You know all that cash Apple is making that isn't getting paid out in dividends? It belongs to the shareholders. Dividends are just a feel-good gesture for mature companies that are in "utility mode". It just says we don't have a better use for this cash, so we're going to pass it on to shareholders.
Sure, the rate is higher in the day - that's the whole point of it. But is it at least lower at night? I'm putting in solar in California, we have a tou tariff too. It's higher than the normal rate *during the day* but it's a good deal for most users since that is when they're making power. Even if your net mrtering rules don't require them to compensate you for that power, if you size your system correctly you should be able to offset your air conditioning load and not have much usage in that tier. My question was how, specifically is your TOU tariff a bad deal?
Can somebody explain that one to me - what is the injustice exactly?
TOU metering is exactly what you want if you have a PV installation, because it allows you to bank credits (or even just offset your usage) at midday when the price is highest, then spend them at night when electricity is cheaper.
Make the CD sound better than an MP3, and we'll buy it. IMHO.
Level compression != data compression. MP3 does not effect loudness, it's applied in the mastering process. And the reason CDs sound like MP3s is because MP3s are ripped from CDs. Indeed, if they fix CDs, our MP3s will sound better too.
The skills "reprogram industrial PLCs" and "find four new zero days in Windows" don't overlap a whole lot.
That's like saying, the skills "sweating copper pipe" and "hydraulic engineering" don't overlap a lot. It's true, but if you can do the latter, you're probably smart enough to figure out the former. And please, don't tell me how hard SCADA is... I've done it (as well as much harder things eg kernel work, VHDL, and analog circuit design), and it's all gluing together simple logic blocks and control busses. The equipment may be specialized, but that only makes it an obscure skill set, not an advanced one.
I don't necessarily disagree with your conclusion though. The aspect of actually making this an _effective_ attack would call for some specific knowledge of how the plant operates. That is of far more interest to me than the technical skills needed to code it. I'm not convinced that this really was an _effective_ hack, in terms of intelligence gained, operations halted for a long time, etc - but who knows the exact objective.
This is very cool! But you have a big problem unless I'm missing something. I only briefly skimmed your code so maybe I missed this, but it doesn't look like you are accounting for power factor. In order to do that you need to measure the voltage on each phase, ideally at a few KHz, and generate CT samples at the same rate which are multiplied by this measurement. This way you properly deal not only with low PF loads, but also variance and distortion in the AC line voltage supplied by the utility.
Some devices can have very low PFs, for example insteon switches and other small loads, and lightly loaded switching power supplies, it can be as little as 0.1x. A ceiling fan running at low speed, or a CFL might be something like 0.4. So the number you are calculating is properly called VA (volt-amps) and is not the same as watts, which is what you're actually consuming and being billed for.
I see you have put calibration factors in for each circuit. You may find that the reason you're needing these at all is because those loads are low PF and are reading higher than they should.
PS when I said automation is precisely the problem, I should have put "problem" in quotes to indicate a hint of irony. I can see how you you would say I had it all backwards since I failed to express that tone.
I agree with every word you said.
I'm NOT suggesting we should all go back to subsistence farming, and of course that there are better things we could do with our time (and for the environment, and our sanity) than commuting. I wish everyone were so creative and ambitious to take on the kinds of rewarding passtimes you suggest.
All I'm saying is that automating the commute solves a problem that people deeply do not want solved. I'm suggesting there is a difference between why people really do it, and why they say they do it.
But automation is precisely the problem. People's basic needs in the 1st world can be met so easily that they truly have nothing better to do than sit in the car and plod from suburb to office building and back. Of course this behavior is now engrained on such a spectacular scale that individuals can't just pick up and say "screw this, I'll telecommute", but as energy costs rise that'll be the ultimate outcome for those who don't absolutely NEED to be in a particular location to work (retail, food service, construction etc).
If we really wanted to not drive, we could take public transit and be free to read or whatever. But driving is the last dying vestige where white collar workers can feel like they're doing something even remotely useful _with autonomy_. Automate driving, and it means we have to find another activity to occupy ourselves since we no longer need scrub our dishes and laundry, much less toil in the fields.
