time we debate technology's superiority to paper:
I cannot remember the last time my map crashed. It may be inaccurate (but so may GPS), it may be out of date (but so may GPS), it may not be intuitive (but so may GPS). But when I turn too fast and pull the plug out of the lighter socket, my paper map will still work. When some jerk is driving next to me with non-FCC licensed equipment drowning out the GPS band, my paper map will still work. It doesn't call out turns a mile ahead, it doesn't show up-to-the-thirty-minutes-ago traffic, all it does is show me where I am and I can use my brain to figure out where I'm going.
A GPS is superior to a map but does not replace it, and becoming reliant on a GPS to the point where I do not consult or bring a paper map is foolhardy.
In contrast to what the summary suggests, there's a big difference between an Arduino and a 555. The 555 was a barebones timer chip which could be misapplied to accomplish a lot, but required a decent understanding of basic electronics to use in any way other than its basic concept allowed (a timer/multivibrator). The Arduino is a toy, mostly for non-engineers, who are interested in learning more. I don't think there is a real analogue (oh, I'm so punny) to the 555 in the microcontroller world except, perhaps, the $5 TI Launchpad as so many other posters have suggested. The Launchpad is a cheap and nearly barebones micro which can be used to do a wide range of simple tasks and can be misapplied to very complicated things given a sufficiently capable user.
I think this is the real root of why most EEs "hate" the Arduino - it's overkill for most of what it is used for. "Proper" engineering implies designing the right solution to a problem, whereas Arduino "engineering" applies an Arduino to a problem. In both cases a solution is achieved, but the Arduino solution is so often needlessly complicated: a microcontroller is not needed to blink an LED when a 555 will suffice.
Based on what gizmag presents I'm not terribly impressed. There are several reasons to put the VR offboard. First is space, second is heat. A VR consumes a lot of both (relative to a microprocessor). You can easily see >5" square of space and >10W of power dissipation next to the processor (and everyone cries about it because of its location).
Since I don't see any resonant components included in the design it appears to me that this is a linear regulator, which will put out a lot of heat. In addition as it stands both Intel and AMD have the ability to dynamically scale the voltage they are being powered from. They can request a higher or lower voltage as well as (of course) draw more or less current, instantly, through their VID pins. So this sure doesn't sound like a great discovery, especially when you consider that the basic concept, as presented in the summary, is widely used and quite well known. But summaries by their definition don't tell the whole story, so perhaps there's more to it. I'll pull up the paper if it's available when I get to work.
As chip-on-chip technology becomes more widespread I will be interested to see what happens. It seems like there may be a place for "on chip" (as far as the enduser is concerned) voltage regulation with some of these all-in-one converter MCMs.
actually in his case it refers to CounterIntelligence as he wrote. This is one of two types of polygraphs used by the US for granting and retaining certain types of defense security clearances.
I bought the Sheevaplug and the Guruplug for some engineering applications, and was sorely disappointed. The Sheevaplug was a decent box, just needed more native IO. The Guruplug was a piece of crap. I had more issues with that box than any embedded box I've ever worked on, including some I've designed (which is saying something when you factor in initial debug time). The Guruplug had major heat issues, even when run from an external 5V supply. I removed the heat spreader (a thin piece of steel) and replaced it with a thicker copper spreader, and that made a big difference, but the unit was never completely stable and could not handle running two GigE interfaces at the same time. And they also had the niggling little problem of selling something different from what they advertised - the sale product did not have an I2C port (I think they finally changed the block diagram to reflect the truth).
By sticking with the same form factor I fear that the Globalscale product will continue to be plagued with heat issues. And based on the history of Globalscale's products, if you need a stable platform that does what the specs say they'll do - look elsewhere.
I'll probably get one to evaluate it, but this time I'm waiting. Someone else can be the early adopter.
I've long had a theory about that. Suppose you include polling information as well. If the polls say that 60% are voting for candidate A, with a 5% margin of error - why vote for either of the big two candidates? Then you certainly ARE wasting your vote, by piling on unnecessarily where your vote is not needed. Eventually, if enough people adopt this strategy, the marginal vote will again make a difference, but by then many more people will be voting (and used to voting) for a 3rd party.