I'm talking about PORTABLE electronics - the little stuff that you charge. They are talking about -plug loads-, which includes vacuum cleaners, a/v equipment, computers, etc.
I'd be surprised if portable devices account for even 0.1% of household energy usage. Who cares? Worry about your HVAC, laundry machines, refrigerator, home server, incandescent lights, etc.
Not really. You still need regulation, and 5-10W AC powered switchers are only a cubic inch or so. And they're produced as modules in such phenomenal volume that they're probably cheaper for this application. A POE supply is not trivial - it still involves a switcher, transformer, optocoupler, and a handful of passives. Basically the same stuff.
When faced with an engineering problem, I can dip into the vast sea of information at my fingertips and instantly find answers instead of spending all day flipping through hardbacks at computer literacy, bullshitting with local sales reps to try and get copies of data sheets faxed to me, or just plain wasting time figuring out something out that's already been solved. This leaves me more time to work on the interesting stuff, or fart around on Facebook if I feel like it.
I'm failing to see the downside. If you're a distractible person you can be even more distracted if you want to. If you're a productive person you can be even more productive if you want to. More information, please.
+1 on the Elk M1. The serial interface has an extensive command set. It is also accessible over ethernet, but their ethernet module is little more than a standard terminal server so you can just as easily use your own.
I have written a Perl module to communicate with it using the POE framework. So you can integrate it with your insteon, A/V, etc. It supports nearly all of the Elk commands. http://www.seanadams.com/ha/automogator
but I would love to see their presentation. It kinds of reminds me of people who think the length of their speaker cables is going to affect their sound stage, because they have no concept of the speed of sound vs an electrical field.
If you were comparing plaintext passwords with strcmp you might expect something on the order of one character comparison per instruction cycle, say 1ns. But across the internet latency variance is very noisy and more like 1ms standard deviation, ie a million times larger. I'd expect you'd have to average billions of tests against each attempted character before that one little nanosecond comes out of the noise, if it's even detectable at all. Most likely there are slower clock domains along the way that will gobble it up completely.
I could imagine in a spectacularly well controlled lab environment with no other traffic on the lan and no other activity on the server, this might be demonstrated, but there's just too much jitter in any production system to do this.
Of course none of this excuses such a braindead authentication scheme. Use a hash!
I ordered a few sets of them in varying widths. It turns out the 15" deep ones are exactly the right size to fit in an Ikea office cabinet with the doors closed. That is a cheap, easy, and tidy way to create a huge amount of well organized space for cables, batteries, adaptors, rubber bands, wire ties, computer parts, etc etc.
Save the receipts from your component purchases in electronic form so you can search them. This is handy not just for reordering but also if you want the exact part number (to look up the specs) which might not be easily determined by looking at the part.
Finally if you are collecting SMT chip caps/resistors, diodes and transistors, you can use these flip-top cases which let you store 128 different values in a small space. http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_164267_-1 There isn't much room for labeling so these are best for resistors and caps where you just need to indicate a value for each item, and the other details for the family can be labeled on the outside of the case.
If we are more efficient at producing one good, and they are more efficient at producing something else, then it doesn't make sense for us to waste money trying to produce it ourselves in the States.
You aren't addressing Grove's point, which is that there is intellectual capital to be gained and retained by continuing to manufacture such commodities ourselves. Textiles vs semiconductors differ in this respect. There is also physical capital we might like to have ready to go, i.e. the fabs, for when China gets tired of taking our Monopoly money.
I have looked at the varying twist rates in CAT6 and it is not as you describe. It cycles down the length of each pair from tight to loose twisting, with each pair being 90 degrees out of phase from the next in this cycle. The "wavelength" is something like 3 feet IIRC. So you could never get a meaningful skew between pairs regardless of cable length - if that happened, gigE would simply break because it uses two synchronous pairs in each direction.
I rather suspect that the effects you're seeing has nothing to do with time skew but is general noise and distortion occurring at the longer lengths. Certainly with analog video this would happen unless you used proper baluns and impedance matching, which is probably the "active electronics" you're thinking of.