Really though, there's hardly any difference between any two candidates A and B, as long as the government isn't controlled by one party. At least it certainly seems that way, except for the relatively petty issues intended to divide up the electorate.
This particular reference led me to an interesting section of US law I was unaware existed - Title 18, Section 793(e):
Whoever, having unauthorized possession of, access to, or control over any document... relating to the national defense, or information relating to the national defense which information the possessor has reason to believe could be used to the injury of the United States or to the advantage of any foreign nation, willfully communicates... or causes to be communicated... the same to any person not entitled to receive it, or willfully retains the same and fails to deliver it to the officer or employee of the United States entitled to receive it... [s]hall be fined not more than $10,000 or imprisoned not more than ten years, or both. (take the widest possible definition of "communicate" because I deleted all the other similar words for clarity)
It's something of a corollary to the UK Official Secrets Act. I've no idea whether this is applicable to anyone subject to US law or just clearance holders (as I am too lazy to read the entire section of code). It would seem to me that all of the US newspaper editors collaborating with Wikileaks are in violation of this section. I'll reserve my personal opinions on that particular idea.
A lot of little things add up to one big thing... [s]o dismissing something because it's insignificant or small is pretty much why poor people tend to remain poor- even with ever increasing incomes.
I'm just going to point you to your own first suggestion. And judging from the site, they can't be wasting to terribly much money on it.
It is always amusing to see a person pointing out logical fallacies who then uses the same fallacy to argue the point. Anyhow, GP's point was that it is silly to cut essential things to save a few relative pennies, especially when cutting such things will cost more in the long term. For example, if you save money on vegetables and just feed the baby from McDonalds, her resulting health problems will cost a great deal later. Society (i.e. government, i.e. all of us) end up footing the bill.
At the same time it is essential to review all of our spending and figure out where we can cut it. I think we will have to eventually bite the bullet and swallow an austerity plan to balance our budget. The crushing costs of both funding nondiscretionary programs such as social security and medicare as well as discretionary programs such as two wars and I don't even know what - these expenses are too great for our current taxes. We must either cut spending or raise taxes, soon, or risk devaluation of our currency when everyone else finally accepts that we can't afford our current lifestyle. But nobody (including me) wants to accept that because it will hurt too much.
But a good start while the country grows to accept the inevitable is to look for the low-hanging fruit - cut or combine overlapping programs. Then cut programs which have minimal impact. If its cost is greater than its benefit, cut it. Assess programs from a complete financial perspective, and then consider the humanitarian impact of those financial decisions, and then decide. This YouCut thing could be a useful means of collecting perspectives on both. Too bad it's a partisan project.
By now this reply is long enough to have committed the cardinal sin of slashdot - I started by critiquing another and have no doubt committed the same sin I critiqued. Should have stopped before I got on my soapbox.
IAAEE and I do design work; I'm in my lab daily and helped design the current lab. My lab has two separate spaces, one for mech work and the other for EE work. The mech lab has a bandsaw, drill press, grinder, lathe and milling machine. For the space you have I'd say get the mill. You can use it to drill, cut and face a small workpiece. Look at Grizzly tools (basically Harbor Freight) for decent deals and ok quality equipment.
The EE lab has a lot of kit. You need at least one dual power supply for each station where you'll do electrical work (look at BK, they have several value items), as well as a soldering iron. For irons the near-instant heat types are truly awesome and IMO well worth the cost. We like Weller but as long as they get hot they'll work fine. One piece of advice - however many soldering stations you intend to have, you should buy them all at one time. Weller loves to keep redesigning the tips so I have to stock five different families of tips in my lab. Don't waste money on benchtop multimeters, for the cost of one great Agilent we got four Fluke 77's - sufficient for most work and more easily portable. You need at least two multimeters per EE station. You should also invest in scopes if you do anything analog. You must have at least one good scope in the lab for every two EE's, and you should also have a bad scope. We have some cheap Teks as well as one awesome MSO4000 series which has a 16 channel mini-LA - expensive but worth every penny. If you do much digital look at the USBee for a cheap LA. Also at least one good function generator is very useful; look at Agilent and SRS for good ones. You may need a spectrum analyzer depending on what you're doing, too.