I'm guessing you have never actually worked with Asian manufacturers. New stuff doesn't just fall out of the sky for whoever is lucky enough to "identify" it. For a customer the size of Apple it is a very close partnership and seldom does the manufacturer fully own the resulting technologies. So either you help them develop the next big thing and you get some degree of exclusivity, or you wait for someone else to pioneer it and then you get it a few years later.
Slashdot Mirror
You know all that cash Apple is making that isn't getting paid out in dividends? It belongs to the shareholders. Dividends are just a feel-good gesture for mature companies that are in "utility mode". It just says we don't have a better use for this cash, so we're going to pass it on to shareholders.
Sure, the rate is higher in the day - that's the whole point of it. But is it at least lower at night? I'm putting in solar in California, we have a tou tariff too. It's higher than the normal rate *during the day* but it's a good deal for most users since that is when they're making power. Even if your net mrtering rules don't require them to compensate you for that power, if you size your system correctly you should be able to offset your air conditioning load and not have much usage in that tier. My question was how, specifically is your TOU tariff a bad deal?
Can somebody explain that one to me - what is the injustice exactly? TOU metering is exactly what you want if you have a PV installation, because it allows you to bank credits (or even just offset your usage) at midday when the price is highest, then spend them at night when electricity is cheaper.
Level compression != data compression. MP3 does not effect loudness, it's applied in the mastering process. And the reason CDs sound like MP3s is because MP3s are ripped from CDs. Indeed, if they fix CDs, our MP3s will sound better too.
That's like saying, the skills "sweating copper pipe" and "hydraulic engineering" don't overlap a lot. It's true, but if you can do the latter, you're probably smart enough to figure out the former. And please, don't tell me how hard SCADA is... I've done it (as well as much harder things eg kernel work, VHDL, and analog circuit design), and it's all gluing together simple logic blocks and control busses. The equipment may be specialized, but that only makes it an obscure skill set, not an advanced one.
I don't necessarily disagree with your conclusion though. The aspect of actually making this an _effective_ attack would call for some specific knowledge of how the plant operates. That is of far more interest to me than the technical skills needed to code it. I'm not convinced that this really was an _effective_ hack, in terms of intelligence gained, operations halted for a long time, etc - but who knows the exact objective.
Ehh.. never mind. I see you are using ADE7753 which watches the line voltage.
Some devices can have very low PFs, for example insteon switches and other small loads, and lightly loaded switching power supplies, it can be as little as 0.1x. A ceiling fan running at low speed, or a CFL might be something like 0.4. So the number you are calculating is properly called VA (volt-amps) and is not the same as watts, which is what you're actually consuming and being billed for.
I see you have put calibration factors in for each circuit. You may find that the reason you're needing these at all is because those loads are low PF and are reading higher than they should.
PS when I said automation is precisely the problem, I should have put "problem" in quotes to indicate a hint of irony. I can see how you you would say I had it all backwards since I failed to express that tone.
I agree with every word you said. I'm NOT suggesting we should all go back to subsistence farming, and of course that there are better things we could do with our time (and for the environment, and our sanity) than commuting. I wish everyone were so creative and ambitious to take on the kinds of rewarding passtimes you suggest. All I'm saying is that automating the commute solves a problem that people deeply do not want solved. I'm suggesting there is a difference between why people really do it, and why they say they do it.
If we really wanted to not drive, we could take public transit and be free to read or whatever. But driving is the last dying vestige where white collar workers can feel like they're doing something even remotely useful _with autonomy_. Automate driving, and it means we have to find another activity to occupy ourselves since we no longer need scrub our dishes and laundry, much less toil in the fields.
I'm talking about PORTABLE electronics - the little stuff that you charge. They are talking about -plug loads-, which includes vacuum cleaners, a/v equipment, computers, etc.
I'd be surprised if portable devices account for even 0.1% of household energy usage. Who cares? Worry about your HVAC, laundry machines, refrigerator, home server, incandescent lights, etc.