Back to the basics, though. You can fit two 6' benches on each 15' wall with space between. Global Industries sells these at reasonable cost but Costco might have some near you for ~$150. A good bench will have a hutch shelf with a row of outlets on it (outlets you can see are really important, but you can also get a 6' industrial power strip instead if you need). I'd get two, and a stand for the mill, as well as storage. A good rolling tool rack (with ball bearing slides! don't skimp on that!) and a cabinet will use most of the other wall. Look at a good shelving cabinet like Akro Mills make if you will stock small parts. Expensive but an excellent way to store and organize your SMT parts or hardware. And think about how/where you'll store your scrap mechanical stock - you might still have room for a good open vertical shelf.
You will want all the storage you can fit in there. Think about a layer of shelves at 6.5' off the ground all the way around the room for rarely-used equipment. And your lab needs power and data. You should have a dedicated power circuit (or two) for the lab and more ethernet ports than usual (preferably two per work station or more). Finally since you are short on space you should try to get a laptop dedicated to lab use, and preferably one with a real db-9 serial port on it (or a base station with one).
That would be my Christmas list for your lab. Hope it helps.
When I used to work in the freight industry, just about everything got kicked or dropped once. A "Fragile" sticker meant kick twice.
In all seriousness, materials need to be packed to survive fairly brutal treatment. The only things we were particularly careful with were live animals, dead people (they want to get shipped home) and radioactive material. We were careful with live animals because they make enough noise already. No way in hell did I want to drive a fork through a dead guy's coffin (bad hoojoo). And the same goes for the hazmat, just for different reasons.
So you would get better treatment if you put a bunch of hazmat stickers on your material, but the associated paperwork would make it quite costly to ship. Pretty hard to masquerade as any of the other things. Best bet - pack your stuff correctly, even at the cost of increased weight or size (which we also charged for as "dimensional weight") if you want it to get there in one piece. If it's worth something to you, spend the time and money to pack it right.
QA is a process of verifying that the part performs to specs. Unless there is a spec which says that "this part will not install malware" they aren't going to look for it. QA is generally an overworked and underappreciated function of a large manufacturer and they don't have time to do extra.
In addition, in most any company large enough to have a real QA department, the QA folks operate under a strict regime of policies, procedures and audits to verify such. Which means they aren't really allowed to screw around looking for something else. And if the spec actually requires that the part not install malware, then you have the case which a sibling poster described... the malware must be crafted in such a way as to hide its presence. I can think of at least one pretty easy way to do that (only attempt to install malware after X power cycles or after running X time...) and there are likely plenty.
An insider capable of installing this capability in nascent hardware would likely be able to circumvent any QA protections.
Slashdot Mirror
When I have signed contracts to purchase things, I have had to sign waivers limiting liability. Those waivers certainly covered reasonable expectations and disclaimed certain possible defects. This is a terrible problem for both sides, because it is just completely unexpected. I have never before heard of a steel beam's magnetization causing such difficulty. TFA is pretty slim on the real effects they are experiencing. I wonder if this is just one of those pseudo-scientific problems (magnetism = evil?) or if it is a real problem, or if it's just my reading comprehension. It would be interesting to see what the field measurements actually looked like. You'd need a very strong magnet to affect a TV from any significant distance.
At least with smaller pieces of metal you can whack them a few times to re-randomize the magnetic domains. I don't know if that actually works for something large enough to support a building (you might have to hit it hard enough to damage it or the structure it supports). Depending on the alignment of the magnetic field it might be possible to form an electromagnet to cancel its field ("degauss" it). Or the structural members can be replaced and removed (I've done this in my house). Most of these options are pretty expensive (except for the first one where you hit it a lot with a hammer).