Not really. You still need regulation, and 5-10W AC powered switchers are only a cubic inch or so. And they're produced as modules in such phenomenal volume that they're probably cheaper for this application. A POE supply is not trivial - it still involves a switcher, transformer, optocoupler, and a handful of passives. Basically the same stuff.
When faced with an engineering problem, I can dip into the vast sea of information at my fingertips and instantly find answers instead of spending all day flipping through hardbacks at computer literacy, bullshitting with local sales reps to try and get copies of data sheets faxed to me, or just plain wasting time figuring out something out that's already been solved. This leaves me more time to work on the interesting stuff, or fart around on Facebook if I feel like it. I'm failing to see the downside. If you're a distractible person you can be even more distracted if you want to. If you're a productive person you can be even more productive if you want to. More information, please.
+1 on the Elk M1. The serial interface has an extensive command set. It is also accessible over ethernet, but their ethernet module is little more than a standard terminal server so you can just as easily use your own. I have written a Perl module to communicate with it using the POE framework. So you can integrate it with your insteon, A/V, etc. It supports nearly all of the Elk commands. http://www.seanadams.com/ha/automogator
It's a figure of speech. The implication is "the programmers should be embarrassed that such tasks aren't taking advantage of multiple cores".
It also happens to be nicely in phase with peak air conditioner usage.
If you were comparing plaintext passwords with strcmp you might expect something on the order of one character comparison per instruction cycle, say 1ns. But across the internet latency variance is very noisy and more like 1ms standard deviation, ie a million times larger. I'd expect you'd have to average billions of tests against each attempted character before that one little nanosecond comes out of the noise, if it's even detectable at all. Most likely there are slower clock domains along the way that will gobble it up completely.
I could imagine in a spectacularly well controlled lab environment with no other traffic on the lan and no other activity on the server, this might be demonstrated, but there's just too much jitter in any production system to do this.
Of course none of this excuses such a braindead authentication scheme. Use a hash!
That puts the share price at a mere 177% of its value 1 year ago. Their investors must be pissed!
I ordered a few sets of them in varying widths. It turns out the 15" deep ones are exactly the right size to fit in an Ikea office cabinet with the doors closed. That is a cheap, easy, and tidy way to create a huge amount of well organized space for cables, batteries, adaptors, rubber bands, wire ties, computer parts, etc etc.
Finally, you didn't ask about cables but I'm sure you have a bunch of test leads as well as common computer cables that are used occasionally. E-Z-hook is the solution: http://catalog.e-z-hook.com/viewitems/accessories/wire-test-lead-cable-wall-brackets?
Save the receipts from your component purchases in electronic form so you can search them. This is handy not just for reordering but also if you want the exact part number (to look up the specs) which might not be easily determined by looking at the part.
Finally if you are collecting SMT chip caps/resistors, diodes and transistors, you can use these flip-top cases which let you store 128 different values in a small space. http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_164267_-1 There isn't much room for labeling so these are best for resistors and caps where you just need to indicate a value for each item, and the other details for the family can be labeled on the outside of the case.
You aren't addressing Grove's point, which is that there is intellectual capital to be gained and retained by continuing to manufacture such commodities ourselves. Textiles vs semiconductors differ in this respect. There is also physical capital we might like to have ready to go, i.e. the fabs, for when China gets tired of taking our Monopoly money.
I rather suspect that the effects you're seeing has nothing to do with time skew but is general noise and distortion occurring at the longer lengths. Certainly with analog video this would happen unless you used proper baluns and impedance matching, which is probably the "active electronics" you're thinking of.
I'm guessing you have never actually worked with Asian manufacturers. New stuff doesn't just fall out of the sky for whoever is lucky enough to "identify" it. For a customer the size of Apple it is a very close partnership and seldom does the manufacturer fully own the resulting technologies. So either you help them develop the next big thing and you get some degree of exclusivity, or you wait for someone else to pioneer it and then you get it a few years later.
For some stupid reason Slashdot filters out greek letters such as , , , , and . That was supposed to say 2m.