It seems unfair for me, as a homebuyer, to get stuck dealing with a house which was built with nonstandard components (in the form of a magnetic structural support). From the builder's perspective it seems like this would be something that they would have to eat and then go after the material seller for their losses, if they can prove when the magnetization occurred.
In answer to OP's question, I suspect that the movie mafia are going after downloaders at large ISPs because the payoff is bigger - they get one process going through one legal department, and a number of names and addresses of suspected downloaders is produced, after which they send out ransom letters. At smaller ISPs, there are fewer targets to send letters to, so the cost is proportionally higher per target.
So in a sense, you could say that customers at smaller ISPs are safer, depending on the movie organizations' intent - if they want to make money off people and get a lot of big settlements to make news, they would focus on large ISPs and their customers. If they want to thoroughly scare people, they'll go after everyone, independent of the cost of doing so.
As of right now the largest FLASH I can find is a 512 Gb unit from Micron (MT29F512G08CUCABH3-12) in a 100 ball LBGA. Couldn't find that package description but maybe a similar one is 9x15.5mm dimension. You'll need more than 32 of these to get to 2 TB, plus a couple of controller ICs.
In short, with tomorrow's technology (what Micron is still developing), you will need a 6" long stick, covered with ICs on both sides. This will not be an inexpensive device for at least a few years.
Hi Kurtis - or anyone else who has spent more time looking -
Is there a schematic or parts list available for the hardware? I looked over the paper but it was pretty high level. I've been wanting to play with exactly this sort of thing for some time. Sorry if I haven't googled enough...
I've been to a few smaller PCB fabs a few years ago, before the days of 4PCB and PCBExpress and the like - mail order, nearly overnight, you fit it into their process flow shops. Anyway, this is back when a 4 layer board run was a $2k/2 week kind of deal rather than the $500 or so you can get now (or cheaper if you can wait). Those places were FILTHY and smelled like all kinds of hell. Nasty business. It's amazing how far these guys have come.
The value is so much better now too. Ten years ago, to get an overnight board we used to mill out two layer boards using a piece of prepreg with copper on either side. A guy would machine off all the copper we didn't want, then drill holes where we needed vias to connect from one side or the other. Then I had to fill the vias with little pieces of wire and solder each side, then stuff the board, then test and debug it; over repeated rework cycles the board would start to peel apart. On top of that, if you get the board hot enough, the vias (wires) would fall out and that was pretty hard to figure out. It was gravity assisted current limit.
Now, you finish your board design and ship it off to one of these guys. During the time you used to spend getting to square 1 with the milled board, you could order parts and then the board shows up from one of these guys like 4PCB here. A 2 day turn on a 4 layer board is no problem and just a few hundred bucks. The time I spent soldering vias into the milled board cost more than the real PCB I can get now. It's amazing. The way they get the price down is a combination of two things - first, you fit into their process flow, as I mentioned earlier. That means that they don't look at your board, they don't think about your board, they just cram it on a panel with some other guys' boards. If you want slots made in the board, you don't get 'em; if you want internal routs cut out of your board you don't get 'em. You get what their process says it does, and so does everyone else. This leads to the second way they get price down - volume. Lots of guys now order from a couple big shops, rather than these little (pretty dirty, as I mentioned) little mom-n-pop PCB houses. And we all order the same process.
It's amazing to see how some of these basic market principals have worked in the past ten years, and it has made a huge change in the R&D industry. It's much easier to do a pilot run of a board, it's much easier and cheaper to make a limited run, and since you are risking less you can order more and try things out. Truly awesome for an electrical guy.
Man, I am so bummed right now. These guys were some of the greatest in the field. Both were very well known in industry and had done a lot in their time to advance the state of the art. I always figured that if I ever got back into the semiconductor industry I'd try to work wherever they were - of course, I don't want to move to California very much, and Pease had sort of retired, but still.
I had no idea Jim Williams had died either. Williams' app notes were both clever and clear, just masterpieces of design and communications. One of my favorites was his AN45, which he worked on while up late with a baby; each circuit was preceded by a number of baby bottles indicating how many bottles he fed his son while working on the circuit. One of them - a CCFL supply built around a Royer oscillator - took more than 30 bottles, drawn lined up three rows deep, convinced me not to use that design in a related project. Just a few months ago Electronic Design or EDN published a paper by Williams describing how to build an ultrasonic thermometer - a technique for measuring temperature using the speed of sound in an olive jar full of dry air. Just neat stuff.
Pease was an interesting guy who I felt I knew better. You could call him up when he was at National and ask him questions - if you had a hard enough question. He had some crafty designs for VFC's and references, but I really remember him for his magazine articles. He once described a proposed highway as a mistake and showed a circuit model for traffic to describe how its construction would make things worse overall. Pease was a neat guy who I knew only through his articles and app notes, and boy would I have liked to go for a ride with him in his old VW. Course, I would have worn a seatbelt (did his car even have them?).
Man am I bummed right now! What a loss.
One of the commenters on the FF extension suggested that the extension introduces a serious vulnerability into your browser - by downloading the XML file containing the list of sites to be redirected, you are basically offering that website the ability to redirect "youtube.com" to "nastysexxxxxychix.com" or whatever. Certainly this would be unpleasant on a work computer, but it could also be used to send you to a malicious site. He also pointed out that every 15 times the extension is actually called you are sent to a "Help Us" page where they probably ask for donations.
The same commenter forked the extension to another called FireIce which has a hardcoded list of sites. I think the ideal way would be with a user-configurable list which the user can easily update from a website as desired, rather than automatically downloading an XML file without user input.
This other extension - which I haven't tried and cannot endorse - is at https://addons.mozilla.org/da/firefox/addon/fireice/
I think your tin foil hat is too thick.
What you are saying is that you shouldn't trust people who know the field best. Certainly it makes sense to keep in mind that the people on there may have hidden agendas, and you should use your brain. A forum populated by people informed on the topic because they actually work in the field is the place I want to go to become informed. These are guys who will present or point you towards raw data and help you learn to interpret it yourself. It is possible to work in a field yet retain integrity and ethics.
I agree with your final point - anything you read on the internet should be taken with a grain of salt.
I don't think the point is quite so simple as that. The marketing people are increasing the perceived value of a product by bundling things together and selling for a higher price. The inherent value of the goods is only what consumers are willing to pay. In this case, consumers (at least one consumer, the parent poster) want to buy the unbundled product without all the extra fixings. Normally the invisible hand of the free market will serve up the product I actually desire, but since we're not dealing with fungible goods here like gasoline or apples, only one vendor sells the product and he can sell it however he wants to. And so the parent poster does not purchase the product.
I find this quite frustrating as well. Sometimes I only want two peppers but they come in a pack of three. Dang.
Well, at least when I got out of the semi business a few years ago you could still do a lot with old 5" and 6" wafers. There's a lot to be said for having your own process line, despite the fabless trend, especially in the analog world. An in-house analog process enables a semi manufacturer to build unique parts that a competitor can't as easily replicate. If you can get a higher voltage or current in a similar sized driver IC you can outsell on features, or you can shrink the die and match features and outsell on price. But if you're both buying the same process from the foundry, what advantage do you have that the competition can't get by offering your engineers more money?
You shouldn't be. You can't do digital without analog, despite what every pointy-headed manager puts in his powerpoint slides. Power is analog and that's a sizeable fraction of your computer budget. Motor control (hard drives), sensors (you name it), a lot of user interface, are all analog. Even signal transmission is analog, although if you set your thresholds just right you can pretend it's digital. In fact this is where a lot of semi companies make their money, by encapsulating the messy analog into the chip so all you have to do is put down two capacitors and hook up the digital interface, because people are escared of analog.
Can you tell I'm an analog guy? I sure hope so.
It's not EMP. It's high energy particles affecting the operation of the ICs inside an electronics box. You get soft errors like what happens when you are probing a chip with the lights on. Or you can get permanent damage from particles stressing the lattice, and there's no Faraday cage that can shield a robot from gammas or neutrons. It's tough to build a complicated system (with numerous chips) which is rad hard.
Modern digital ICs with very small feature sizes are highly susceptible to this damage, so you're talking about older technology. I've heard of paralleling functions (computers, reset circuits, etc.) up and making them vote on each decision, which might be why they did that on the moon missions.
See http://en.wikipedia.org/wiki/Rad-hard for more info.
I will make the same comment I make every time we debate technology's superiority to paper:
I cannot remember the last time my map crashed. It may be inaccurate (but so may GPS), it may be out of date (but so may GPS), it may not be intuitive (but so may GPS). But when I turn too fast and pull the plug out of the lighter socket, my paper map will still work. When some jerk is driving next to me with non-FCC licensed equipment drowning out the GPS band, my paper map will still work. It doesn't call out turns a mile ahead, it doesn't show up-to-the-thirty-minutes-ago traffic, all it does is show me where I am and I can use my brain to figure out where I'm going.
A GPS is superior to a map but does not replace it, and becoming reliant on a GPS to the point where I do not consult or bring a paper map is foolhardy.
In contrast to what the summary suggests, there's a big difference between an Arduino and a 555. The 555 was a barebones timer chip which could be misapplied to accomplish a lot, but required a decent understanding of basic electronics to use in any way other than its basic concept allowed (a timer/multivibrator). The Arduino is a toy, mostly for non-engineers, who are interested in learning more. I don't think there is a real analogue (oh, I'm so punny) to the 555 in the microcontroller world except, perhaps, the $5 TI Launchpad as so many other posters have suggested. The Launchpad is a cheap and nearly barebones micro which can be used to do a wide range of simple tasks and can be misapplied to very complicated things given a sufficiently capable user.
I think this is the real root of why most EEs "hate" the Arduino - it's overkill for most of what it is used for. "Proper" engineering implies designing the right solution to a problem, whereas Arduino "engineering" applies an Arduino to a problem. In both cases a solution is achieved, but the Arduino solution is so often needlessly complicated: a microcontroller is not needed to blink an LED when a 555 will suffice.
Based on what gizmag presents I'm not terribly impressed. There are several reasons to put the VR offboard. First is space, second is heat. A VR consumes a lot of both (relative to a microprocessor). You can easily see >5" square of space and >10W of power dissipation next to the processor (and everyone cries about it because of its location).
Since I don't see any resonant components included in the design it appears to me that this is a linear regulator, which will put out a lot of heat. In addition as it stands both Intel and AMD have the ability to dynamically scale the voltage they are being powered from. They can request a higher or lower voltage as well as (of course) draw more or less current, instantly, through their VID pins. So this sure doesn't sound like a great discovery, especially when you consider that the basic concept, as presented in the summary, is widely used and quite well known. But summaries by their definition don't tell the whole story, so perhaps there's more to it. I'll pull up the paper if it's available when I get to work.
As chip-on-chip technology becomes more widespread I will be interested to see what happens. It seems like there may be a place for "on chip" (as far as the enduser is concerned) voltage regulation with some of these all-in-one converter MCMs.
actually in his case it refers to CounterIntelligence as he wrote. This is one of two types of polygraphs used by the US for granting and retaining certain types of defense security clearances.
http://en.wikipedia.org/wiki/List_of_U.S._security_clearance_terms
I bought the Sheevaplug and the Guruplug for some engineering applications, and was sorely disappointed. The Sheevaplug was a decent box, just needed more native IO. The Guruplug was a piece of crap. I had more issues with that box than any embedded box I've ever worked on, including some I've designed (which is saying something when you factor in initial debug time). The Guruplug had major heat issues, even when run from an external 5V supply. I removed the heat spreader (a thin piece of steel) and replaced it with a thicker copper spreader, and that made a big difference, but the unit was never completely stable and could not handle running two GigE interfaces at the same time. And they also had the niggling little problem of selling something different from what they advertised - the sale product did not have an I2C port (I think they finally changed the block diagram to reflect the truth).
By sticking with the same form factor I fear that the Globalscale product will continue to be plagued with heat issues. And based on the history of Globalscale's products, if you need a stable platform that does what the specs say they'll do - look elsewhere.
I'll probably get one to evaluate it, but this time I'm waiting. Someone else can be the early adopter.
I've long had a theory about that. Suppose you include polling information as well. If the polls say that 60% are voting for candidate A, with a 5% margin of error - why vote for either of the big two candidates? Then you certainly ARE wasting your vote, by piling on unnecessarily where your vote is not needed. Eventually, if enough people adopt this strategy, the marginal vote will again make a difference, but by then many more people will be voting (and used to voting) for a 3rd party.
Really though, there's hardly any difference between any two candidates A and B, as long as the government isn't controlled by one party. At least it certainly seems that way, except for the relatively petty issues intended to divide up the electorate.
Never posted one of these before but it is the most insightful comment in the thread and deserves to be seen.
The NDA is definitely required as one of the three components necessary for access to classified information.
http://www.archives.gov/isoo/training/standard-form-312.html
This particular reference led me to an interesting section of US law I was unaware existed - Title 18, Section 793(e):
Whoever, having unauthorized possession of, access to, or control over any document... relating to the national defense, or information relating to the national defense which information the possessor has reason to believe could be used to the injury of the United States or to the advantage of any foreign nation, willfully communicates... or causes to be communicated... the same to any person not entitled to receive it, or willfully retains the same and fails to deliver it to the officer or employee of the United States entitled to receive it... [s]hall be fined not more than $10,000 or imprisoned not more than ten years, or both.
(take the widest possible definition of "communicate" because I deleted all the other similar words for clarity)
It's something of a corollary to the UK Official Secrets Act. I've no idea whether this is applicable to anyone subject to US law or just clearance holders (as I am too lazy to read the entire section of code). It would seem to me that all of the US newspaper editors collaborating with Wikileaks are in violation of this section. I'll reserve my personal opinions on that particular idea.
A lot of little things add up to one big thing... [s]o dismissing something because it's insignificant or small is pretty much why poor people tend to remain poor- even with ever increasing incomes.
I'm just going to point you to your own first suggestion. And judging from the site, they can't be wasting to terribly much money on it.
It is always amusing to see a person pointing out logical fallacies who then uses the same fallacy to argue the point. Anyhow, GP's point was that it is silly to cut essential things to save a few relative pennies, especially when cutting such things will cost more in the long term. For example, if you save money on vegetables and just feed the baby from McDonalds, her resulting health problems will cost a great deal later. Society (i.e. government, i.e. all of us) end up footing the bill.
At the same time it is essential to review all of our spending and figure out where we can cut it. I think we will have to eventually bite the bullet and swallow an austerity plan to balance our budget. The crushing costs of both funding nondiscretionary programs such as social security and medicare as well as discretionary programs such as two wars and I don't even know what - these expenses are too great for our current taxes. We must either cut spending or raise taxes, soon, or risk devaluation of our currency when everyone else finally accepts that we can't afford our current lifestyle. But nobody (including me) wants to accept that because it will hurt too much.
But a good start while the country grows to accept the inevitable is to look for the low-hanging fruit - cut or combine overlapping programs. Then cut programs which have minimal impact. If its cost is greater than its benefit, cut it. Assess programs from a complete financial perspective, and then consider the humanitarian impact of those financial decisions, and then decide. This YouCut thing could be a useful means of collecting perspectives on both. Too bad it's a partisan project.
By now this reply is long enough to have committed the cardinal sin of slashdot - I started by critiquing another and have no doubt committed the same sin I critiqued. Should have stopped before I got on my soapbox.
IAAEE and I do design work; I'm in my lab daily and helped design the current lab. My lab has two separate spaces, one for mech work and the other for EE work. The mech lab has a bandsaw, drill press, grinder, lathe and milling machine. For the space you have I'd say get the mill. You can use it to drill, cut and face a small workpiece. Look at Grizzly tools (basically Harbor Freight) for decent deals and ok quality equipment.
The EE lab has a lot of kit. You need at least one dual power supply for each station where you'll do electrical work (look at BK, they have several value items), as well as a soldering iron. For irons the near-instant heat types are truly awesome and IMO well worth the cost. We like Weller but as long as they get hot they'll work fine. One piece of advice - however many soldering stations you intend to have, you should buy them all at one time. Weller loves to keep redesigning the tips so I have to stock five different families of tips in my lab. Don't waste money on benchtop multimeters, for the cost of one great Agilent we got four Fluke 77's - sufficient for most work and more easily portable. You need at least two multimeters per EE station. You should also invest in scopes if you do anything analog. You must have at least one good scope in the lab for every two EE's, and you should also have a bad scope. We have some cheap Teks as well as one awesome MSO4000 series which has a 16 channel mini-LA - expensive but worth every penny. If you do much digital look at the USBee for a cheap LA. Also at least one good function generator is very useful; look at Agilent and SRS for good ones. You may need a spectrum analyzer depending on what you're doing, too.
Back to the basics, though. You can fit two 6' benches on each 15' wall with space between. Global Industries sells these at reasonable cost but Costco might have some near you for ~$150. A good bench will have a hutch shelf with a row of outlets on it (outlets you can see are really important, but you can also get a 6' industrial power strip instead if you need). I'd get two, and a stand for the mill, as well as storage. A good rolling tool rack (with ball bearing slides! don't skimp on that!) and a cabinet will use most of the other wall. Look at a good shelving cabinet like Akro Mills make if you will stock small parts. Expensive but an excellent way to store and organize your SMT parts or hardware. And think about how/where you'll store your scrap mechanical stock - you might still have room for a good open vertical shelf.
You will want all the storage you can fit in there. Think about a layer of shelves at 6.5' off the ground all the way around the room for rarely-used equipment. And your lab needs power and data. You should have a dedicated power circuit (or two) for the lab and more ethernet ports than usual (preferably two per work station or more). Finally since you are short on space you should try to get a laptop dedicated to lab use, and preferably one with a real db-9 serial port on it (or a base station with one).
That would be my Christmas list for your lab. Hope it helps.
When I used to work in the freight industry, just about everything got kicked or dropped once. A "Fragile" sticker meant kick twice.
In all seriousness, materials need to be packed to survive fairly brutal treatment. The only things we were particularly careful with were live animals, dead people (they want to get shipped home) and radioactive material. We were careful with live animals because they make enough noise already. No way in hell did I want to drive a fork through a dead guy's coffin (bad hoojoo). And the same goes for the hazmat, just for different reasons.
So you would get better treatment if you put a bunch of hazmat stickers on your material, but the associated paperwork would make it quite costly to ship. Pretty hard to masquerade as any of the other things. Best bet - pack your stuff correctly, even at the cost of increased weight or size (which we also charged for as "dimensional weight") if you want it to get there in one piece. If it's worth something to you, spend the time and money to pack it right.
There's a lot more detail in the symantec virus "dossier". A very interesting and detailed read.
QA is a process of verifying that the part performs to specs. Unless there is a spec which says that "this part will not install malware" they aren't going to look for it. QA is generally an overworked and underappreciated function of a large manufacturer and they don't have time to do extra.
In addition, in most any company large enough to have a real QA department, the QA folks operate under a strict regime of policies, procedures and audits to verify such. Which means they aren't really allowed to screw around looking for something else. And if the spec actually requires that the part not install malware, then you have the case which a sibling poster described... the malware must be crafted in such a way as to hide its presence. I can think of at least one pretty easy way to do that (only attempt to install malware after X power cycles or after running X time...) and there are likely plenty.
An insider capable of installing this capability in nascent hardware would likely be able to circumvent any QA protections.